阅读说明：本技术 高效研磨解决方案 (Efficient grinding solution ) 是由 布拉赫玛南达马·V·塔尼凯拉 克里斯多佛·阿尔科纳 托马斯·H·奥斯本 约瑟夫·P·沙利文 于 2020-03-28 设计创作，主要内容包括：本申请涉及用于获得实时磨损数据的系统和方法。示例性计算机实现的方法可包括在计算设备处接收来自一个或多个传感器的传感器数据。所述一个或多个传感器设置在磨料产品或与所述磨料产品相关联的工件附近。所述一个或多个传感器配置为收集与磨料操作相关联的磨损操作数据,所述磨料操作涉及所述磨料产品或所述工件。所述计算机实现的方法可进一步包括基于所述传感器数据来训练机器学习系统,以确定所述磨料产品的产品特定信息和/或工件特定信息。所述计算机实现的方法还可包括使用所述计算设备来提供经训练的机器学习系统。(The present application relates to systems and methods for obtaining real-time wear data. An example computer-implemented method may include receiving, at a computing device, sensor data from one or more sensors. The one or more sensors are disposed proximate to an abrasive product or a workpiece associated with the abrasive product. The one or more sensors are configured to collect wear operation data associated with an abrasive operation involving the abrasive product or the workpiece. The computer-implemented method may further include training a machine learning system based on the sensor data to determine product-specific information and/or workpiece-specific information for the abrasive product. The computer-implemented method may also include providing, using the computing device, a trained machine learning system.)

1. A computer-implemented method, the computer-implemented method comprising:

receiving, at a computing device, sensor data from one or more sensors, wherein the one or more sensors are disposed proximate to an abrasive product or a workpiece associated with the abrasive product, wherein the one or more sensors are configured to collect wear operation data associated with an abrasive operation involving the abrasive product or the workpiece;

Training, with the aid of the computing device and based on the sensor data, a machine learning system to determine product-specific information of the abrasive product or workpiece-specific information of the workpiece; and

providing, by way of the computing device, a trained machine learning system.

2. The computer-implemented method of claim 1, further comprising: tagging at least a portion of the sensor data to provide tagged sensor data, wherein the tagged sensor data includes one or more tags, each tag identifying different product-specific information of the abrasive product.

3. The computer-implemented method of claim 2, wherein the one or more markers identify a pattern of wear operation data for a duration of time prior to an abrasive product related event.

4. The computer-implemented method of claim 3, wherein the pattern of the wear operation data comprises one or more phases, each phase associated with one or more sensor thresholds, wherein the one or more markers are associated with a phase based on the one or more sensor thresholds.

5. The computer-implemented method of claim 1, wherein training the machine learning system comprises training one or more machine learning models, wherein each model is trained with sensor data from an abrasive product having a unique identifier from a shared set of identifiers.

6. The computer-implemented method of claim 1, wherein the sensor data from one or more sensors is aggregated by a local computing device to provide aggregated sensor data, and wherein receiving the sensor data comprises receiving the aggregated sensor data from the local computing device.

7. The computer-implemented method of claim 1, wherein the one or more sensors are disposed in a wearable device, wherein the sensor data from one or more sensors is aggregated by the wearable device to provide aggregated sensor data, and wherein receiving the sensor data comprises receiving the aggregated sensor data from the wearable device.

8. The computer-implemented method of claim 7, wherein the sensor data includes information indicative of a Revolutions Per Minute (RPM) value of the abrasive product.

9. A computer-implemented method, the computer-implemented method comprising:

receiving, at a computing device, sensor data from one or more sensors, wherein the one or more sensors are disposed proximate to an abrasive product or a workpiece associated with the abrasive product, wherein the one or more sensors are configured to collect wear operation data associated with an abrasive operation involving the abrasive product or the workpiece, and wherein the computing device has access to a trained machine learning system configured to receive input sensor data and output product-specific information of the abrasive product or workpiece-specific information of the workpiece;

Determining product-specific information of the abrasive product or workpiece-specific information of the workpiece by applying the trained machine learning system to the sensor data; and

providing the product-specific information or the workpiece-specific information to one or more client devices.

10. The computer-implemented method of claim 9, wherein the one or more sensors are disposed in a wearable device, wherein the sensor data from one or more sensors is aggregated by the wearable device to provide aggregated sensor data, and wherein receiving the sensor data comprises receiving the aggregated sensor data from the wearable device.

11. The computer-implemented method of claim 10, wherein the sensor data includes information indicative of a Revolutions Per Minute (RPM) value of the abrasive product.

12. The computer-implemented method of claim 9, wherein the one or more client devices comprise at least one of: a wearable device, a mobile device, a dashboard device, a network server, an analytics processing engine, or a third party server.

13. The computer-implemented method of claim 9, wherein providing the product-specific information comprises providing information associated with a new abrasive product or a newer abrasive product, wherein the information comprises, at least in part, instructions for constructing the new abrasive product or the newer abrasive product.

14. The computer-implemented method of claim 9, wherein providing the product-specific information or the artifact-specific information comprises providing a notification to the one or more client devices.

15. The computer-implemented method of claim 9, wherein providing the product-specific information or the workpiece-specific information comprises providing one or more product-specific solutions to the one or more client devices for solving a problem with the abrasive product.

16. The computer-implemented method of claim 15, wherein the one or more client devices are configured to:

displaying the one or more product-specific solutions on a graphical user interface via which a selection of one of the one or more product-specific solutions is received;

determining training data for the trained machine learning system based on the selected product-specific solution; and

transmitting the training data to the computing device.

17. The computer-implemented method of claim 9, wherein the abrasive product is a hand-held abrasive product operated by a user.

18. The computer-implemented method of claim 17, wherein the abrasive operation data comprises at least one of: vibration data associated with the handheld abrasive product or acceleration data associated with the handheld abrasive product.

19. The computer-implemented method of claim 17, wherein providing the product-specific information or the workpiece-specific information comprises providing a notification to a graphical interface of a wearable device worn by the user.

20. The computer-implemented method of claim 17, wherein the product-specific information comprises at least one of: time spent performing a task assigned to the user, idle time of the user, or work time of the user.

21. The computer-implemented method of claim 17, wherein the product-specific information comprises at least one of: an operating angle of the user relative to the hand-held abrasive product, an operating angle of the hand-held abrasive product relative to the workpiece, a gripping force of the user on the hand-held abrasive product, or a pressure exerted by the user on the hand-held abrasive product.

22. The computer-implemented method of claim 17, wherein the product-specific information includes an end-of-life estimate of the handheld abrasive product, wherein the end-of-life estimate includes an estimated amount of time that the user can safely use the handheld abrasive product.

23. The computer-implemented method of claim 9, wherein the abrasive product is an automated abrasive product operated by a controller.

24. The computer-implemented method of claim 23, wherein the one or more sensors comprise a spark-constant sensor configured to collect an operating speed of the automated abrasive product.

25. The computer-implemented method of claim 23, wherein providing the product specific information comprises providing a determination that: one or more abrasive articles of the automated abrasive product are damaged or fail.

26. The computer-implemented method of claim 25, wherein in providing the determination, the computing device is further configured to:

identifying, by a product database, one or more replacement abrasive articles of the automated abrasive product; and

In response to identifying the one or more replacement abrasive articles, a request is made for one or more replacement articles or a refurbishment process.

27. The computer-implemented method of claim 23, wherein providing the product-specific information comprises transmitting at least one control instruction to the controller of the automated abrasive product, wherein the at least one control instruction comprises at least one of: adjusting a rotational speed of the automated abrasive product, providing a notification to the automated abrasive product, turning the automated abrasive product on, or turning the automated abrasive product off.

28. The computer-implemented method of claim 9, wherein determining the product-specific information of the abrasive product or workpiece-specific information of the workpiece comprises determining one or more of: a predicted future condition of the abrasive product or a predicted future condition of the workpiece.

29. The computer-implemented method of claim 28, wherein providing the product-specific information or the workpiece-specific information comprises providing at least one of: the predicted future condition of the abrasive product or the predicted future condition of the workpiece.

30. The computer-implemented method of claim 9, wherein determining the product-specific information of the abrasive product or workpiece-specific information of the workpiece comprises determining a normative action.

31. The computer-implemented method of claim 30, wherein providing the product-specific information or the workpiece-specific information comprises providing the normative action, wherein the normative action comprises at least one of: adjusting an operating parameter of a grinding mill, performing a maintenance operation, repairing the abrasive product, or replacing the abrasive product.

32. A computing system, the computing system comprising:

a trained machine learning system configured to receive input sensor data and to output product-specific information or workpiece-specific information based on the input sensor data; and

a computing device configured to:

receiving sensor data from one or more sensors, wherein the one or more sensors are disposed proximate to an abrasive product or a workpiece associated with the abrasive product, wherein the one or more sensors are configured to collect wear operation data associated with an abrasive operation involving the abrasive product or the workpiece;

Determining product-specific information of the abrasive product or workpiece-specific information of the workpiece by applying the trained machine learning system to the sensor data; and

providing the product-specific information or the workpiece-specific information to one or more client devices.

33. The computing system of claim 32, further comprising:

a display configured to provide a graphical user interface having a search interface, wherein the search interface includes a plurality of user-selectable criteria, wherein the user-selectable criteria include at least one of: a location menu, a device menu, a date range, or a workpiece menu, wherein the graphical user interface is configured to:

receiving user-selected search criteria from the user-selectable criteria;

determining one or more metrics based on the user-selected search criteria; and

displaying the one or more metrics via the graphical user interface.

34. The computing system of claim 33, wherein the one or more metrics comprise at least one of: a grinding time metric, a best grinding metric, a vibration metric, a depth of cut, a current trace, a tool identifier, or a part count.

35. The computing system of claim 33, wherein the graphical user interface is further configured to receive a desired metric selected from a menu of metrics, wherein the one or more metrics are displayed based on the desired metric.

36. The computing system of claim 33, wherein the display is further configured to provide a cycle comparison interface via the graphical user interface, wherein the cycle comparison interface is configured to display at least a portion of the sensor data in an overlapping arrangement of a plurality of periodic time series.

37. The computing system of claim 32, wherein the computing device is further configured to:

comparing a plurality of periodic time series of at least a portion of the sensor data, wherein the product-specific information of the abrasive product or workpiece-specific information of the workpiece is determined based at least in part on the comparison.

Background

Abrasive tools are useful for a variety of material removal operations. Such tools have been equipped with sensors that can monitor tool usage. For example, a power sensor may be incorporated into the tool in order to monitor the electrical power consumed by the load. While power sensors incorporated into tools may provide useful information to users of the tools about the tools, the sensors may not fully capture the operation of the tools and/or the user's experience. For example, power sensor data cannot be effectively used to determine whether a component of a tool has been damaged or failed.

Disclosure of Invention

The present disclosure relates generally to systems and methods for obtaining, analyzing, and utilizing real-time data in abrasive and abrasive equipment applications.

In a first aspect, a computer-implemented method is provided. A computer-implemented method includes receiving, at a computing device, sensor data from one or more sensors. The one or more sensors are disposed in proximity to at least one of: one or more abrasive products or one or more workpieces. The one or more sensors are configured to collect wear operation data associated with abrasive operations involving the one or more abrasive products or the one or more workpieces. The computer-implemented method includes training a machine learning system using a computing device to determine product-specific information and/or workpiece-specific information for the one or more abrasive products based on the sensor data. The computer-implemented method also includes providing, using the computing device, a trained machine learning system.

In a second aspect, a computer-implemented method is provided. A computer-implemented method includes receiving, at a computing device, sensor data from one or more sensors. The one or more sensors are disposed in proximity to at least one of: an abrasive product or a workpiece. The one or more sensors are configured to collect wear operation data associated with an abrasive operation involving an abrasive product or workpiece. The computing device has access to a trained machine learning system configured to receive input sensor data and output product-specific information or workpiece-specific information of the abrasive product based on the input sensor data. The computer-implemented method includes determining, by a trained machine learning system, product-specific information or workpiece-specific information for an abrasive product based on sensor data. The computer-implemented method additionally includes providing product-specific information or workpiece-specific information of the abrasive product to one or more client devices.

In a third aspect, a system is provided. The system includes a trained machine learning system configured to receive input sensor data and output, based on the input sensor data, at least one of: product-specific information or workpiece-specific information of the abrasive product. The system also includes a computing device configured to receive sensor data from the one or more sensors from the client device. The one or more sensors are disposed proximate to the abrasive product or workpiece. The one or more sensors are configured to collect wear operation data associated with an abrasive operation involving the abrasive product or the workpiece. The computing device is further configured to determine product-specific information or workpiece-specific information of the abrasive product by applying the trained machine learning system to the sensor data. The computing device is further configured to provide the product-specific information and/or the workpiece-specific information of the abrasive product to one or more client devices.

In a fourth aspect, a computer-implemented method is provided. A computer-implemented method includes receiving, at a computing device, sensor data from one or more sensors. The one or more sensors are disposed proximate a hand-held abrasive product operated by a user. The one or more sensors are configured to collect wear operation data associated with an abrasive operation involving a hand-held abrasive product. The computing device has access to a trained machine learning system configured to receive input sensor data and output product-specific information for the abrasive product based on the input sensor data. The computer-implemented method includes determining, by a trained machine learning system, product-specific information of abrasive product information based on sensor data. The computer-implemented method also includes providing product-specific information related to abrasive operation of the abrasive product to one or more client devices.

In a fifth aspect, a computer-implemented method is provided. A computer-implemented method includes receiving, at a computing device, sensor data from one or more sensors. The one or more sensors are disposed proximate to an abrasive product operated by an automated process. The one or more sensors are configured to collect wear operation data associated with an abrasive operation involving an abrasive product. The computing device has access to a trained machine learning system configured to receive input sensor data and output product-specific information for the abrasive product based on the input sensor data. The computer-implemented method includes determining, by a trained machine learning system, product-specific information for an abrasive product based on sensor data. The computer-implemented method also includes providing product-specific information related to the abrasive product to one or more client devices.

In a sixth aspect, a computer-implemented method is provided. A computer-implemented method includes receiving, at a computing device, sensor data from one or more sensors. The one or more sensors are disposed proximate to the one or more abrasive products. The one or more sensors are configured to collect wear operation data associated with abrasive operations involving the one or more abrasive products. The one or more abrasive products are located in a plurality of enterprises. The computer-implemented method further includes training, using the computing device, the machine learning system to determine product-specific information or workpiece-specific information for the one or more abrasive products based on the sensor data. The computer-implemented method additionally includes providing, using the computing device, a trained machine learning system.

Drawings

Fig. 1A includes a perspective view illustration of a shaped abrasive particle according to an example embodiment.

Fig. 1B includes a top view of the shaped abrasive particle of fig. 1A according to an example embodiment.

Fig. 1C includes a perspective view illustration of a shaped abrasive particle according to an example embodiment.

Fig. 2A includes a perspective view illustration of a highly controlled abrasive particle (CHAP) according to an example embodiment.

Fig. 2B includes a perspective view illustration of a non-shaped particle according to an example embodiment.

FIG. 3 includes a cross-sectional illustration of a coated abrasive article incorporating a particulate material according to an example embodiment.

FIG. 4 includes a top view of a portion of a coated abrasive according to an example embodiment.

FIG. 5 illustrates a cross-section of a portion of a coated abrasive according to an example embodiment.

FIG. 6 includes a perspective view illustration of a bonded abrasive article according to an example embodiment.

FIG. 7A illustrates a cross-sectional view of an abrasive article according to an example embodiment.

Fig. 7B illustrates a cross-sectional view of an electronic assembly, according to an example embodiment.

Fig. 8A illustrates a top view of a releasable coupling of an electronic component on a body, according to an example embodiment.

Fig. 8B illustrates an abrasive system according to an example embodiment.

Fig. 9A-9C illustrate microscopic interactions according to an example embodiment.

FIG. 10 illustrates a block diagram of a computing device, according to an example embodiment.

FIG. 11 illustrates an arrangement of a machine learning platform according to an example embodiment.

FIG. 12 depicts a scenario involving an enterprise, a machine learning system, an input request, and an output prediction, according to an example embodiment.

FIG. 13 illustrates a scenario representing an input device to a machine learning system, according to an example embodiment.

FIG. 14 depicts an apparatus to receive predictions from a machine learning system according to an example embodiment.

Fig. 15 is a flow chart of a method according to an example embodiment.

Fig. 16 is a flow chart of a method according to an example embodiment.

Fig. 17 is a flow chart of a method according to an example embodiment.

Fig. 18A illustrates a grinding wheel according to an example embodiment.

Fig. 18B illustrates a grinding wheel according to an example embodiment.

FIG. 19 illustrates labeled components according to an example embodiment.

FIG. 20 illustrates a communication environment, according to an example embodiment.

FIG. 21 illustrates a communication environment, according to an example embodiment.

FIG. 22 illustrates a method according to an example embodiment.

FIG. 23 illustrates a state diagram in accordance with an example embodiment.

Fig. 24A-24C illustrate views of a mobile device according to an example embodiment.

FIG. 25 illustrates a model according to an example embodiment.

FIG. 26 illustrates a view of a web application, according to an example embodiment.

Fig. 27 illustrates several displays of a wearable device according to an example embodiment.

Figures 28A-28C illustrate panes of an analysis platform according to an example embodiment,

FIG. 29 illustrates a graph according to an example embodiment.

FIG. 30 illustrates a graph according to an example embodiment.

Detailed Description

Example methods, devices, and systems are described herein. It should be understood that the words "example" and "exemplary" are used herein to mean "serving as an example, instance, or illustration. Any embodiment or feature described herein as "exemplary" or "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments or features. Other embodiments may be utilized, and other changes may be made, without departing from the scope of the subject matter presented here.

Accordingly, the example embodiments described herein are not meant to be limiting. The aspects of the present disclosure, as generally described herein, and illustrated in the figures, may be arranged, substituted, combined, separated, and designed in a wide variety of different configurations, all of which are contemplated herein.

Further, the features shown in each figure may be used in combination with each other, unless the context indicates otherwise. Thus, the drawings are to be generally regarded as constituting aspects of one or more general embodiments, with the understanding that not all illustrated features are required for each embodiment.

I. Overview

Abrasive products are widely used in various industrial and domestic operations ranging from home decoration projects to high-tech precision projects. With the aid of these products, operators can perform grinding, polishing, buffing, and other operations to shape and finish many different types of materials.

Typically, abrasive product manufacturers collect operational data from customers to improve the productivity and safety of their abrasive products. For example, a manufacturer may equip an abrasive product with sensors to generate data streams regarding the input (e.g., components forming the product), operation (e.g., power, speed, and/or vibration of the product), and output (e.g., end material surface finish) of the abrasive product. By combining multiple data streams (e.g., through an internet of things (IoT) aggregation tool), a manufacturer can obtain diagnostic information from various abrasive products, thereby enhancing the ability of the manufacturer to perform process monitoring, address customer issues, and improve development of future abrasive products.

To provide more value, it may be beneficial for a manufacturer to convert diagnostic information into a form that is easy for a customer to use. For example, assuming that the customer may be an enterprise, a production manager from the enterprise may be concerned with production rate and quality information for the abrasive product, while an operator from the enterprise may be concerned with real-time safety information. Thus, it may be advantageous for a manufacturer to convert diagnostic information into a text notification and transmit the notification to a graphical interface used by production managers and operators.

To efficiently translate and transmit diagnostic information, manufacturers may benefit from a remote hosted platform that can learn about individuals/entities operating within an enterprise and distribute diagnostic information to related individuals/entities in real-time. The goal of such platforms is to develop predictive intelligence and analysis frameworks for the customer's program so that the customer can focus on producing high value material with the product, rather than wasting time analyzing the abrasive product data.

To achieve this goal, a machine learning platform is described herein that can intelligently provide predictive analysis to customers of abrasive product manufacturers. Such machine learning platforms may be hosted remotely from the customer, but may access data and services from the customer by way of a secure connection. The machine learning platform may be web-based and accessible from various internet-enabled client devices. For example, a machine learning platform may have a mobile application component (iOS/Android) and a web services component, which allows customers easy access to features provided by the platform.

Such machine learning platforms may have several desirable capabilities and characteristics. For example, by utilizing aggregated diagnostic information across multiple customers, the machine learning platform can develop in-depth insights that provide real-time feedback to recommend and/or adjust customer operations and predicted statistics in real-time to drive future business decisions for the customer. These features of the machine learning platform can also be exploited by manufacturers to develop new business models, including abrasive services and abrasive products, to drive further growth of manufacturers. Other features, functions, and benefits of such machine learning platforms may exist and will be appreciated and understood from the following discussion.

Accordingly, disclosed herein are methods and systems for using abrasive operational data indicative of the behavior of an abrasive product. As described herein, the abrasive operation data can be sent to a machine learning platform to train one or more machine learning models. Each machine learning model may be configured to predict one or more behaviors of the abrasive product based on the abrasive operation data. The machine learning platform may be transmitted to an interface on the abrasive product, to a mobile computing device, or to an analysis platform for product-specific information or workpiece-specific information related to the abrasive product. This information may include providing ergonomic recommendations to an operator using the product, determining the end of life of the product, and/or determining operational improvements (e.g., best practices for the workflow).

Exemplary abrasive particles

As used herein, the term abrasive tool includes any tool configured for use with an abrasive article. The abrasive article can comprise a fixed abrasive article including at least a substrate and abrasive particles attached to (e.g., contained within or overlying) the substrate. The abrasive articles of the embodiments herein can be bonded abrasives, coated abrasives, nonwoven abrasives, thin wheels, cutting wheels, reinforced abrasive articles, superabrasive (superabraive), single layer abrasive articles, and the like. Such abrasive articles may include one or more various types of abrasive particles including, for example, but not limited to, shaped abrasive particles, constant height abrasive particles, unformed abrasive particles (e.g., crushed, pressed, or blasted abrasive particles), and the like.

Fig. 1A includes a perspective view illustration of a shaped abrasive particle according to an embodiment. Shaped abrasive particle 100 can include a body 101 including a major surface 102, a major surface 103, and a side surface 104 extending between major surface 102 and major surface 103. As shown in fig. 1A, the body 101 of the shaped abrasive particle 100 can be a thin body, wherein the major surface 102 and the major surface 103 are larger than the side surface 104. Further, the body 101 may include a longitudinal axis 110 extending from the point to the base and extending through a midpoint 150 on the major surface 102 or 103. The longitudinal axis 110 can define a longest dimension of the body along the major surface and through a midpoint 150 of the major surface 102.

In some particles, if the midpoint of the major surface of the body is not readily apparent, the major surface may be viewed from above, a nearest fit circle drawn around the two-dimensional shape of the major surface, and the center of the circle used as the midpoint of the major surface.

Fig. 1B includes a top view illustration of the shaped abrasive particle of fig. 1A. Notably, the main body 101 includes a main surface 102 having a triangular two-dimensional shape. A circle 160 is drawn around the triangle to facilitate positioning the midpoint 150 on the major surface 102.

Referring again to fig. 1A, the body 101 can further include a lateral axis 111 defining a width of the body 101 extending substantially perpendicular to the longitudinal axis 110 on the same major surface 102. Finally, as shown, the body 101 may include a vertical axis 112, which in the context of a thin body may define the height (or thickness) of the body 101. For a thin body, the length of the longitudinal axis 110 is greater than the vertical axis 112. As shown, the thickness along vertical axis 112 may extend along side surface 104 between major surface 102 and major surface 103, and is perpendicular to a plane defined by longitudinal axis 110 and lateral axis 111. It should be understood that references herein to the length, width and height of abrasive particles may refer to an average of suitable sample amounts of abrasive particles taken from a larger group, including, for example, a group of abrasive particles attached to a fixed abrasive.

Shaped abrasive particles of the embodiments herein, including thin shaped abrasive particles, can have a primary aspect ratio, expressed as length to width, such that the length can be greater than or equal to the width. Further, the length of the body 101 may be greater than or equal to the height. Finally, the width of the body 101 may be greater than or equal to the height. According to an embodiment, the primary aspect ratio (length: width) may be at least 1:1, such as at least 1.1:1, at least 1.2:1, at least 1.5:1, at least 1.8:1, at least 2:1, at least 3:1, at least 4:1, at least 5:1, at least 6:1, or even at least 10: 1. In another non-limiting embodiment, the body 101 of the shaped abrasive particles can have a primary aspect ratio, expressed as length to width, of no greater than 100:1, no greater than 50:1, no greater than 10:1, no greater than 6:1, no greater than 5:1, no greater than 4:1, no greater than 3:1, no greater than 2:1, or even no greater than 1:1. It should be understood that the primary aspect ratio of the body 101 may be in a range including between any of the minimum to maximum ratios noted above.

However, in certain other embodiments, the width may be greater than the length. For example, in those embodiments in which the body 101 is an equilateral triangle, the width may be greater than the length. In such embodiments, the major aspect ratio, expressed as length to width, may be at least 1:1.1, or at least 1:1.2, or at least 1:1.3, or at least 1:1.5, or at least 1:1.8, or at least 1:2, or at least 1:2.5, or at least 1:3, or at least 1:4, or at least 1:5, or at least 1: 10. Additionally, in one non-limiting embodiment, the major aspect ratio length to width may be no greater than 1:100, or no greater than 1:50, or no greater than 1:25, or no greater than 1:10, or no greater than 5:1, or no greater than 3: 1. It should be understood that the primary aspect ratio of the body 101 may be in a range including between any of the minimum to maximum ratios noted above.

Further, the body 101 may have a second aspect ratio, expressed in width to height, which may be at least 1:1, such as at least 1.1:1, at least 1.2:1, at least 1.5:1, at least 1.8:1, at least 2:1, at least 3:1, at least 4:1, at least 5:1, at least 8:1, or even at least 10: 1. Additionally, in another non-limiting embodiment, the second aspect ratio (width: height) of the body 101 can be no greater than 100:1, such as no greater than 50:1, no greater than 10:1, no greater than 8:1, no greater than 6:1, no greater than 5:1, no greater than 4:1, no greater than 3:1, or even no greater than 2: 1. It should be understood that the second aspect ratio, expressed as width to height, may be within a range including any of the minimum and maximum ratios above.

In another embodiment, the body 101 may have a third aspect ratio, expressed as length to height, which may be at least 1.1:1, such as at least 1.2:1, at least 1.5:1, at least 1.8:1, at least 2:1, at least 3:1, at least 4:1, at least 5:1, at least 8:1, or even at least 10: 1. Additionally, in another non-limiting embodiment, the third aspect ratio (length: height) of the body 101 can be no greater than 100:1, such as no greater than 50:1, no greater than 10:1, no greater than 8:1, no greater than 6:1, no greater than 5:1, no greater than 4:1, no greater than 3: 1. It should be understood that the third aspect ratio of the body 101 may be within a range including any of the minimum and maximum ratios above.

The abrasive particles (including shaped abrasive particles) of the embodiments herein can comprise crystalline materials, and more particularly, polycrystalline materials. Notably, the polycrystalline material may include abrasive grains. In one embodiment, the body of abrasive particles (e.g., the body comprising shaped abrasive particles) can be substantially free of organic materials, such as binders. In at least one embodiment, the abrasive particles can consist essentially of polycrystalline material. In another embodiment, abrasive particles, such as shaped abrasive particles, may be silane-free and, in particular, may not have a silane coating.

The abrasive particles may be made of certain materials including, but not limited to, nitrides, oxides, carbides, borides, oxynitrides, oxyborides, diamond, carbonaceous materials, and combinations thereof. In particular instances, the abrasive particles can include oxide compounds or composites such as alumina, zirconia, titania, yttria, chromia, strontium oxide, silica, magnesia, rare earth oxides, and combinations thereof. The abrasive particles may be superabrasive particles.

In a particular embodiment, the abrasive particles can comprise a majority content of alumina. For at least one embodiment, the abrasive particles may comprise at least 80 wt% alumina, such as at least 90 wt% alumina, at least 91 wt% alumina, at least 92 wt% alumina, at least 93 wt% alumina, at least 94 wt% alumina, at least 95 wt% alumina, at least 96 wt% alumina, or even at least 97 wt% alumina. Additionally, in at least one particular embodiment, the abrasive particles can comprise not greater than 99.5wt alumina, such as not greater than 99wt alumina, not greater than 98.5wt alumina, not greater than 97.5wt alumina, not greater than 97wt alumina, not greater than 96wt alumina, or even not greater than 94wt alumina. It should be understood that the abrasive particles of the embodiments herein can comprise an alumina content within a range including between any of the minimum and maximum percentages noted above. Further, in particular instances, the shaped abrasive particles can be formed from a seeded sol-gel. In at least one embodiment, the abrasive particles can consist essentially of alumina and certain dopant materials described herein.

In particular, the abrasive particles of the embodiments herein can include a compact, which can be suitable for use as an abrasive. For example, the abrasive particles may have a body with a density of at least 95% theoretical density, such as at least 96% theoretical density, at least 97% theoretical density, at least 98% theoretical density, or even at least 99% theoretical density.

The abrasive particles (i.e., crystallites) contained within the body of the abrasive particles can have an average grain size (i.e., average crystal size) generally not greater than about 100 microns. In other embodiments, the average grain size may be smaller, such as not greater than about 80 microns, or not greater than about 50 microns, or not greater than about 30 microns, or not greater than about 20 microns, or not greater than about 10 microns, or not greater than 6 microns, or not greater than 5 microns, or not greater than 4 microns, or not greater than 3.5 microns, or not greater than 3 microns, or not greater than 2.5 microns, or not greater than 2 microns, or not greater than 1.5 microns, or not greater than 1 micron, or not greater than 0.8 microns, or not greater than 0.6 microns, or not greater than 0.5 microns, or not greater than 0.4 microns, or not greater than 0.3 microns, or even not greater than 0.2 microns. Additionally, the average grain size of the abrasive particles contained within the body of the abrasive particles can be at least about 0.01 microns, such as at least about 0.05 microns, or at least about 0.06 microns, or at least about 0.07 microns, or at least about 0.08 microns, or at least about 0.09 microns, or at least about 0.1 microns, or at least about 0.12 microns, or at least about 0.15 microns, or at least about 0.17 microns, or at least about 0.2 microns, or even at least about 0.3 microns. It will be appreciated that the abrasive particles can have an average grain size (i.e., average crystal size) within a range between any minimum and maximum values noted above.

The average grain size (i.e., average crystal size) can be measured based on an uncorrected interception method using a Scanning Electron Microscope (SEM) micrograph. The preparation method of the abrasive grain sample comprises the following steps: the bakelite scaffolds were made in epoxy resin and then polished with diamond polishing slurry using a Struers Tegramin 30 polishing unit. After polishing, the epoxy was heated on a hot plate, and then the polished surface was subjected to thermal etching at 150 ℃ below the sintering temperature for 5 minutes. Individual dies (5-10) were mounted on SEM scaffolds and then gold plated in preparation for SEM observation. SEM micrographs of three individual abrasive particles were taken at a magnification of about 50,000X, and then the uncorrected crystallite size was calculated using the following steps: 1) drawing a diagonal line from one corner to the opposite corner of the crystal structure view, but excluding the black data band at the bottom of the photograph; 2) measuring the lengths L1 and L2 of the diagonals to the nearest 0.1 cm; 3) calculating the number of grain boundaries intersected by each diagonal (i.e., grain boundary intersections I1 and I2) and recording the number for each diagonal; 4) the calculated number of bars is determined by measuring the length of the micrometer bars (in centimeters) at the bottom of each micrograph or viewing screen (i.e., "bar length"), then dividing the bar length (in micrometers) by the bar length (in centimeters); 5) adding the total centimeters of the diagonals drawn on the photomicrograph (L1+ L2) to obtain the sum of the length of the diagonals; 6) adding the intersection points of the grain boundaries of the two diagonal lines (I1+ I2) to obtain the sum of the intersection points of the grain boundaries; 7) the sum of the diagonal lengths (L1+ L2) (in centimeters) was divided by the sum of the number of grain boundary intersections (I1+ I2), and this number was then multiplied by the calculated number. This process was completed at least three different times for three different, randomly selected samples to obtain an average crystallite size.

According to certain embodiments, certain abrasive particles may be composite articles that include at least two different types of grains within the body of the abrasive particle. It should be understood that the different types of grains are grains having different compositions from each other. For example, the body of abrasive particles can be formed such that it includes at least two different types of grains, wherein the two different types of grains can be nitrides, oxides, carbides, borides, oxynitrides, oxyborides, diamond, and a combination thereof.

According to one embodiment, the abrasive particles can have an average particle size of at least about 100 microns as measured in the largest dimension (i.e., length). In fact, the abrasive particles can have an average particle size of at least about 150 microns, such as at least about 200 microns, at least about 300 microns, at least about 400 microns, at least about 500 microns, at least about 600 microns, at least about 800 microns, or even at least about 900 microns. However, the abrasive particles of the embodiments herein can have an average particle size of not greater than about 5mm, such as not greater than about 3mm, not greater than about 2mm, or even not greater than about 1.5 mm. It will be appreciated that the abrasive particles can have an average particle size within a range between any minimum and maximum values noted above.

Fig. 1A includes an illustration of a shaped abrasive particle having a two-dimensional shape defined by the plane of the upper major surface 102 or major surface 103, which has a generally triangular two-dimensional shape. It should be understood that the shaped abrasive particles of the embodiments herein are not so limited and may include other two-dimensional shapes. For example, the shaped abrasive particles of the embodiments herein can include particles having a body with a two-dimensional shape defined by a major surface of the body, the two-dimensional shape selected from the group consisting of: polygons, regular polygons, irregular polygons including curved or bent sides or side portions, ovals, numbers, greek alphabet characters, latin alphabet characters, russian alphabet characters, kanji characters, complex shapes having a combination of polygonal shapes, shapes (e.g., stars) including a central region and a plurality of arms (e.g., at least three arms) extending from the central region, and combinations thereof. Particular polygonal shapes include rectangles, trapezoids, quadrilaterals, pentagons, hexagons, heptagons, octagons, nonagons, decagons, and any combination thereof. In another example, the finally-formed shaped abrasive particles can have a body with a two-dimensional shape such as a trapezoid, an irregular rectangle, an irregular trapezoid, an irregular pentagon, an irregular hexagon, an irregular heptagon, an irregular octagon, an irregular nonagon, an irregular decagon, and combinations thereof. An irregular polygonal shape refers to a polygonal shape that defines at least one side of the polygonal shape differing in size (e.g., length) relative to another side. As shown in other embodiments herein, the two-dimensional shape of certain shaped abrasive particles can have a particular number of outer points or angles. For example, the body of the shaped abrasive particle can have a two-dimensional polygonal shape when viewed in a plane defined by a length and a width, wherein the body comprises a two-dimensional shape having at least 4 exterior points (e.g., a quadrilateral), at least 5 exterior points (e.g., a pentagon), at least 6 exterior points (e.g., a hexagon), at least 7 exterior points (e.g., a heptagon), at least 8 exterior points (e.g., an octagon), at least 9 exterior points (e.g., a nonagon), and the like.

Fig. 1C includes a perspective view illustration of a shaped abrasive particle according to another embodiment. Notably, the shaped abrasive particle 170 can include a body 171 that includes a surface 172 and a surface 173, which can be referred to as an end face 172 and an end face 173. The body can further include a major surface 174, a major surface 175, a major surface 176, and a major surface 177 extending between and coupled to the end faces 172, 173. The shaped abrasive particles of fig. 1C are elongated shaped abrasive particles having a longitudinal axis 180 extending along the major surface 175 and extending through a midpoint 184 between the end face 172 and the end face 173. For particles having an identifiable two-dimensional shape, such as the shaped abrasive particles of fig. 1A-1C, the longitudinal axis is a well understood dimension that is used to define the length of the body through a midpoint on the major surface. For example, in fig. 1C, as shown, the longitudinal axis 180 of the shaped abrasive particle 170 extends between the end face 172 and the end face 173 parallel to the edges defining the major surface. Such a longitudinal axis coincides with how the length of the bar is defined. Notably, the longitudinal axis 180 extends non-diagonally between the corners of the connecting end surfaces 172 and 173 and the edges defining the major surface 175, even though such a line may define the dimension of greatest length. To the extent that the major surface has undulations or minor defects relative to a perfectly flat surface, the longitudinal axis can be determined using a top view two-dimensional image with the undulations omitted.

It should be understood that surface 175 is selected to represent longitudinal axis 180 because body 171 has a generally square cross-sectional profile defined by end face 172 and end face 173. Thus, surface 174, surface 175, surface 176, and surface 177 may have substantially the same dimensions relative to one another. However, in the context of other elongated abrasive particles, surface 172 and surface 173 may have different shapes, such as rectangular shapes, and thus, at least one of surface 174, surface 175, surface 176, and surface 177 may be larger relative to one another. In such cases, the largest surface may define a major surface, and the longitudinal axis will extend along the largest surface thereof through the midpoint 184 and may extend parallel to the edges defining the major surface. As further shown, the body 171 can include a transverse axis 181 that extends perpendicular to the longitudinal axis 180 within the same plane defined by the surface 175. As further shown in the figures, the body 171 can further include a vertical axis 182 defining a height of the abrasive particles, the vertical axis 182 extending in a direction perpendicular to a plane defined by the longitudinal axis 180 and the lateral axis 181 of the surface 175.

It should be understood that similar to the thin shaped abrasive particles of fig. 1A-1B, the elongated shaped abrasive particles of fig. 1C can have various two-dimensional shapes, such as those defined with respect to the shaped abrasive particles of fig. 1A-1B. The two-dimensional shape of the body 171 may be defined by the shape of the perimeter of the end faces 172 and 173. The elongated shaped abrasive particles 170 can have any of the properties of the shaped abrasive particles of the embodiments herein.

Fig. 2A includes a perspective view illustration of a highly controlled abrasive particle (CHAP) according to an embodiment. As shown, the CHAP 200 may include a body 201 including a first major surface 202, a second major surface 203, and a side surface 204 extending between the first major surface 202 and the second major surface 203. As shown in fig. 2A, the main body 201 may have a thin, relatively planar shape in which the first main surface 202 and the second main surface 203 are larger than the side surface 204 and are substantially parallel to each other. Further, the body 201 may include a longitudinal axis 210 extending through the midpoint 220 and defining a length of the body 201. The body 201 can further include a lateral axis 211 on the first major surface 202 that extends perpendicular to the longitudinal axis 210 through a midpoint 220 of the first major surface 202 and defines a width of the body 201.

The body 201 may further include a vertical axis 212, which may define the height (or thickness) of the body 201. As shown, a vertical axis 212 may extend between the first and second major surfaces 202 and 203 along the side surface 204 in a direction generally perpendicular to a plane defined by the axes 210 and 211 on the first major surface. For thin bodies such as CHAP 200 shown in fig. 2A, the length may be equal to or greater than the width, and the length may be greater than the height. It should be understood that references herein to the length, width and height of abrasive particles may refer to an average of a suitable sample amount of abrasive particles taken from a batch of abrasive particles.

Unlike the shaped abrasive particles of fig. 1A, 1B, and 1C, CHAP 200 of fig. 2A does not have an easily identifiable two-dimensional shape based on the perimeter of the first major surface 202 or the second major surface 203. Such abrasive particles can be formed in a variety of ways, including but not limited to fracturing thin layers of material to form abrasive particles having a controlled height but with irregularly formed planar major surfaces. For such particles, the longitudinal axis is defined as the longest dimension on the major surface that extends through a midpoint on the surface. To the extent that the major surface has undulations, the longitudinal axis can be determined using a top view two-dimensional image with the undulations omitted. Further, as mentioned in fig. 1B, a closest fitting circle may be used to identify the midpoint of the major surface and to identify the longitudinal and lateral axes.

Fig. 2B includes an illustration of a non-shaped particle, which may be an elongated non-shaped abrasive particle or a secondary particle, such as a diluent grain, filler, agglomerate, or the like. The shaped abrasive particles can be formed by specific processes including molding, printing, casting, extrusion, and the like. The shaped abrasive particles can be formed such that each particle has substantially the same surface and edge arrangement relative to each other. For example, the surfaces and edges of a group of shaped abrasive particles typically have the same arrangement and orientation and/or two-dimensional shape as one another. As a result, the shaped abrasive particles have relatively high shape fidelity and consistency in the arrangement of the surfaces and edges relative to each other. In addition, the high abrasive particles (CHAP) may also be formed by specific processes that facilitate the formation of thin bodies that may have irregular two-dimensional shapes when viewed from above looking down at the major surface. CHAP may have less shape fidelity than shaped abrasive particles, but may have generally flat and parallel major planes separated by side surfaces.

In contrast, non-shaped particles can be formed by a different process and have different shape properties than shaped abrasive particles and CHAP. For example, non-shaped particles are typically formed by a pulverization process, wherein a mass of material is formed, then crushed and sieved to obtain abrasive particles of a certain size. However, the non-shaped particles will have a generally random arrangement of surfaces and edges, and generally lack any identifiable two-dimensional or three-dimensional shape in the surface and edge arrangement. Furthermore, the non-shaped particles do not have to have shapes that are consistent with each other, and thus have much lower shape fidelity than shaped abrasive particles or CHAP. Non-shaped particles are generally defined by the random arrangement of each particle relative to the surfaces and edges of other non-shaped particles.

Fig. 2B includes a perspective view illustration of a non-shaped particle. The non-shaped particles 250 may have a body 251 that includes a generally random arrangement of edges 255 extending along an outer surface of the body 251. The body may further include a longitudinal axis 252 defining the longest dimension of the particle. The longitudinal axis 252 defines the longest dimension of the body viewed in two dimensions. Thus, unlike the shaped abrasive particle and CHAP, where the longitudinal axis is measured on the major surface, the longitudinal axis of the non-shaped particle is defined by the points on the body that are furthest from each other because the particle is viewed in two dimensions using an image or vantage point that provides a view of the longest dimension of the particle. That is, elongated particles such as shown in fig. 2B, but not shaped particles, should be viewed in perspective to make the longest dimension apparent for proper assessment of the longitudinal axis. Body 251 may further include a transverse axis 253 that extends perpendicular to longitudinal axis 252 and defines the width of the particle. Transverse axis 253 may extend perpendicular to longitudinal axis 252 through a midpoint 256 of the longitudinal axis in the same plane used to identify longitudinal axis 252. The abrasive particles may have a height (or thickness) as defined by the vertical axis 254. Vertical axis 254 may extend through midpoint 256, but in a direction perpendicular to the plane defining longitudinal axis 252 and lateral axis 253. To evaluate height, it may be necessary to change the perspective of the abrasive particle to view the abrasive particle from a favorable angle different from the evaluation of length and width.

It should be understood that the abrasive particles may have a length defined by a longitudinal axis 252, a width defined by a lateral axis 253, and a height defined by a vertical axis 254. It should be understood that the body 251 may have a major aspect ratio, expressed as length to width, such that the length is equal to or greater than the width. Further, the length of the body 251 may be greater than or equal to the height. Finally, the width of the body 251 may be greater than or equal to the height. According to one embodiment, the primary aspect ratio (length: width) may be at least 1.1:1, at least 1.2:1, at least 1.5:1, at least 1.8:1, at least 2:1, at least 3:1, at least 4:1, at least 5:1, at least 6:1, or even at least 10: 1. In another non-limiting embodiment, the body 251 of elongated shaped abrasive particles can have a primary aspect ratio, expressed as length to width, of no greater than 100:1, no greater than 50:1, no greater than 10:1, no greater than 6:1, no greater than 5:1, no greater than 4:1, no greater than 3:1, or even no greater than 2: 1. It should be understood that the major aspect ratio of the body 251 may be in a range including between any of the minimum to maximum ratios noted above.

Further, the body 251 may include a second aspect ratio, expressed in width to height, which may be at least 1.1:1, such as at least 1.2:1, at least 1.5:1, at least 1.8:1, at least 2:1, at least 3:1, at least 4:1, at least 5:1, at least 8:1, or even at least 10: 1. Additionally, in another non-limiting embodiment, the second aspect ratio (width: height) of the body 251 can be no greater than 100:1, such as no greater than 50:1, no greater than 10:1, no greater than 8:1, no greater than 6:1, no greater than 5:1, no greater than 4:1, no greater than 3:1, or even no greater than 2: 1. It should be understood that the second aspect ratio, expressed as width to height, may be within a range including between any of the minimum to maximum ratios noted above.

In another embodiment, the body 251 may have a third aspect ratio, expressed in length to height, which may be at least 1.1:1, such as at least 1.2:1, at least 1.5:1, at least 1.8:1, at least 2:1, at least 3:1, at least 4:1, at least 5:1, at least 8:1, or even at least 10: 1. Additionally, in another non-limiting embodiment, the third aspect ratio (length: height) of the body 251 can be no greater than 100:1, such as no greater than 50:1, no greater than 10:1, no greater than 8:1, no greater than 6:1, no greater than 5:1, no greater than 4:1, no greater than 3: 1. It should be understood that the third aspect ratio of the body 251 may be within a range including between any of the minimum and maximum ratios noted above.

The non-shaped abrasive particles 250 can have any of the attributes of the abrasive particles described in the embodiments herein, including, for example, but not limited to, composition, microstructural characteristics (e.g., average grain size), hardness, porosity, and the like.

The abrasive articles of the embodiments herein may incorporate different types of particles, including different types of abrasive particles, different types of secondary particles, or any combination thereof. For example, in one embodiment, a coated abrasive article may comprise: a first type of abrasive particles, the first type of abrasive particles comprising shaped abrasive particles; and a second type of abrasive particles. The second type of abrasive particles can be shaped abrasive particles or non-shaped abrasive particles.

FIG. 3 includes a cross-sectional illustration of a coated abrasive article incorporating a particulate material according to an embodiment. As shown, coated abrasive 300 may include a substrate 301 and a make coat 303 covering a surface of substrate 301. Coated abrasive 300 may further include: a first type of particulate material 305 in the form of a first type of shaped abrasive particles; a second type of particulate material 306 in the form of a second type of shaped abrasive particles; and a third type of particulate material 307, which may be secondary particles, such as diluent abrasive particles, non-shaped abrasive particles, fillers, and the like. The coated abrasive 300 may further include a size coat 304 overlying and bonded to the abrasive particulate materials 305, 306, 307, and the make coat 304. It is understood that other layers or materials may be added to other component layers of the substrate, including but not limited to front fillers, back fillers, etc. known to those of ordinary skill in the art.

According to one embodiment, the substrate 301 may include organic materials, inorganic materials, and combinations thereof. In certain examples, the substrate 301 may comprise a woven material. However, the substrate 301 may be made of a nonwoven material. Particularly suitable substrate materials may include organic materials including polymers, particularly polyesters, polyurethanes, polypropylenes, polyimides such as KAPTON from DuPont, paper, or any combination thereof. Some suitable inorganic materials may include metals, metal alloys, and especially foils of copper, aluminum, steel, and combinations thereof. In the context of a nonwoven substrate, which may be an open web of fibers, the abrasive particles may be adhered to the fibers by one or more adhesive layers. In such nonwoven products, the abrasive particles coat the fibers, but do not necessarily form a conformal layer overlying a major surface of the substrate, as shown in FIG. 3. It is to be understood that such nonwoven products are included in the examples herein.

The make coat 303 may be applied to the surface of the substrate 301 in a single process, or alternatively, the particulate materials 305, 306, 307 may be combined with the make coat 303 material and the combination of the make coat 303 and the particulate materials 305, 307 may be applied as a mixture to the surface of the substrate 301. In some cases, controlled deposition or placement of the particles 305-307 in the make coat may be better suited by separating the process of applying the make coat 303 from the process of depositing the abrasive particulate material 305-307 in the make coat 303. Additionally, it is contemplated that such processes may be combined. Suitable primer layer 303 materials can include organic materials, particularly polymeric materials, including, for example, polyesters, epoxies, polyurethanes, polyamides, polyacrylates, polymethacrylates, polyvinyl chloride, polyethylene, polysiloxanes, silicones, cellulose acetate, cellulose nitrate, natural rubber, starch, shellac, and mixtures thereof. In one embodiment, primer layer 303 may comprise a polyester resin. The coated substrate may then be heated to cure the resin and abrasive particulate material to the substrate. Typically, during this curing process, the coated substrate 301 may be heated to a temperature of about 100 ℃ to less than about 250 ℃.

The particulate material 305-307 may include different types of abrasive particles as described in embodiments herein. The different types of abrasive particles can include different types of shaped abrasive particles, different types of secondary particles, or combinations thereof. The different types of particles may differ from each other in composition, two-dimensional shape, three-dimensional shape, grain size, particle size, hardness, brittleness, agglomeration, or combinations thereof. As shown, the coated abrasive 300 can include a first type of shaped abrasive particles 305 having a generally pyramidal shape and a second type of shaped abrasive particles 306 having a generally triangular two-dimensional shape. Coated abrasive 300 can include different amounts of first type of shaped abrasive particles 305 and second type of shaped abrasive particles 306. It should be understood that the coated abrasive may not necessarily include different types of shaped abrasive particles, but may consist essentially of a single type of shaped abrasive particles. It is to be appreciated that the shaped abrasive particles of the embodiments herein can be incorporated into a variety of fixed abrasives (e.g., bonded abrasives, coated abrasives, nonwoven abrasives, thin wheels, cutting wheels, reinforced abrasive articles, etc.), including in the form of blends, which can include different types of shaped abrasive particles, secondary particles, and the like.

The particles 307 can be secondary particles different from the first type of shaped abrasive particles 305 and the second type of shaped abrasive particles 306. For example, the secondary particles 307 may include crushed abrasive grains representing non-shaped abrasive particles.

After the make coat 303 is sufficiently formed to contain the abrasive particulate material 305, 307 therein, a size coat 304 may be formed to cover and bond the abrasive particulate material 305 in place. Size layer 304 may include an organic material, may be made substantially of a polymeric material, and notably, polyesters, epoxies, polyurethanes, polyamides, polyacrylates, polymethacrylates, polyvinyl chloride, polyethylene, polysiloxanes, silicones, cellulose acetate, cellulose nitrate, natural rubber, starch, shellac, and mixtures thereof may be used.

Exemplary abrasive articles

FIG. 4 includes a top view of a portion of a coated abrasive according to an embodiment. Coated abrasive 400 may include a plurality of regions, such as first region 410, second region 420, third region 430, and fourth region 440. Each of the regions 410, 420, 430, and 440 may be separated by channel regions 450, wherein the channel regions 450 define particle-free regions of the backing. The channel region 450 may be of any size and shape, and may be particularly useful for chip removal and for improving grinding operations. The length (i.e., longest dimension) and width (i.e., shortest dimension perpendicular to the length) of a channel region may be greater than the average spacing between immediately adjacent abrasive particles in any of regions 410, 420, 430, and 440. For any of the embodiments herein, the channel region 450 is an optional feature.

As further shown, the first region 410 can include a set of shaped abrasive particles 411 having a substantially random rotational orientation relative to one another. The constituent shaped abrasive particles 411 may be arranged in a random distribution relative to one another such that there is no discernable short-range or long-range order with respect to the placement of the shaped abrasive particles 411. Notably, the compositionally shaped abrasive particles 411 can be substantially uniformly distributed within the first region 410, thereby limiting the formation of agglomerates (two or more particles in contact with each other). It should be understood that the particle weight of the compositionally shaped abrasive particles 411 in the first region 410 can be controlled based on the intended application of the coated abrasive.

The second region 420 can include a set of shaped abrasive particles 421 that are arranged in a controlled distribution relative to one another. Further, the compositionally shaped abrasive particles 421 can have a regular and controlled rotational orientation relative to one another. As shown, the compositionally shaped abrasive particles 421 can have substantially the same rotational orientation as defined by the same rotational angle on the backing of the coated abrasive 401. Notably, the compositionally shaped abrasive particles 421 can be substantially uniformly distributed within the second region 420, thereby limiting the formation of agglomerates (two or more particles in contact with each other). It should be understood that the particle weight of the compositionally shaped abrasive particles 421 in the second region 420 can be controlled based on the intended application of the coated abrasive.

The third region 430 can include multiple sets of shaped abrasive particles 421 and secondary particles 432. The compositionally shaped abrasive particles 431 and secondary particles 432 can be arranged in a controlled distribution relative to each other. In addition, the compositionally shaped abrasive particles 431 can have a regular and controlled rotational orientation relative to each other. As shown, the compositionally shaped abrasive particles 431 can generally have one of two types of rotational orientation on the backing of coated abrasive 401. Notably, the compositionally shaped abrasive particles 431 and secondary particles 432 can be substantially uniformly distributed within the third region 430, thereby limiting the formation of agglomerates (two or more particles in contact with each other). It should be understood that the grain weight of the compositionally shaped abrasive particles 431 and secondary particles 432 in the third region 430 can be controlled based on the intended application of the coated abrasive.

The fourth region 440 may include a set of shaped abrasive particles 441 and secondary particles 442 having a substantially random distribution relative to each other. In addition, the compositionally shaped abrasive particles 441 may have a random rotational orientation relative to one another. The compositionally shaped abrasive particles 441 and secondary particles 442 can be arranged in a random distribution relative to each other such that there is no discernable short-range or long-range order. Notably, the compositionally shaped abrasive particles 441 and secondary particles 442 can be substantially uniformly distributed within the fourth region 440, thereby limiting the formation of agglomerates (two or more particles in contact with each other). It should be understood that the particle weights of the compositionally shaped abrasive particles 441 and secondary particles 442 in the fourth region 440 can be controlled based on the intended application of the coated abrasive.

As shown in fig. 4, coated abrasive article 400 may include different regions 410, 420, 430, and 440, each of which may include different sets of particles, such as shaped particles and secondary particles. Coated abrasive article 400 is intended to illustrate different types of groupings, arrangements, and distributions of particles that may be produced using the systems and processes of the embodiments herein. This description is not intended to be limited to only those groups of particles, and it should be understood that a coated abrasive article may be made that includes only one region, as shown in FIG. 4. It should also be understood that other coated abrasive articles may be manufactured that include different combinations or arrangements of one or more of the regions shown in FIG. 4.

According to another embodiment, a coated abrasive article can be formed that includes different sets of abrasive particles, where the different sets have different angles of inclination relative to each other. For example, as shown in FIG. 5, a cross-sectional illustration of a portion of a coated abrasive is provided. Coated abrasive 500 can comprise backing 501 and first set of abrasive particles 502, wherein each abrasive particle in first set of abrasive particles 502 has a first average angle of inclination. The coated abrasive 500 can further include a second set of abrasive particles 503, wherein each of the abrasive particles 503 of the second set of abrasive particles 503 has a second average tilt angle. According to one embodiment, the first set of abrasive particles 502 and the second set of abrasive particles 503 may be separated by a channel region 505. Further, the first average tilt angle may be different from the second average tilt angle. In a more specific embodiment, the first set of abrasive particles can be oriented in an upright orientation and the second set of abrasive particles can be oriented in an inclined orientation. While not wishing to be bound by a particular theory, it is believed that the controlled variation of the tilt angle may facilitate improved performance of the coated abrasive for different sets of abrasive particles in different regions of the coated abrasive.

According to a particular aspect, the amount of abrasive particles overlying the backing can be controlled based on the intended application. For example, the abrasive particles may cover at least 5% such as at least 10%, or at least 20%, or at least 30%, or at least 40%, or at least 50%, or at least 60%, or at least 70%, or at least 80%, or at least 90% of the total surface area of the backing. In yet another embodiment, the coated abrasive article may be substantially free of silane.

Further, the abrasive articles of the embodiments herein can have a particular content of particles overlying the substrate. Furthermore, it should be noted that for certain levels of particles on the backing (such as the sparse coating density), the industry has found that it is challenging to achieve certain levels of particles in the desired perpendicular orientation. In one embodiment, the particles can define a loose coated abrasive product having particles (i.e., abrasive particles, secondary particles, or both abrasive particles and secondary particles) with a coating density of no greater than about 70 particles/cm². In other instances, the density of the shaped abrasive particles per square centimeter of the abrasive article can be not greater than about 65 particles/cm²Such as not greater than about 60 particles/cm²No greater than about 55 particles/cm ²Or even not greater than about 50 particles/cm². In addition, in one non-limiting embodiment, the density of the coated abrasive using the sparse coating of shaped abrasive particles herein can be at least about5 particles/cm²Or even at least about 10 particles/cm². It will be appreciated that the density of shaped abrasive particles per square centimeter of the abrasive article can be within a range between any of the minimum and maximum values noted above.

In certain instances, the sparse coating density of the abrasive article may be no greater than about 50% of the particles (i.e., abrasive particles or secondary particles or all of the abrasive particles and secondary particles) covering the outer abrasive surface of the article. In other embodiments, the area of the abrasive particles can be not greater than about 40%, such as not greater than about 30%, not greater than about 25%, or even not greater than about 20% relative to the total area of the surface on which the particles are disposed. Additionally, in one non-limiting embodiment, the coating percentage of the particles relative to the total area of the surface can be at least about 5%, such as at least about 10%, at least about 15%, at least about 20%, at least about 25%, at least about 30%, at least about 35%, or even at least about 40%. It will be appreciated that the percent coverage of particles for the total area of the abrasive surface can be within a range between any of the minimum and maximum values noted above.

For a given area (e.g., 1 order-30.66 m, where²) Some abrasive articles may have a particular content of particles (i.e., abrasive particles or secondary particles or all of the abrasive particles and secondary particles). For example, in one embodiment, the abrasive article may utilize at least about 1 pound per ream (14.8 grams/m)²) Such as a normalized particle weight of at least 5 pounds per ream, or at least 10 pounds per ream, or at least about 15 pounds per ream, or at least about 20 pounds per ream, or at least about 25 pounds per ream, or even at least about 30 pounds per ream. Additionally, in one non-limiting embodiment, the abrasive article can comprise not greater than about 90 pounds per ream (1333.8 grams/m)²) Such as a normalized particle weight of no greater than 80 pounds per ream, or no greater than 70 pounds per ream, or no greater than 60 pounds per ream, or no greater than about 50 pounds per ream, or even no greater than about 45 pounds per ream. It will be appreciated that the abrasive articles of the embodiments herein can utilize a normalized particle weight within a range between any minimum and maximum value noted above.

In some instances, the abrasive article may be used on a particular workpiece. Suitable exemplary workpieces may include inorganic materials, organic materials, natural materials, and combinations thereof. According to particular embodiments, the workpiece may comprise a metal or metal alloy, such as an iron-based material, a nickel-based material, or the like. In one embodiment, the workpiece may be steel, and more particularly, may consist essentially of stainless steel (e.g., 304 stainless steel).

In another embodiment, the fixed abrasive article may be a bonded abrasive comprising abrasive particles contained within a three-dimensional volume of a bond material, which may differ from certain other fixed abrasive articles, including, for example, coated abrasive articles that typically comprise a single layer of abrasive particles contained in a binder (such as a make coat and/or size coat). In addition, coated abrasive articles typically include a backing as a support for the layer of abrasive particles and binder. In contrast, bonded abrasive articles are typically self-supporting articles comprising a three-dimensional volume of abrasive particles, a bond material, and optionally some porosity. The bonded abrasive article may not necessarily include a substrate, and may be substantially free of a substrate.

FIG. 6 includes a perspective view illustration of a bonded abrasive article according to an embodiment. As shown, the bonded abrasive article 620 can have a generally cylindrical body 601 including an upper surface 624, a bottom surface 626, and a side surface 603 extending between the upper and bottom surfaces 624, 626. It should be understood that the fixed abrasive article of fig. 6 is a non-limiting example, and that other shapes of the body may be utilized, including but not limited to a cone, a cup, a concave center wheel (e.g., T42), and the like. Finally, as further shown, the body 601 may include a central opening 685 that may be configured to accept a rod or shaft for mounting the body 601 on a machine configured to rotate the body 601 and facilitate material removal operations.

The bonded abrasive article 620 can have a body 601 including abrasive particles including, for example, groups 605 and 628 of abrasive particles contained within a volume of the body 601. The abrasive particles may be contained within the three-dimensional volume of the body 601 by a bond material 607 that may extend throughout the entire three-dimensional volume of the body 601. According to one embodiment, the bonding material 607 may include materials such as glass, polycrystalline, single crystal, organic (e.g., resin), metal alloys, and combinations thereof.

In particular embodiments, the abrasive particles may be encapsulated within the bonding material 607. As used herein, "encapsulated" refers to a state in which at least one abrasive particle is completely surrounded by a uniform or substantially uniform bond material composition. In one embodiment, the bonded abrasive article 620 can be substantially free of a consolidation layer. In particular instances, the bonded abrasive article 620 can be substantially uniform throughout the volume of the body 601. In more particular instances, the body 601 may have a substantially uniform composition throughout the volume of the body 601.

According to one embodiment, the abrasive particles contained within the bonded abrasive article 620 may comprise abrasive materials according to those described in the embodiments herein.

The bonded abrasive article 620 can include a combination of abrasive particles including one or more types of abrasive particles, such as a primary type of abrasive particles and a secondary type of abrasive particles. The primary type and the secondary type may refer to a content of the abrasive particles within a body of the fixed abrasive article, wherein the content of the primary type of abrasive particles is higher than the content of the secondary type of abrasive particles. In other cases, the distinction between the primary type of abrasive particles and the secondary type of abrasive particles may be based on the location of the abrasive particles within the body, where the primary abrasive particles may be positioned to perform an initial stage of material removal or to perform a majority of material removal as compared to the secondary abrasive particles. In still other instances, the distinction between primary and secondary abrasive particles may relate to the abrasive properties (e.g., hardness, friability, fracture mechanics, etc.) of the abrasive particles, where the primary properties of the primary abrasive particles are typically stronger than the secondary type of abrasive particles. Some suitable examples of abrasive particles that may be considered secondary types of abrasive particles include diluent particles, agglomerated particles, unagglomerated particles, naturally occurring materials (e.g., minerals), synthetic materials, and combinations thereof.

In some instances, the bonded abrasive article 620 can include a particular content of abrasive particles within the body 601 that can facilitate suitable material removal operations. For example, the body 601 can include abrasive particles in an amount of at least 0.5 vol% and not greater than 60 vol% of the total volume of the body.

In addition, the body 601 of the bonded abrasive article 620 can include a particular content of the bonding material 607 that can facilitate proper operation of the bonded abrasive article 620. For example, the body 601 can include the bonding material 607 in an amount of at least 0.5 vol% and not greater than about 90 vol% of the total volume of the body.

In certain instances, the fixed abrasive article may have a body 601 that includes a certain porosity content. The porosity may extend through at least a portion of the entire volume of the body 601, and in some cases, may extend substantially uniformly through the entire volume of the body 601. For example, porosity may include closed porosity or open porosity. The closed porosity may be in the form of discrete pores separated from each other by bond material and/or abrasive particles. Such closed porosity may be formed by pore formers. In other instances, the porosity can be an open porosity defining an interconnected network of channels extending through at least a portion of the three-dimensional volume of the body 601. It should be understood that the body 601 may include a combination of closed porosity and open porosity.

According to one embodiment, a fixed abrasive article may have a body 601 that includes a particular content of porosity that may facilitate a suitable material removal operation. For example, the body 601 can have a porosity of at least 0.5 vol% and no greater than 80 vol% of the total volume of the body.

According to another embodiment, it should be appreciated that the bonded abrasive article 620 may include a body 601 containing certain additives that may facilitate certain abrading operations. For example, the body 601 may include a variety of additives such as fillers, grinding aids, pore inducing agents, hollow materials, catalysts, coupling agents, curing agents, antistatic agents, suspending agents, anti-loading agents, lubricants, wetting agents, dyes, fillers, viscosity modifiers, dispersants, defoamers, and combinations thereof.

As further shown in fig. 6, the body 601 may have a diameter 683, which may vary depending on the desired material removal operation. The diameter may refer to the largest diameter of the body, particularly if the body 601 has a conical or cup-shaped profile.

Further, body 601 may have a particular thickness 681 extending along side surface 603 along axial axis 680 between upper surface 624 and bottom surface 626. The body 601 may have a thickness 681, which may be an average thickness of the body 601, which may be no greater than 1 m.

According to one embodiment, body 601 may have a particular relationship between diameter 683 and thickness 681, defining a diameter to thickness ratio that may be suitable for certain material removal operations. For example, the body 601 may have a diameter to thickness ratio of at least 10:1, such as at least 15:1, at least 20:1, at least 50:1, or even 100: 1. It is to be understood that the body may have a diameter to thickness ratio of no greater than 10,000:1, or no greater than 1000: 1.

The bonded abrasive article 620 can include at least one reinforcing member 641. In certain instances, the reinforcing material 641 can extend for a majority of the entire width (e.g., diameter 683) of the body 601. In other cases, however, reinforcing member 641 may extend only a small portion of the entire width (e.g., diameter 183) of body 601. In some cases, a stiffening member 641 may be included to add suitable stability to the body for certain material removal operations. According to one embodiment, the stiffening member 641 may comprise a material such as a woven material, a nonwoven material, a composite material, a laminate material, a monolith material, a natural material, a synthetic material, and combinations thereof. More specifically, in some cases, the stiffening member 641 may include a material such as a single crystalline material, a polycrystalline material, a glass material, an amorphous material, a glass (e.g., glass fiber), a ceramic, a metal, an organic material, an inorganic material, and combinations thereof. In particular instances, the reinforcing material 641 may include glass fibers and may be formed substantially of glass fibers.

In particular instances, the reinforcing material 641 can be substantially contained within a three-dimensional volume of the body 601, and more particularly, within a three-dimensional volume of the bonding material 607. In some cases, the reinforcing material 641 may intersect an outer surface of the body 601 including, but not limited to, the upper surface 624, the sides 603, and/or the bottom surface 626. For example, the reinforcing material 641 may intersect the upper surface 624 or the bottom surface 626. In at least one embodiment, the reinforcing material 641 can define an upper surface 624 or a lower surface 626 of the body 601 such that the bonding material 607 is disposed between one or more reinforcing materials. It should be understood that although a single stiffening member 641 is illustrated in the embodiment of fig. 6, a plurality of stiffening members may be provided within body 601 in a variety of arrangements and orientations that may be suitable for the intended material removal application.

As further shown, the body 601 may include certain axes and planes that define a three-dimensional volume of the body 601. For example, the body 601 of the fixed abrasive article 620 can include an axial axis 680. As further shown along axial axis 680, body 601 may include a first axial plane 631 extending along axial axis 680 and extending through a particular diameter of body 601 at a particular angular orientation (designated herein as 0 °). The body 601 may further include a second axial plane 632 different from the first axial plane 631. The second axial plane 632 may extend along an axial axis 680 and pass through a diameter of the body 601 at an angular position, as exemplarily designated herein as 30 °. First axial plane 631 and second axial plane 632 of body 601 can define a particular axial collection of abrasive particles within body 601, including, for example, an axial collection of abrasive particles 691 within axial plane 631 and an axial collection of abrasive particles 692 within axial plane 632. Further, the axial planes of the body 601 may define sectors therebetween, including, for example, sector 684, which is defined as the area between the axial planes 631 and 632 within the body 601. The sectors may include groups of abrasive particles that may be specified, which facilitate improved material removal operations. References herein to features of the abrasive particle portion within the body, including, for example, abrasive particles in the axial plane, will also relate to the abrasive particle groups contained within one or more sectors of the body.

As further shown, the body 601 may include a first radial plane 621 that extends along a plane substantially parallel to the upper surface 624 and/or the bottom surface 626 at a particular axial location along the axial axis 680. The body may further include a second radial plane 622 that may extend in a substantially parallel manner to the upper surface 624 and/or the bottom surface 626 at a particular axial location along the axial axis 680. The first and second radial planes 621 and 622 may be separated from each other within the body 601, and more particularly, the first and second radial planes 621 and 622 may be axially separated from each other. As further shown, in some cases, one or more reinforcing members 641 may be disposed between the first radial plane 621 and the second radial plane 622. The first radial plane 621 and the second radial plane 622 may include one or more specific abrasive particle groups, including, for example, the abrasive particle group 628 of the first radial plane 621 and the abrasive particle group 605 of the second radial plane 622, which may have certain characteristics relative to each other that may facilitate improved abrading performance.

The abrasive particles of the embodiments herein can include a particular type of abrasive particle. For example, the abrasive particles can include shaped abrasive particles and/or elongated abrasive particles, wherein the elongated abrasive particles can have an aspect ratio in length to width or length to height of at least 1.1: 1. Various methods may be utilized to obtain shaped abrasive particles. The particles may be obtained from commercial sources or manufactured. Some suitable processes for making shaped abrasive particles may include, but are not limited to, deposition, printing (e.g., screen printing), molding, pressing, casting, slicing, cutting, dicing, punching, stamping, drying, curing, coating, extruding, rolling, and combinations thereof. Similar processes can be used to obtain elongated abrasive particles. Elongated, unshaped abrasive particles can be formed by comminution and sieving techniques.

FIG. 7A illustrates a cross-sectional view of an abrasive article 730 according to an example embodiment. The abrasive article 730 includes an abrasive portion 732 and a non-abrasive portion 731, and an electronic assembly 720 coupled to the non-abrasive portion 731 of the abrasive article 730. The non-abrasive portion 731 may have a first surface 733, a second surface 734, and a side surface 735 extending between the first surface 733 and the second surface 734. The first surface 733 and the second surface 734 may be primarily flat surfaces. The second surface 734 may be a primarily flat surface having the same size or a different size relative to the first surface 733. As further shown, the non-abrasive portion 731 can include an opening 705, such as a spindle bore. The electronic component 720 may be coupled to the first surface 733. The electronic components 720 may include electronics 722 and packaging 721 as described in embodiments herein. In one embodiment, the electronic assembly 720 may include at least one electronic device 722 that may be housed within a package 721. The package 721 may be suitable for attaching the electronic assembly 720 to the body of the abrasive article 730 and may provide some suitable protection for the one or more electronic devices contained therein. In a particular example, the electronic device 722 can be enclosed within a package 721.

According to one embodiment, the electronic device 722 may be configured to write information, store information, or provide information to other objects during a read operation. Such information may be related to the manufacture of the abrasive article, the operation of the abrasive article, or the conditions encountered by the electronic assembly 720. Reference herein to an electronic device is to be understood as a reference to at least one electronic device, which may comprise one or more electronic devices. In at least one embodiment, the electronic device 722 may include at least one device selected from the group consisting of: an integrated circuit and chip, a data transponder, a radio frequency based tag or sensor with or without a chip, an electronic tag, an electronic memory, a sensor, an analog-to-digital converter, a transmitter, a receiver, a transceiver, a modulator circuit, a multiplexer, an antenna, a near field communication device, a power supply, a display (e.g., an LCD screen or an OLED screen), an optical device (e.g., an LED), a Global Positioning System (GPS) or device, or any combination thereof. In some examples, the electronic device may optionally include a substrate, a power source, or both. In one particular embodiment, the electronic device 722 may include a tag, such as a passive Radio Frequency Identification (RFID) tag. In another embodiment, the electronic device 722 may include an active Radio Frequency Identification (RFID) tag. Active RFID tags may include a power source, such as a battery or an inductive-capacitive (LC) tank circuit. In another embodiment, the electronic device 1722 may be wired or wireless.

According to one aspect, the electronic device 722 may include a sensor. The sensors may be selectively operated by any system and/or individual in the supply chain. For example, the sensor may be configured to sense one or more process conditions during formation of the abrasive article. In another embodiment, the sensor may be configured to sense a condition during use of the abrasive article. In yet another embodiment, the sensor may be configured to sense a condition in the environment of the abrasive article. The sensors may include acoustic sensors (e.g., ultrasonic sensors), force sensors, vibration sensors, temperature sensors, humidity sensors, pressure sensors, gas sensors, timers, accelerometers, gyroscopes, or any combination thereof. The sensor may be configured to issue the particular condition sensed by the sensor to any system and/or individual associated with the abrasive article, such as a manufacturer and/or customer. The sensors may be configured to generate an alarm signal to one or more systems and/or individuals in the supply chain (including but not limited to manufacturers, distributors, customers, users, or any combination thereof).

In another embodiment, electronic device 722 may comprise a near field communication device. A near field communication device may be any device capable of transmitting information via electromagnetic radiation within a certain defined radius of the device, typically less than 20 meters. The near field communication device may be coupled to one or more electronic devices, including, for example, sensors. In one particular embodiment, the sensor may be coupled to a near field communication device and configured to relay information to one or more systems and/or individuals in the supply chain via the near field communication device.

In an alternative embodiment, the electronic device 722 may include a transceiver. A transceiver may be a device that may receive information and/or transmit information. Unlike passive RFID tags or passive near field communication devices, which are typically read-only devices that store information for read operations, the transceiver can actively transmit information without having to perform active read operations. Further, the transceiver may transmit information over various selected frequencies, which may improve the ability of the electronic component to communicate with various systems and/or individuals in the supply chain.

Fig. 7B illustrates a cross-sectional view of an electronic assembly, according to an example embodiment. According to an aspect, electronic component 720 may include one or more electronic devices including, for example, electronic device 756 and electronic device 757. In some cases, the electronic assembly 720 can include a substrate 759 upon which the one or more electronic devices 756 and 757 can be disposed. In other cases, the electronic component 720 may further include a first portion 771 and a second portion 772. The first 771 and second 772 portions may be part of a package that may cover at least a portion of the electronic component 720. The package 721 may consist essentially of a first portion 771 and a second portion 772. For example, as shown in fig. 7B, the first portion 771 can underlie the substrate 759 and the one or more electronic devices 756 and 757. In some cases, the first portion 771 can be coupled to (e.g., in direct contact with) the second portion 772. In yet another embodiment, the electronic assembly 720 can include a first portion 771 that underlies and partially surrounds at least a portion of the substrate 1759 and the one or more electronic devices 756 and 757. The second portion 772 can cover at least a portion of the one or more electronic devices 756 and 757. The second portion 772 can be indirectly coupled or directly coupled (e.g., directly contacted or bonded) to the first portion 771. As shown, the first 771 and second 772 portions may substantially surround the entire one or more electronic devices 756 and 757 and substrate 759.

The first portion 771 can underlie at least a portion (e.g., at least 50%) of the electronic device 757. The first portion 771 can electrically insulate and isolate the electronic device 757 from the non-abrasive portion to which it is coupled. In particular instances, the first portion 771 can be disposed between and electrically insulate at least one of the at least one or more electronic devices 756 and 757 from the body of the abrasive article. More specifically, the electronic devices 756 and/or 757 may include at least one antenna, and the first portion 771 may be disposed between and electrically isolate the antenna from the body of the abrasive article.

In some cases, the second portion 772 may serve as a protective layer. In some examples, the substrate may serve as a protective layer or may facilitate bonding of the electronic component to the body to avoid the use of a protective layer disposed below the substrate. In another example, a protective layer may be disposed below the electronic device, and the upper surface and the side surfaces of the electronic device 757 or 756 may not be covered by the protective layer. In another embodiment, the electronic component 720 may include an additional protective layer disposed above and/or below the second portion for providing additional protection. The second portion 772 may act as a protective layer to limit the effects of coolant and swarf on the electronic assembly. In other cases, the protective layer may protect the electronic device from mechanical or chemical damage during remodeling, dressing, maintenance, etc. of the abrasive or non-abrasive portion.

Fig. 8A illustrates a top view of a releasable coupling of an electronic component on a body, according to an example embodiment. As shown, the body 801 may include an upper surface 802. Electronic components 803 may be contained within cavities 820 in the body 801. The electronic component 803 may be press fit in the cavity 820. The securing assembly 830 including the securing element 831 can be configured to translate from an engaged position to a disengaged position. In the engaged position, as shown in fig. 8A, the securing element 831 can cover and engage the electronic component 803, thereby securing the electronic component 803 to the body 801. In the disengaged position, the securing element 831 can be spaced apart and disengaged from the electronic component 803. The fixation element 831 may articulate between an engaged position and a disengaged position by translating in the Y direction. In the disengaged position, the electronic component 803 is in an unsecured position and can be easily removed from the body 801. In such cases, removal of the electronic component 803 from the body 801 may be accomplished without the need to apply heat or other chemical additives to remove or dissolve the adhesive.

Fig. 8B illustrates an abrasive system 850 according to an example embodiment. Abrasive system 850 includes a housing 851 and a body 852 contained within housing 851. The body 852 may include an electronic component 853 coupled to the body 952. As shown, the body 852 may be a particular type of edge grinding tool, wherein the workpiece 1961 may be a piece of glass. The housing 851 can further include a coolant 854 applied to the abrasive interface during the material removal operation. In one embodiment, the housing 851 may include at least one electronic device 855. The at least one electronic device 855 may be coupled to a surface or embedded in the material of the housing 851. The electronic components 853 include one or more electronic devices configured to communicate with the one or more electronic devices 855 in the housing 851. The information received by the electronic device 855 may relate to a remote electronic device 856 located outside the housing 851.

As further shown, the workpiece 861 can include one or more electronic devices 857 coupled to the workpiece 861 and configured to send and/or receive information from one of the other electronic devices (such as electronic component 853, electronic device 855, and/or electronic device 856). In certain instances, it may be suitable for the electronic component 853 to include a protective layer configured to prevent the corrosive effects of the coolant 854.

In an alternative embodiment, the electronic component 853 may also be coupled to, partially embedded in, or fully embedded in the surface 858 of the body 852. The placement and positioning of electronic components may facilitate improved communication with electronic devices 855, 856, and/or 857. Further, in some cases, electronics 855, 856, 857, and/or electronics 853 can utilize a vertically polarized antenna, a booster antenna, a 3D polarized antenna, or any combination thereof. It should also be understood that in some cases, it may be appropriate to use multiple electronic components located at different positions and orientations on the body 852.

Exemplary microscopic interactions

Fig. 9A depicts various types of interactions associated with the grinding process. Each type of interaction may be incorporated into the analytical models and/or machine learning methods and systems described herein. For example, abrasive interactions may include cutting (material removal), plowing (material displacement), or sliding (surface modification) effects. The wear process involves sliding a hard material (e.g., abrasive particles) against a softer material, during which the softer material may deform and surface modify. In some cases, this may be due to deeper scoring or plowing, such as moving the work material without material removal or slippage between the abrasive particles and the workpiece. If the depth of penetration of the abrasive particles into the work material is sufficiently large, the abrasive particles can act as cutting edges, resulting in the creation of new surfaces and the removal of debris called "swarf" from the working surface. If the penetration depth is insufficient, the hard abrasive particles may locally deform the work material. This interaction or deformation is commonly referred to as plowing. Finally, if the depth of penetration of the abrasive particles into the work material is extremely shallow, the result will be that the abrasive particles slide against the work material, despite the high contact stresses. The surface created at the end is the cumulative effect of all these abrasive/workpiece interactions during the abrading process. Additionally or alternatively, various chip/binder, chip/workpiece, binder/workpiece, and/or other sliding interactions are also possible and contemplated.

Fig. 9B depicts an exemplary force interaction and associated analytical model for a grinding process. As an example, the Material Removal Rate (MRR) may increase in proportion to the tangential force and/or the normal force. Further, fig. 9B illustrates MRR as a function of the components of the applied force (e.g., tangential force Ft and normal force Fn), which follows a relationship governed by principles of machining and tribology. As an example, for a given material removal rate, the tangential force Ft is the force Ft required to form the chip_c+ frictional force Ft_fThreshold force Ft when + time is zero_th(0) Threshold force at + time t Ft_th(t) of (d). A similar relationship applies to the normal force component.

Fig. 9C depicts an exemplary power interaction and associated analytical model of the grinding process. For example, fig. 9C illustrates the change in grinding power after time "t" at a given MRR and its associated four components: (1) initial threshold power P_th(0) (2) variation of threshold Power with time P_th(t), (3) Power P for cutting or chip production_cAnd (4) P caused by chip friction effect_cVariation P_f(t) of (d). It should also be understood that fig. 9A-9C present conceptual representations of exemplary microscopic interactions and are not intended to be limiting with respect to the types of microscopic interactions, analytical models, or grinding processes that may be used with the present disclosure.

V. exemplary computing device

FIG. 10 illustrates a block diagram of a computing device 1000, according to an example embodiment. In particular, computing device 1000 may be configured to perform at least functions related to and/or related to machine learning platform 1110, enterprise 1120, external provider 1130, 3 rd party user 1140, machine learning system 1210, manual abrasive device 1310, wearable device 1320, automated abrasive device 1330, server device 1340, other sensors 1350, abrasive product 1410, remote device 1420, provider 1430, analysis platform 1440, method 1500, method 1600, method 1700, microcontroller 1810B, controller 2020, remote network 2110, client network 2120, mobile device 2400, and/or other elements described herein.

The computing device 1000 may include: one or more sensors 1016 to collect data; a data store 1004 that can store collected data and can include instructions 1014; one or more processors 1002; a communication interface 1006 for communicating with a remote source (e.g., a server or another device/sensor); and a display 1008. Additionally, computing device 1000 may include an audio output device (e.g., a speaker) and a haptic feedback device (e.g., an Eccentric Rotating Mass (ERM) actuator, a Linear Resonant Actuator (LRA), or a piezoelectric actuator, among other examples).

The processor 1002 may include one or more general-purpose processors or special-purpose processors (e.g., GPUs). The processor 1002 may be configured to execute the computer-readable instructions 1014. For example, the processor 1002 may control the one or more sensors 1016 based at least in part on the computer readable instructions 1014. The processor 1002 may be configured to process real-time data collected by the one or more sensors 1016.

The data storage 1004 is a non-transitory computer readable medium that may include, but is not limited to, magnetic disks, optical disks, organic memory, and/or any other volatile (e.g., RAM) or non-volatile (e.g., ROM) storage system that is readable by the processor 1002. Data store 1004 may include a data store for storing data indications such as sensor readings, machine learning models, program settings (e.g., to adjust the behavior of computing device 1000), user inputs (e.g., from a user interface on device 1000 or transmitted from a remote device), and so forth. The data memory 1004 may also include program instructions 1014 that are executed by the processor 1002 to cause the device 1000 to perform the operations specified by the instructions. The operation may include any of the methods described herein.

Communication interface 1006 may include hardware to enable communication within computing device 1000 and/or between computing device 1000 and one or more other devices. The hardware may include, for example, a transmitter, a receiver, and an antenna. Communication interface 1006 may be configured to facilitate communication with one or more other devices according to one or more wired or wireless communication protocols. For example, communication interface 1006 may be configured to facilitate wireless data communication for computing device 1000 in accordance with one or more wireless communication standards, such as one or more IEEE 801.11 standards, a ZigBee standard, a bluetooth standard, etc. For example, communication interface 1006 may include WiFi connectivity and access rights for cloud computing and/or cloud storage capabilities. As another example, communication interface 1006 may be configured to facilitate wired data communication with one or more other devices.

Display 1008 may be any type of display component configured to display data. As one example, the display 1008 may comprise a touch screen display. As another example, the display 1008 may include a flat panel display, such as a Liquid Crystal Display (LCD) or a Light Emitting Diode (LED) display.

The user interface 1010 may include one or more pieces of hardware for providing data and control signals to the computing device 1000. For example, the user interface 1010 may include a mouse or pointing device, a keyboard or keypad, a microphone, a touchpad, or a touch screen, among other possible types of user input devices. In general, the user interface 1010 may enable an operator to interact with a Graphical User Interface (GUI) provided by the computing device 1000 (e.g., displayed by the display 1008). As an example, the user interface 1010 may allow an operator to provide input data to the computing device 1000. As another example, the operator may provide input indicative of a product to be used to perform an operation and/or input indicative of a workpiece on which the operator may perform an abrasive operation.

In some embodiments, a user may utilize the GUI to provide a desired level of operation (e.g., a maximum desired vibration level, a maximum desired noise level, etc.), which may be based on user preferences and/or user comfort, for example. It should be understood that the user may provide information indicating the desired level of operation in other ways as well.

One or more sensors 1016 may be configured to collect data from or associated with the environment of computing device 1000 in real-time. Real-time collection of data may involve the sensor collecting data periodically or continuously. For example, the one or more sensors 1016 can include a sound detection device (e.g., a microphone) configured to detect sound in the environment of the sensor (e.g., from an abrasive device operating in proximity to the sensor). Additionally and/or alternatively, the sensors 1016 may be configured to collect data from or associated with an operator of the computing device 1000. For example, the one or more sensors 1016 may include an accelerometer. As described herein, data collected by the one or more sensors 1016 may be used to determine abrasive operation data, which may then be used to obtain real-time data regarding grinding/abrasive operations, capture user experience of a user using the device, and/or determine operations and/or business improvements (e.g., based on data collected over a period of time).

The one or more sensors 1016 may also include other sensors for detecting motion, such as IMUs and gyroscopes. Further, the one or more sensors 1016 may include other types of sensors, such as location tracking sensors (e.g., GPS or other positioning devices), light intensity sensors, thermometers, clocks, force sensors, pressure sensors, photoelectric sensors, hall sensors, vibration sensors, sound pressure sensors, magnetometers, infrared sensors, cameras, and piezoelectric sensors, among other examples. The sensor and its components can be miniaturized.

Exemplary machine learning platform

Fig. 11 illustrates an arrangement 1100 of a machine learning platform 1110 according to an example embodiment. As shown in fig. 11, machine learning platform 1110 is communicatively coupled to enterprise 1120, external vendor 1130, and party 3 user 1140. The machine learning platform 1110 can include, for example, a machine learning system 1112, a database device 1114, a server device 1116, and an analytics platform 1118. Machine learning platform 1110 can utilize machine learning to process and/or analyze sensor data collected by enterprise 1120. The machine learning platform 1110 can store the received sensor data and then analyze the data to provide product-specific information for the abrasive product and/or workpiece-specific information associated with the abrasive product on the enterprise 1120. As used herein, product-specific information may refer to any information relating to an element of an abrasive product/apparatus or an element of any abrasive operation/process performed by the abrasive product/apparatus. For example, machine learning platform 1110 can determine the best operating practices for enterprise 1120. In another example, machine learning platform 1110 can determine different value metrics (e.g., productivity, product life, etc.) for different abrasive products. As used herein, an abrasive product may refer to a device associated with or embodied by an abrasive tool.

Machine learning system 1112 may include one or more machine learning models configured to receive sensor data from enterprise 1120. For example, the sensor data may relate to abrasive products and to abrasive operating modes, specific workpieces, specific abrasive tools, or specific abrasive conditions from the enterprise 1120. In response to receiving the sensor data, the machine learning system 1112 may train the one or more machine learning models to predict product-specific information and/or workpiece-specific information related to the received sensor data. After one or more machine learning models have been trained, machine learning system 1112 may be applied at runtime to predict or infer predicted conditions based on real-time data received from enterprise 1120. As described herein, a predicted condition may trigger, prompt, or initiate various events, such as a notification, a report, a command, or another type of action

The database device 1114 may include one or more computing devices configured to store data in one or more databases. For example, the database device may include one or more relational databases (e.g., SQL), graphical databases (e.g., neo4j), document databases (e.g., MongoDB), columnar databases (e.g., Cassandra), and/or other database models. Database device 1114 may act as a data store for the components of machine learning platform 1110. For example, database device 1114 may be configured to receive and store sensor data from enterprise 1120 and provide the sensor data to machine learning system 1112 for use in training one or more machine learning models. In some instances, the database device 1114 may be configured to serve as a primary data source for the analytics platform 1118. In other examples, the database device 1114 may be configured to store one or more trained models (e.g., learned parameters).

Server device 1116 may include one or more network servers, file servers, and/or computing servers. The server device may facilitate communication between machine learning platform 1110 and enterprise 1120, external providers 1130, and 3 rd party users 1140. Communications may be facilitated by known network communication protocols, such as TCP/IP. In some embodiments, the server device 1116 may be used by the machine learning system 1112 or the analytics platform 1118 for computing tasks. For example, the devices in server device 1116 may be part of a MapReduce cluster used as part of a distributed training architecture for machine learning system 1112.

The analytics platform 1118 may include a web application configured to utilize information collected from the machine learning system 1112 and the database device 1116. After processing the collected information, analysis platform 1118 may generate various forecasted future conditions for enterprise 1120 and various regulatory actions for enterprise 1120. As used herein, predicting a future condition refers to an estimate of a future event that may occur at enterprise 1120. Examples of future events may include predicted failure of the abrasive product/workpiece, prediction of potential damage to the abrasive product/workpiece, or prediction that the workpiece quality does not meet a predetermined quality level, among other possibilities. Further, as used herein, a normative action refers to a recommendation for an optimal course of action given the current state and/or current situation of the abrasive product and/or given the current state and/or current situation of the enterprise 1120. Examples of normative actions may include a command to shut down the abrasive product when the abrasive product exhibits abnormal behavior, a command to adjust the speed of the grinding wheel, a notification to replace the abrasive article of the abrasive product, or a notification to dress a damaged abrasive product, among other possibilities.

In some embodiments, analysis platform 1118 includes a simulated environment programmed with a digital version of a physical abrasive product used by enterprise 1120 (e.g., "digital twins"). The simulation environment may use these digital versions to estimate productivity, cost, and/or damage due to the addition/reconfiguration/removal of different digital abrasive products from the simulation environment. In some embodiments, the analysis platform 1118 is configured to graphically display metrics associated with one or more abrasive products and/or one or more workpieces in the enterprise 1120. More details regarding the analysis platform 1118 are provided below.

Notably, the configuration of the machine learning platform 1110 is provided as an example. In some cases, machine learning platform 1110 can include one or more additional devices. For example, machine learning platform 1110 can include a firewall to allow access from authorized users, deny access from unauthorized users, provide intrusion detection, facilitate virus scanning, and/or provide other network security services. As another example, machine learning platform 1110 can include one or more load balancers to distribute incoming network traffic or requests across multiple computing devices within machine learning platform 1110 (e.g., such that no single device is overwhelmed by task requests). In other examples, machine learning platform 1110 can include one or more routers, virtual machines, proxy servers, and/or other common network devices. The machine learning platform 1110 may also be connected to one or more client devices (e.g., personal computers or mobile phones). In some examples, machine learning platform 1110 can provide Virtual Private Network (VPN) services.

Additionally and/or alternatively, components of machine learning platform 1110 can be replicated across multiple computing devices to provide data replication and increase the capacity of the service. These computing devices may be located in different physical locations to ensure high availability in the event of a failure at one location. Thus, machine learning platform 1110 may be configured across different physical locations and hundreds of computing devices.

Enterprise 1120 may include, for example, one or more abrasive products 1122, wearable device 1124, server device 1126, and remote device 1128. Enterprise 1120 may represent a single geographic location containing multiple grinders or may represent multiple grinders located in several geographic locations. Further, enterprise 1120 may represent a single enterprise of a plurality of enterprises that utilize products manufactured or maintained by entities operating machine learning platform 1110. Thus, the machine learning platform 1110 can act as a remote customer support system for these products.

Abrasive product 1122 may include one or more devices or tools that perform an abrading operation on a workpiece. As described above, abrasive product 1122 may be manufactured or maintained by an entity operating machine learning platform 1110. Abrasive product 1122 can include one or more sensors that collect wear operation data associated with a grinding operation or related to a workpiece being ground. For example, the one or more sensors may transmit the collected wear operation data to server device 1126 via bluetooth, TCP/IP, or other networking protocols. In another example, the one or more sensors may transmit the collected wear operation data to the machine learning platform 1110.

Wearable device 1124 can include a wearable computing device having one or more sensors that continuously or periodically collect data from or associated with the environment of abrasive product 1122 and/or collect data from or associated with the operator's abrasive product 1122. For example, data collected by wearable device 1124 can be used to determine abrasive operation data. In some instances, the collected data may be sent to server device 1126 (e.g., via bluetooth, TCP/IP, or other networking protocols). In other examples, the collected data may be transmitted directly to machine learning platform 1110.

Server devices 1126 may include one or more computing devices located on enterprise 1120. The server device may be configured to receive and aggregate sensor data from abrasive product 1122 and wearable device 1124. Server device 1126 may be operated by machine learning platform 1110 or by enterprise 1120. Upon receiving the sensor data, server device 1126 may apply data filters to the sensor data, such as removing outlier sensor data from one or more wearable devices 1124 or abrasive products 1122 and/or ignoring sensor data from the one or more wearable devices or abrasive products. In some instances, server device 1126 may be configured to convert sensor data into a different data format that is more suitable for machine learning platform 1110, such as into JavaScript object notation (JSON). As another example, server device 1126 may allow a human operator to tag sensor data with tags, as described further herein. Server device 1126 may receive product-specific information and/or workpiece-specific information from machine learning platform 1110 and distribute that information to remote device 1128, abrasive product 1122, wearable device 1124, or may store that data for later access by members of enterprise 1120.

In some embodiments, server device 1126 may provide sensor data to machine learning platform 1110 by grouping data in batches. Batches may be transferred periodically, for example every 10 minutes or 30 minutes. In other instances, server device 1126 may send sensor data machine learning platform 1110 in a streaming format in real-time. In some embodiments, server device 1126 may be configured to monitor sensors disposed in abrasive product 1122 and wearable device 1124. For example, server device 1126 may send a heartbeat message to a sensor, which may in turn be configured to respond with a responsive heartbeat message. This may ensure that the sensors are operational and do not stop sending data to server device 1126, for example, because of a failure or power outage.

Remote device 1128 may include an interface to one or more computing devices located within enterprise 1120. For example, the remote device 1128 may include a wearable device (e.g., a smart watch), a mobile device (e.g., a mobile phone or tablet computer), and/or a monitor (e.g., a computer screen). Remote device 1128 can receive data from server device 1126 or machine learning platform 1110 and display output data or issue alerts, warnings, notifications, reports, commands, and/or other types of actions on a Graphical User Interface (GUI).

External provider 1130 may represent one or more computing systems managed by partners of the entity operating machine learning platform 1110. In an example embodiment, machine learning platform 1110 can transmit new order requests, delivery requests, and/or other logistics requests to external vendor 1130 based on predictions made by machine learning system 1112. These requests may be made automatically by machine learning platform 1110 on behalf of enterprise 1120.

The 3 rd party user 1140 may include one or more individuals or organizations that utilize the capabilities of the analytics platform 1118. For example, the 3 rd party user 1140 may access the analysis platform 1118 via a web browser and may be able to access data provided to the analysis platform 1118 by the machine learning platform 1110. For example, 3 rd party user 1140 may be authorized for access through a subscription-based model. The analytics platform 1118 may provide multiple levels of access rights for the 3 rd party user 1140, each level of access being based on subscriptions purchased by the 3 rd party user 1140. For example, each access level may provide a more sensitive or larger amount of data.

It is noted that the components of arrangement 1100 are for illustrative purposes. Other components and arrangements are also possible.

Exemplary machine learning System

FIG. 12 depicts a scenario 1200 that includes an enterprise 1202, a machine learning system 1210, an input request 1220, and an output prediction 1230, according to an example embodiment. The scenario 1200 may occur as part of the arrangement 1100. Thus, enterprise 1202 may be embodied by enterprise 1120, machine learning system 1210 may be embodied by machine learning system 1112, and training data 1212 may be embodied by database device 1114.

Enterprise 1202 may represent, for example, an organization that utilizes products manufactured or maintained by operators of machine learning system 1210. As discussed above, enterprise 1202 may include sensors that generate wear operation data associated with abrasive operations involving one or more abrasive products or one or more workpieces. Data from these sensors may be transmitted from enterprise 1202 to machine learning system 1210 for determining product-specific information and workpiece-specific information for the one or more abrasive products based on the sensor data.

Machine learning system 1210 may utilize machine learning techniques to train one or more machine learning models with training data to detect patterns and provide output predictions about the training data. The resulting trained machine learning model may be referred to as a trained machine learning model. For example, scenario 1200 illustrates machine learning model 1214 trained on training data 1212 to become trained machine learning model 1216. During the prediction time, the trained machine learning model 1216 may receive the input request 1220 and responsively provide an output prediction 1230.

Training data 1212 may include one or more databases designed to receive and store sensor data from enterprise 1202 and provide sensor data to train one or more machine learning models. For example, the training data 1212 may include a relational database (e.g., SQL), a graphical database (e.g., neo4j), a document database (e.g., MongoDB), a columnar database (e.g., Cassandra), and/or other database models.

The machine learning models 1214 may include, but are not limited to, algorithms such as logical or linear regression, Support Vector Machines (SVMs), bayesian networks, Artificial Neural Networks (ANN), including Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), naive bayes classifiers, K-nearest neighbors, autoencoders, Hidden Markov Models (HMMs), markov decision processes, decision trees, random forests, integration methods, including boosting and bagging, and/or heuristic machine learning models. Machine learning models incorporating rule-based algorithms (e.g., association rule models, learning classifier models, etc.) or iterative algorithms (e.g., expectation-maximization algorithms) are also contemplated and are possible within the scope of the present application.

The machine learning model 1214 may be trained using online or offline learning. During training, the machine learning model 1214 may utilize the training data 1212 to adjust the weights and/or other parameters of one or more models. The machine learning model 1214 may use various regularization techniques during training (such as L1 regularization, L2 regularization, early termination of training, and/or a breakout method) to reduce overfitting. The machine learning model 1214 may use various optimization methods during training, such as gradient-based methods (bulk gradient descent, random gradient descent, Adam optimization), search-based techniques (genetic algorithms, grid search, random search), and/or other techniques to learn open or more parameters and/or hyper-parameters.

In some embodiments, the machine learning model 1214 may use supervised learning to determine the output prediction based on the labeled training data 1212. These output predictions may be accepted or corrected based on the correct results associated with the labels on the training data 1212. As an example, linear regression may be used to predict idle time for an operator of an abrasive product given input RPM data for the abrasive product. This may include, for example, optimizing a square loss function based on a difference between the operator's predicted idle time and the operator's actual tag idle time associated with the input RPM data.

In some embodiments, the machine learning model 1214 may use unsupervised learning to learn patterns, structures based on features of the unlabeled training data 1212. As an example, a clustering algorithm (e.g., k-means clustering, hierarchical clustering) may be used to group sensor data with similar characteristics into clusters, which may be used for anomaly detection (e.g., a sensor with characteristics of any cluster not belonging to may receive an alert). In another example, an auto-encoder may be used to learn a new representation of sensor data (typically with reduced dimensionality). These new representations of sensor data may then be used for classification tasks.

In some embodiments, the machine learning model 1214 may use semi-supervised learning by having labels for some, but not all, of the training data 1212. Thus, supervised learning may be used for a portion of the training data 1212 having a label, while unsupervised learning may be used for a portion of the training data 1212 not having a label. In some embodiments, the machine learning model 1214 may use reinforcement learning to receive reward values in response to actions in the environment. For example, during reinforcement learning, the machine learning model 1214 may take actions, such as providing a textual notification to the client device indicating the end of life, and receiving a reward value from the enterprise 1202. In response to the reward, the machine learning model 1214 may attempt to maximize the reward value by taking additional action (providing another textual notification) or exploring new potential actions (providing different types of notifications to the client device).

In some embodiments, the machine learning model 1214 may be trained based on various user experiences and/or knowledge. For example, operational limits may be set or suggested based on experience from a developer and/or application engineer. In such scenarios, the machine learning model 1214 may utilize organizational "know-how" and/or feedback regarding some or all aspects of the abrasive operation, such as, but not limited to, operator experience, manager experience, and/or customer feedback.

In some embodiments, the training of the machine learning model 1214 may be accelerated using a special-purpose processor, such as a Graphics Processing Unit (GPU), Tensor Processing Unit (TPU), and/or Digital Signal Processor (DSP). In some embodiments, training the machine learning model 1214 may include training different machine learning models for different sets of sensor data. The set of sensor data may be based on a unique identifier of the abrasive product from which the sensor data was collected. For example, a set of sensor data may include all sensor data collected from abrasive products having a unique identifier beginning with "X _12," which may represent abrasive products sharing similar characteristics. As another example, a set of sensor data may include all sensor data collected from abrasive products sharing the same number and type of sensors (e.g., vibration sensors).

In some embodiments, machine learning model 1214 may be configured to be periodically retrained with new sensor data received from enterprise 1202. For example, sensor data from enterprise 1202 may be associated with a timestamp indicating the time at which the sensor collected the data. Thus, the machine learning model 1214 may be configured to retrain on a subset of new sensor data determined by the time stamp (such as all data collected last week, last month, or last year). In some examples, periodically retraining the machine learning model 1214 may include removing old sensor data from the training data 1212 based on timestamps of the old sensor data. For example, sensor data collected over two, three, or four years may be removed from the training data 1212.

After training the machine learning model 1214 to become a trained machine learning model 1216 on the training data 1212, the trained machine learning model 1216 may be located and executed on the machine learning system 1210 to provide a prediction of requests from one or more computing devices. In some cases, the trained machine learning model 1216 may be located and executed on one or more computing devices to make predictions of requests from the one or more computing devices.

In some embodiments, trained machine learning model 1216 may receive input requests 1220 from enterprise 1202, generate one or more output predictions 1230 for input requests 1220, and provide the one or more output predictions to enterprise 1202. For example, the input request 1220 can be a request from the enterprise 1202 for predicting the end of service life of an abrasive device. Thus, input request 1220 may include RPM data for the abrasive device. In such examples, enterprise 1202 may be a provider of training data 1212. In other embodiments, the trained machine learning model 1216 may receive an input request 1220 from an external source 1240, generate one or more output predictions 1230 for the input request 1220, and transmit the one or more output predictions to the external source 1240. For example, the input request 1220 can be a request from an external source 1240 to predict the end of the useful life of the abrasive device. Thus, input request 1220 may include RPM data for the abrasive device. In such instances, external source 1240 may not be a provider of training data 1212, and thus may rely on training data from enterprise 1202 to make predictions.

In some embodiments, the output predictions 1230 may be labeled and utilized during a subsequent training phase to further refine the trained machine learning model 1216. For example, the output predictions 1230 may be provided as one or more product-specific solutions for solving the problem with the abrasive product. These product-specific solutions may be provided to enterprise 1202, which in turn selects a product-specific solution from the one or more product-specific solutions. Upon selecting a solution, enterprise 1202 may record the selected product-specific solution and may determine a label for the abrasive product as the selected solution. This solution may then be transmitted to training data 1212 for use in improving the trained machine learning model 1216.

Exemplary input

Fig. 13 illustrates a scenario 1300 representing an input device to a machine learning system, according to an example embodiment. Scenario 1300 may occur in enterprise 1120 and include a manual abrasive device 1310, a wearable device 1320, an automated abrasive device 1330, and a server device 1340. Thus, manual abrasive apparatus 1310 and automated abrasive apparatus 1330 may be embodied by abrasive product 1122, and server apparatus 1340 may be embodied by server apparatus 1126. As used herein, an abrasive apparatus may refer to an apparatus associated with or embodied by an abrasive tool.

Manual abrasive device 1310 may be any tool configured to perform a manual abrading operation on a workpiece (not shown in fig. 13). Such grinding operations may include grinding, polishing, buffing, honing, cutting, drilling, sharpening, filing, lapping, sanding, and/or other similar tasks. However, other types of manual mechanical operations are contemplated that may include vibration and/or noise. For example, hammering, chiseling, crimping, hammering, or other manual operations are possible within the context of the present disclosure. Accordingly, manual abrasive apparatus 1310 may be an apparatus configured to perform one or more abrasive operations. For example, manual abrasive device 1310 may be a right angle grinding tool, a power drill, a hammer drill and/or a rapping hammer, a saw, a planer, a screwdriver, a router, a sander, an angle grinder, a garden appliance, and/or a multi-function tool, among other examples.

In an example, automated abrasive device 1330 may be any tool configured to perform an automated abrading operation on a workpiece (not shown in fig. 13). Such grinding operations may include grinding, polishing, buffing, honing, cutting, drilling, sharpening, filing, lapping, sanding, and/or other similar tasks. For example, automated abrasive device 1330 can include a cutting tool, a sanding tool, a grinding tool, and/or a polishing tool.

Although an operator typically operates manual abrasive apparatus 1310, in the present disclosure, a controller operates automated abrasive apparatus 1330. The controller may take the form of computing device 1000 and may be configured to perform various actions, such as turning automated abrasive device 1330 on, turning automated abrasive device 1330 off, setting a rotational speed of automated abrasive device 1330, applying a desired abrading or cutting force at a desired position and/or angle relative to a workpiece, and other possibilities. In some embodiments, the actions of the controller are configured based on output from the machine learning system 1210. For example, machine learning system 1210 may make predictions based on abrasive operation data (e.g., vibration data, current data, wheel speed data, etc.) received from a controller of automated abrasive apparatus 1330. After making the prediction, the machine learning system 1210 may reconfigure the controller according to the prediction, for example, by setting a new maximum rotational speed at the controller. In some examples, automated abrasive apparatus 1330 can take the form of a Computer Numerically Controlled (CNC) grinder, and the aforementioned controller can be a CNC controller. However, other types of automated abrasive devices are possible and contemplated.

Manual abrasive apparatus 1310 and automated abrasive apparatus 1330 may include one or more components that enable the tool to perform one or more abrasive operations. In particular, manual abrasive apparatus 1310 and automated abrasive apparatus 1330 may include abrasive articles for performing the one or more operations. The abrasive article may include one or more materials useful for shaping or finishing a workpiece. The one or more materials may include abrasive minerals such as calcite (calcium carbonate), corundum (impure corundum), diamond, CBN, diamond powder (e.g., synthetic diamond), homogeneous quartzite, pumice, rouge, sand, corundum, garnet, sandstone, diatomaceous rock, powdered feldspar, staygorskite, boron nitride, ceramics, ceramic alumina, ceramic iron oxide, corundum, glass powder, steel abrasives, silicon carbide (corundum), zirconia alumina, boron carbide, and slag. Additionally and/or alternatively, the one or more materials may include a composite material including coarse particle aggregates that are pressed and bonded together with a binder. The composite material may include clay, resin, glass, rubber, alumina, silicon carbide, tungsten carbide, garnet, and/or garnet ceramic.

Further, the abrasive article may have one of a variety of shapes. For example, the article may take the form of a sheet, block, stick, wheel, ring or disc, among other examples. In the example shown in fig. 13, manual abrasive apparatus 1310 may include a wheel-shaped abrasive article 1316. In addition, manual abrasive apparatus 1310 and automated abrasive apparatus 1330 may include a power source that may be configured to actuate the abrasive article to perform an operation. In an example, the power source may be an electric motor, a gasoline engine, or compressed air. Manual abrasive apparatus 1310 and automated abrasive apparatus 1330 may also include a housing that houses a power source. The housing may be formed of hard plastic, phenolic resin, or medium durometer rubber, among other examples.

In some embodiments, manual abrasive device 1310 and automated abrasive device 1330 may include identification features, such as a scannable identifier (e.g., a QR code, a barcode, a serial number, etc.) that may be engraved or attached. The identification feature may be used to identify the type of tool, the manufacturer of the tool, the model of the tool, and/or a unique identifier of the tool. Additionally and/or alternatively, various components of manual abrasive apparatus 1310 and automated abrasive apparatus 1330 may include identification features. For example, the abrasive article may include an identification feature engraved in and/or attached to the abrasive article. The identification feature may be used to identify the type of abrasive article, the manufacturer of the abrasive article, the model of the abrasive article, and/or a unique identifier of the abrasive article.

Manual abrasive apparatus 1310 and automated abrasive apparatus 1330 may include one or more sensors that may collect data in real-time from the environment of the tool and/or from the tool itself (e.g., within the handle, body, and/or coupled to the abrasive product of the tool). In some embodiments, the system may additionally include a remote sensor disposed in the environment in which the operation is being performed. These sensors and embedded sensors may be configured to transmit the collected data to server device 1340. The sensor may be configured to transmit operational data of the tool and a unique identifier of the tool.

Wearable device 1320 is in the form of a wrist-worn device that is mounted on the wrist of the user's hand 1322. The user's hand 1322 may be the dominant hand that the operator prefers when performing tasks. Here, an operator may use a hand 1322 (on which wearable apparatus 1320 is mounted) to grasp handle 1312 of manual abrasive apparatus 1310. Wearable device 1320 may be configured to obtain real-time data that may be used to determine abrasive operation data. To obtain real-time data, wearable device 1320 may include sensors that can collect data in real-time from the environment of the abrasive product and/or from the abrasive product itself. Wearable device 1320 may be configured to communicate with a remote sensor on the abrasive product and/or with the one or more sensors associated with the abrasive product. Additionally, the wearable device may include a communication interface to transmit the collected data to the server device 1340, for example, in a real-time and/or asynchronous manner. In some embodiments, wearable device 1320 may be used to run web applications, which may include event driven scripts that operate in a node.

In some embodiments, a sensor of wearable device 1320 may be configured to read or scan an identification feature of manual abrasive device 1310 or an abrasive article of manual abrasive device 1310. For example, the sensor may include an image capture device (e.g., a camera) that may capture and analyze an image of the manual abrasive device 1310 in order to determine the type or scannable identification feature 1314. Identifying the tool and/or components thereof may allow the wearable device 1320 to provide information associated with the tool and/or components thereof to an operator. Additionally and/or alternatively, the identification may allow wearable device 1320 to associate data collected by sensors in the environment with manual abrasive device 1310 and/or particular components for performing a desired operation.

In scenario 1300, sensors of manual abrasive device 1310, wearable device 1320, automated abrasive device 1330, and other sensors 1350 may continuously or periodically collect data from or associated with the operation of manual abrasive device 1310 and automated abrasive device 1330. For example, the collected abrasive operation data may include sound data, acceleration data, vibration data, gyroscope data, and/or data extrapolated from sound, acceleration, and/or vibration data (e.g., applied force data, RPM data, usage, etc.). In example embodiments, the collected abrasive operation data may be related to material, material removal rate, operating conditions, consumed power, or specific grinding energy.

In addition, sensors of manual abrasive device 1310, wearable device 1320, automated abrasive device 1330, and other sensors 1350 may be positioned at various locations relative to the abrasive device/tool and/or the workpiece being operated upon. For example, a vibration sensor, a gyroscope, a microphone, and/or any other sensor may be embedded within the tool or the handle of the tool. In other examples, the sensor may be located near the tool and/or the workpiece. In yet another example, the sensor may be mounted on a work surface on which the workpiece may be placed. In even further examples, the sensor may be mounted at a wall or ceiling location. It should be understood that multiple sensors may be located at various locations near the tool and/or workpiece to provide a "stereo" or multi-sensor combination. Such multiple sensor combinations may provide information about which tool is being used and/or disambiguate a particular sound based on stereo or multi-view sensing.

In some embodiments, the sensor may include an accelerometer that may be used to measure and record acceleration information in three axes (x, y, and z). For example, wearable device 1320 may include an accelerometer configured to collect acceleration data related to a user's hand 1322 while operating manual abrasive device 1310. Thus, the accelerometer may measure acceleration of the hand due to vibration. Because the hand vibration is a result of the tool vibration, the acceleration information collected by the accelerometer may be indicative of the tool vibration. In such a scenario, the acceleration information may be used to determine the degree of vibration of the tool. Vibration data is one example of abrasive operation data that may be used to extrapolate other abrasive operation data. As an example, the vibration data may be used to determine operational information of the tool, such as operational state and operational time. For example, the operating states may include "OFF", "IDLE", and "SANDING", among other possibilities. As another example, the vibration data may be used to determine lapping information for the performed abrasive operation, such as working angle, gripping force, applied pressure, angular velocity (e.g., revolutions per minute, RPM), and other variables.

In some embodiments, the sensor may include a microphone configured to collect acoustic data from the abrasive operation. By analyzing the correlation of the magnitude of the acoustic data to known acoustic magnitudes for different RPM values of the abrasive product, the collected acoustic data can be used to determine the RPM at which the abrasive product is operating.

In some embodiments, the sensor may comprise an optical or magnetic sensor operably configured to detect a visible/magnetic attachment to the grinding wheel of the disc and provide information about the angular velocity (RPM) of the grinding wheel or disc.

In some embodiments, the sensor may include a spark-constant sensor configured to detect rotation of a sensing target attached to the grinding wheel or disc and provide information about the angular velocity (RPM) of the grinding wheel or disc. In some embodiments, the sensor may include a pressure sensor to determine the air pressure applied to the abrasive tool.

In an example embodiment, the determined RPM value may be used to extrapolate abrasive operation data. For example, the RPM value may determine the grinding power of a particular tool by using data (e.g., a table) indicating a correlation between the RPM of the tool and the grinding power applied by the tool. As another example, the RPM value may determine the force applied to the workpiece by using data (e.g., a table) indicating a correlation between the RPM of a particular tool and the grinding power applied by that tool.

In some embodiments, a sensor may collect information indicative of a workpiece. For example, the sensor may include an image capture device (e.g., a camera) configured to capture an image of the workpiece. The image may be analyzed to determine a state of the workpiece, including a type of the workpiece, a material of the workpiece, a size of the workpiece, a surface characteristic of the workpiece, and/or a placement of the workpiece in the environment (e.g., an orientation, an angle, a position relative to a reference point in the environment). In some embodiments, information about the workpiece may be determined based on other types of sensor information. For example, during the abrading process, the interaction between the abrading tool and various workpiece materials may provide different noise and vibration data. In an example embodiment, the database of labeled noise and vibration data may specifically identify various workpiece materials. In such scenarios, subsequently captured noise and vibration data may be compared to entries in a labeled noise and vibration database in order to provide information about a given workpiece material.

In some embodiments, the sensor may collect information indicative of microscopic interactions (e.g., tribology) occurring in the abrasive region. (i.e., the area between the grinding wheel and the work material). For example, the sensor may include a vibration sensor configured to extrapolate the depth of penetration of the workpiece by the abrasive particles during each step of the abrading process. Each penetration depth may then be associated with a particular microscopic interaction, such as cutting (e.g., material removal), plowing (e.g., material displacement), sliding (e.g., surface modification). Notably, the surface created at the end of the grinding process may be a cumulative effect of different types of microscopic interactions.

Server device 1340 may include one or more computing devices configured to receive and aggregate sensor data from manual abrasive device 1310, wearable device 1320, sensors of automated abrasive device 1330, and other sensors 1350. Communication between these sensors and server device 1340 may be facilitated via: a wireless fidelity (Wi-Fi) connection, a Bluetooth connection, an optical fidelity (Li-Fi) connection, an infrared connection, a Near Field Communication (NFC) connection, or some other wireless connection. In some embodiments, the sensors may be configured to communicate with server device 1340 using Message Queue Telemetry Transport (MQTT) or another type of messaging protocol.

Upon receiving the sensor data, server device 1340 may apply data filters to the sensor data or convert the sensor data into a different data format. Server device 1340 may transmit sensor data to a machine learning platform (e.g., machine learning platform 1110) in order to determine product-specific information and/or workpiece-specific information related to manual abrasive device 1310, wearable device 1320, automated abrasive device 1330, and/or other sensors 1350.

In some embodiments, server device 1340 may be configured to allow a human operator to label sensor data. In particular, the labels may be used as labels or additional training features for training one or more machine learning models. For example, the indicia may identify product-specific information for the one or more abrasive products associated with the sensor data. This may include operating conditions related to vibration data. As an example, an operational state of "walking" may be assigned to vibration data having a small peak. In another example, an "idle" operating state may be assigned to vibration data having a steady slope.

In some embodiments, the indicia may identify the pattern of wear operation data immediately prior to the abrasive product related event. For example, an abrasive product related event may include severe injury to an operator using the abrasive product. Thus, the indicia may be associated with vibration sensor data from a wearable device of the operator prior to the injury. For example, a "critical" marker may be associated with vibration data one hour prior to injury, while a "dangerous" marker may be associated with vibration data minutes or less prior to injury.

In some embodiments, server device 1340 may provide a graph of sensor data to assist a human operator in extrapolating patterns indicative of particular conditions or performance metrics of the abrasive product. As explained herein, a human operator may use one or more data analysis methods to extrapolate the pattern. As an example, the sensor profile may be used to extrapolate a correlation between a power signal supplied to the tool during operation and vibration of the tool during operation in order to assign a signature to the vibration data. In particular, vibration data having an amplitude greater than a threshold value during a certain time period/phase may indicate that an "energized" mark is assigned to the abrasive product during that time period/phase. Further, vibration data having an amplitude greater than the second threshold for a period/phase of time may indicate that a "harsh condition" mark was assigned to the abrasive product during that period/phase of time. In another example, when the abrasive product is operated under harsh conditions, the acceleration data may include a higher peak value than when the abrasive product is operated under normal conditions. Thus, a peak greater than a threshold in the vibration data may indicate a "harsh condition" marker.

In some embodiments, a human operator may use methods such as the following to assist in the tagging of sensor data: machine learning (e.g., bayesian classifiers, support vector machines, linear classifiers, k-nearest neighbor classifiers, decision trees, random forests), Fast Fourier Transforms (FFTs), Artificial Intelligence (AI) methods (e.g., neural networks, fuzzy logic, cluster analysis, or pattern recognition), filtering, peaking, mean, standard deviation, skewness, and/or kurtosis. For example, a human operator may use a machine learning platform (e.g., machine learning platform 1110) to assist in creating indicia of sensor data. Notably, the created labels can then be used to train one more machine learning models on the same machine learning platform.

Other sensors 1350 may include other sensors disposed in the environment around manual abrasive device 1310, wearable device 1320, and automated abrasive device 1330. For example, other sensors may include temperature sensors to detect the temperature of the environment surrounding manual abrasive apparatus 1310 and automated abrasive apparatus 1330. In some examples, sensors in manual abrasive device 1310, wearable device 1320, and automated abrasive device 1330, as well as other sensors 1350, may be in direct communication with a machine learning platform (e.g., machine learning platform 1110).

IX. exemplary output

Fig. 14 depicts a scenario 1400 representing a device receiving predictions from machine learning system 1210, according to an example embodiment. Scenario 1400 may include abrasive product 1410, remote device 1420, vendor 1430, and analysis platform 1440. Notably, abrasive product 1410 may be embodied by abrasive product 1122, analysis platform 1440 may be embodied by analysis platform 1118, supplier 1430 may be embodied by external supplier 1130, and remote device 1420 may be embodied by remote device 1128.

Abrasive product 1410 may include, for example, an automated abrasive device 1412, a manual abrasive device 1414, and a wearable device 1416 communicatively coupled to manual abrasive device 1414, as described herein. Accordingly, machine learning system 1210 may output predictive information related to abrasive operation data of abrasive product 1410. Such abrasive operational data may be collected by sensors disposed on or near abrasive product 1410. For example, abrasive operation data may include information about the angular velocity (RPM) of the grinding wheel, the severity of the operation, and the vibrations experienced by the tool. In another example, the abrasive operating data may include grinding parameters of manual abrasive device 1414, including working angle, grip tightness, and applied pressure. As another example, a noise sensor and a spark constancy sensor may provide electrical power information for automated abrasive device 1412 and/or manual abrasive device 1414. In a further example, the abrasive operation data can include gyroscope data indicative of a workpiece on which the operation is being performed (e.g., based on sensor data indicative of the workpiece, such as an image).

Based on the abrasive operation data, machine learning system 1210 may determine product-specific information for abrasive product 1410 and/or workpiece-specific information for a workpiece operated on by abrasive product 1410. Machine learning system 1210 may then provide one or more notifications to abrasive product 1410 alerting of product-specific information or workpiece-specific information. In response to receiving the indication, abrasive product 1410 may output a visual, tactile, and/or audio alert or may take automated action. For example, abrasive product 1410 may include a Graphical User Interface (GUI) or an attachment indicator light for displaying notifications. In another example, abrasive product 1410 may include a device (e.g., an embedded speaker) that converts the electrical pulse into sound for hearing an audio alarm. As a further example, abrasive product 1410 may include an embedded computing device that may automatically change the abrasive operation of abrasive product 1410 in response to an indication from machine learning system 1210. In yet another example, the wearable device 1416 can include a vibration mechanism such that the alert can be transmitted to the wearable device 1416 as a vibration.

In some embodiments, machine learning system 1210 may determine whether the operating speed of a particular grinding wheel of abrasive product 1410 is within a safe or productive operating range for a particular workpiece, a particular grinding wheel, or a combination thereof, being operated on. For example, machine learning system 1210 may provide a warning signal in the form of a warning light to abrasive product 1410. In another example, machine learning system 1210 may provide automated instructions to abrasive product 1410 to adjust RPM, open, and/or close. For example, the automatic command may be a signal to a fluid control valve (e.g., an air valve) to adjust the speed of the mill; or an indication of the use of an abrasive dresser to dress the surface of the grinding wheel.

In some embodiments, machine learning system 1210 can determine that an abrasive article of abrasive product 1410 is damaged or malfunctioning. For example, machine learning system 1210 may analyze acceleration and/or noise data to determine that an abrasive article is damaged and/or malfunctioning. This may involve detecting one or more patterns in the acceleration and/or noise data, which may indicate that the abrasive article is damaged or malfunctioning. For example, a first pattern of spikes or peaks may indicate abrasive product damage, while a second pattern of spikes or peaks may indicate abrasive product failure. After making the determination, machine learning system 1210 may provide an indication to abrasive product 1410 that the abrasive article is damaged or malfunctioning. Such an indication may be a visual, tactile, and/or audio alert. In some embodiments, the machine learning system 1210 is configured to send an indication when acceleration and/or noise data exceeds a predetermined threshold (e.g., a maximum expected tool vibration level, a maximum expected tool noise level, etc.). In such embodiments, the indication may be an instruction to shut down one or more abrasive products 1410. In addition, the alert may provide the user with an option to order a replacement article or request maintenance of an article via the GUI.

In some embodiments, machine learning system 1210 may determine, based on the gyroscope data, that the user is positioning manual abrasive device 1414 at an angle other than the recommended angle. Based on this abrasive operation data, machine learning system 1210 may determine that a user of manual abrasive apparatus 1414 is performing an operation incorrectly. Machine learning system 1210 may then provide an indication to manual abrasive device 1414 and/or wearable device 1416 that the user is performing the operation incorrectly. In response to receiving the indication, manual abrasive device 1414 and/or wearable device 1416 may output a visual, tactile, and/or audio alert indicating to the user that the user is performing the operation incorrectly. Additionally and/or alternatively, machine learning system 1210 may continuously provide feedback to a user indicating proper performance of an operation by providing notifications related to abrasive operations.

In some embodiments, machine learning system 1210 may determine an ergonomic state of a user operating manual abrasive apparatus 1414. For example, the determination may be based on an analysis of data captured from manual abrasive device 1414. Machine learning system 1210 may determine that the user has performed an operation for a period of time that exceeds a recommended time and/or may determine that the user is operating manual abrasive device 1414 at a force, vibration, and/or RPM level that is above/below a desired level. For example, an operator and/or abrasive product manufacturer may provide upper and lower limits of force, vibration, and/or RPM to machine learning system 1210. The upper and lower limits may be determined via the machine learning system 1210, an analysis platform (e.g., analysis platform 1118), and/or may be based on occupational safety standards implemented today or in the future. For example, the upper and lower limits may be based on standards set by the Occupational Safety and Health Administration (OSHA), the National Institute for Occupational Safety and Health (NIOSH), the european office for work safety and health (EU-OSHA), or the international organization for standardization (ISO). Accordingly, machine learning system 1210 may determine whether the force, vibration, and/or RPM data collected from a user operating manual abrasive device 1414 is within an "optimal zone," e.g., falling between an upper limit and a lower limit. If the percentage of time the user is operating within the optimal area is low enough, machine learning system 1210 may provide information to manual abrasive device 1414 and/or wearable device 1416 to increase the percentage of time in the "optimal area" (in some embodiments, by outputting a visual, tactile, and/or audio alert that provides an operational improvement, a recommended angle of operation, and/or the like). In such a manner, the systems and methods described herein may monitor operating conditions related to worker safety and/or compliance with international, federal (e.g., OSHA), national and/or local regulations and guidelines. In some embodiments, a safety margin (e.g., 1% -10% of the maximum limit) may be included in order to ensure compliance and avoid inadvertent violations due to, for example, sensor calibration errors or other minor sensor errors.

In some embodiments, the optimal regions associated with different abrasive operations and/or tools may be operationally related. For example, the optimal region of vibration data received from a manual abrasive device may be different than the optimal region of vibration data received from a handheld abrasive device. As another example, the optimal region of vibration data received from a lighter handheld abrasive device may be different than the optimal region of vibration data received from a heavier handheld abrasive device.

Further, the optimal region may be a measurement value calculated based on a maximum metric of the abrasive apparatus. For example, if the maximum RPM of the abrasive device is MAX _ PRM, the optimal area of the abrasive device may be calculated as anywhere between 0.6 MAX _ RPM and 0.7 MAX _ RPM.

It should be noted that while the optimal region is discussed based on data collected from manual abrasive apparatus 1414, the optimal region may similarly be developed for data collected from automated abrasive apparatus 1412.

In some embodiments, machine learning system 1210 may determine an end-of-life estimate for abrasive product 1410. For example, the determination may be based on an analysis of historical RPM data or product life data of similar abrasive products that failed and/or were deactivated. Machine learning system 1210 may then provide an indication to abrasive product 1410 or a supervisor of an operator of abrasive product 1410 that abrasive product 1410 is nearing the end of service life. The indication may include an estimated amount of time that an operator may safely use abrasive product 1410. In some cases, machine learning system 1210 may also be configured to automatically order replacement equipment for abrasive product 1410.

The remote device 1420 may include, for example, a mobile computing device, a database device, a tablet computing device, and/or other computing devices that facilitate operations including aggregating received data, filtering received data, and/or displaying received data. The remote device 1420 may also include the ability to execute web applications. Machine learning system 1210 may output predictive information related to aggregate abrasive operation data for a plurality of abrasive devices in an enterprise to remote device 1420. Thus, this information may be displayed by the remote device 1420 for analysis by one or more supervisors. For example, the aggregate abrasive operation data may include a length of time it takes to perform the assigned task, idle time of an abrasive equipment operator, and/or work time of the abrasive equipment operator. For example, sound data and/or vibration data may be used to determine that the abrasive apparatus is in operation.

Machine learning system 1210 may be configured to collect aggregate abrasive operational data and maintain a classification of the collected data by: the abrasive operation performed, the length of the abrasive operation, the workpiece associated with the abrasive operation, the abrasive equipment used, the operator performing the abrasive operation, and the time at which the abrasive operation was performed, feedback on the abrasive operation (e.g., from a manager or customer), vibration, noise, productivity, product life, and the like.

Based on aggregated abrasive operation data collected from a plurality of abrasive devices in an enterprise, machine learning system 1210 may provide predictions to improved workplace/enterprise operations. For example, machine learning system 1210 may predict a workflow and/or best practices for performing a particular type of task by utilizing a database of knowledge base articles that include information related to the task, information related to best practices in performing the task, and information describing how certain abrasive devices are used to complete the task. In another example, machine learning system 1210 can provide metrics associated with one or more abrasive devices in an enterprise. Metrics may include usage, total operating time, down time, number of failures, number of repair requests, and the like. These metrics may be compared between abrasive equipment used in a given task and between operators.

In an exemplary embodiment, the remote device 1420 may be programmed to display predictions of best practices, workflows, and metrics received from the machine learning system 1210. These may include graphical visualizations (e.g., histograms, bar graphs, trends in plots over time), Key Performance Indicators (KPIs) of the enterprise, and so forth. In some embodiments, remote device 1420 may be programmed to display virtual rewards that occur in response to positive actions performed by the abrasive device operator, such as positive safety actions (e.g., modifying the angle of cut, performing recommended safety checks, performing recommended maintenance, etc.) or positive operational actions (e.g., having the highest/fastest throughput, producing the most parts per minute, etc.), among other possibilities. In such embodiments, a central server device (e.g., server device 1126 or machine learning platform 1110) may establish a virtual competition between remote devices 1420, wherein the virtual rewards associated with each remote device 1420 are compared to one another, potentially encouraging each operator to increase their virtual rewards by performing further positive actions. The remote device 1420 may also have access to an analysis platform 1440, as discussed below.

The provider 1430 may represent one or more computing systems managed by an operator of the machine learning system 1210, partners of the operator of the machine learning system 1210, and/or partners of an enterprise using the machine learning system 1210. For example, provider 1430 may include computing systems associated with: a manufacturer of components responsible for assembling one or more abrasive devices, a logistics center responsible for delivering components to/from an enterprise, a development (R & D) center responsible for improving abrasive device operation, an OSHA inspector responsible for abrasive device safety, an abrasive device maintenance technician responsible for maintaining a defective abrasive device, and the like. To allow flexibility among potential vendors, machine learning system 1210 may provide the ability to add and/or remove vendors from communication. Adding a vendor may involve configuring machine learning system 1210 to process data in a vendor-provided format, and transmitting the data in a vendor-known format. Adding a vendor may also involve configuring the machine learning system 1210 to resolve conflicting actions between vendors. For example, machine learning system 1210 may resolve conflicting actions by defaulting to actions of the primary provider.

Communication with the provider 1430 may enable the machine learning system 1210 to automatically adjust supply chain/manufacturing operations in response to the predicted output. As an example, machine learning system 1210 may collect temporal abrasive operation data during the life of an operator or abrasive equipment to provide an estimate of abrasive product life, abrasive product condition, and/or operator condition. Based on the predicted output from the temporal abrasive operation data, machine learning system 1210 may be configured to transmit the command directly to supplier 1430. Such orders may include automatically ordering new components of the abrasive product near the predicted end of service life, automatically requesting OSHA checks to comply with regulatory laws based on the predicted safety hazard, and/or canceling the order based on extension of the predicted end of service life.

In some examples, the machine learning system 1210 can collect data associated with microscopic interactions between the abrasive equipment and the workpiece during the abrading process. Each set of microscopic interactions may be marked by the type of surface generated at the end of the milling process. Thus, the machine learning system 1210 may utilize the generated surface of the type to predict how microscopic interactions may be manipulated, individually or in some combination, to create a desired surface based on machine tool capabilities, abrasive particles, binders, abrasive/binder interactions between the surface and the tool, or structure, machined material properties, abrasive equipment characteristics and specifications, and operational factors (such as dressing, cycle design, coolant application). In addition, the power of the abrasive device, the grinding efficiency of the device tool, and the cutting efficiency of the device can be collected and used to create the desired surface.

In example embodiments, machine learning system 1210 may utilize such relationships, provide predictions based on the relationships, and/or transmit predictions to an abrasive manufacturing vendor to help produce improved abrasive products or abrading processes that achieve a particular set of micro-interactions, construct new abrasive particles or abrasive tools having different micro-interactions with a workpiece, and/or create customized abrasive products and abrading processes to meet customer requirements.

In some embodiments, machine learning system 1210 may provide the enterprise with the option to configure the automatic alert type sent to provider 1430. For example, the machine learning system 1210 may be configured to order new parts based only on the type of abrasive article that requires the part, the cost of the part, the lead time of the part, and the like.

Analysis platform 1440 may include a web application configured to receive wear operation data and predictions made by machine learning system 1210 across multiple enterprises. By utilizing data across multiple enterprises, the analysis platform 1440 may provide services that simulate the operation of abrasive products, estimate the cost of purchasing new abrasive products, and/or perform other analytical operations. For example, an enterprise that does not have access to a particular abrasive product may utilize the analysis platform 1440 to estimate a cost or productivity associated with the particular abrasive product. Analysis platform 1440 may be accessible by one or more users through a web browser.

In some embodiments, the analytics platform 1440 may provide layered services. Each tier may provide a different portion of information to the user. For example, a higher layer may give a user access to all services on the analytics platform 1440, while a lower layer may give a user access to only some of the services on the analytics platform 1440. The user may be assigned a tier based on a fee or subscription or another pay-for-service mechanism.

In some embodiments, the analytics platform 1440 may be configured to anonymize wear operation data and predictions received from the machine learning system 1210. For example, by ensuring that data from at least k businesses is indistinguishable, wear operation data may be separated from the businesses using de-identification methods such as k anonymization. As another example, the analytics platform 1440 may be configured to obtain limits from each enterprise that indicate types of wear operation data that may be shared and types of wear operation data that may not be shared.

In various embodiments, the analysis platform 1440 comprises a simulated environment programmed with a digital version of a physical abrasive product used by an enterprise (e.g., a "digital twin"). The simulation environment may use these digital versions to estimate productivity, cost, and/or damage due to the addition/reconfiguration/removal of different digital abrasive products from the simulation environment. For example, by means of, for example, a ProModel ^TMTo model a simulated environment.

In some examples, the simulated environment may use a digital version of the physical abrasive product to generate the synthetic sensor data. The composite sensor data may reflect actual sensor data generated from the physical abrasive product. However, unlike actual sensor data, synthetic sensor data may have the advantage of being fast and easy to generate. The synthesized sensor data may be provided to the enterprise for cost estimation, throughput analysis, energy usage, and the like. In addition, the synthesized sensor data can be provided to a machine learning system (e.g., machine learning system 1210) and used to train one or more machine learning models.

Further, the simulation environment may provide the ability to simulate the configuration of new abrasive products. For example, the simulation environment may be provided via a computer-aided design (CAD) program (e.g., AutoCAD)^TM) The ability to model entirely new abrasive products or new variants of existing abrasive products. Machine learning system 1210 may then estimate abrasive operation data (e.g., RPM) and associated predictions (e.g., cost, idle time) that may result from operation of a new abrasive product. This may help the user decide whether to actually create and manufacture a new abrasive product.

X. exemplary Process

Fig. 15 is a flow chart of a method 1500 according to an example embodiment. The method 1500 may begin at block 1502, where a wearable device (e.g., wearable device 1320) may collect sensor data from an abrasive operation. For example, an abrasive operation may include an operator having a wearable device grinding on a workpiece (e.g., a piece of sheet metal) using a manual angle grinder. During the milling operation, the wearable device may determine an angular velocity value of the wearable device relative to X, Y and the Z-axis based on the embedded gyroscope sensor. These values can be used to estimate the user's location of the hand-operated angle grinder. Further, the wearable device may determine the frequency, amplitude, and wavelength of sound emitted from the rotating grinding wheel on the angle grinder based on communication with sound sensors embedded in the angle grinder. This data can be used to estimate the operating state of the angle grinder. Notably, the sound data and gyroscope data are merely examples of sensor data that may be collected at block 1502. In fact, any of the sensor data described with respect to fig. 13 and 14 may be utilized.

At block 1504, machine learning system 1210 may receive sensor data from one or more server devices configured to aggregate sensor data from a wearable device (e.g., server device 1340) or directly from the one or more sensors on the wearable device. Machine learning system 1210 may then convert the received sensor data into feature vectors for input into one or more machine learning models. For example, machine learning system 1210 may be configured to create a feature vector (with an associated tag) for sensor data collected at a given timestamp. In other words, machine learning system 1210 may create feature vector V _1 to describe the sensor data collected at time T _1 and may create feature vector V _2 to describe the data collected at time T _ 2. Machine learning system 1210 may also convert classification data (e.g., product type) collected from sensors into a one-hot code. At block 1506, the machine learning system 1210 may pre-process the feature vectors to achieve more accurate predictions. For example, the preprocessing may include normalization techniques (scaling to unit norm), discretization of continuous features (e.g., binning), and/or other known methods.

At block 1508, the machine learning system 1210 may assign the feature vectors to the trained machine learning model for execution. As described above, machine learning system 1210 may include a plurality of trained machine learning models, each trained machine learning model being trained to make predictions for a particular subset of abrasive products. For example, the machine learning system 1210 can use a unique identifier (e.g., the identification feature 1314, which can be transmitted by the wearable device along with the sensor data) to recognize that the input feature vector corresponds to a manual angle grinder. Thus, the machine learning system 1210 can select a machine learning model that has been trained with sensor data from a comparable manual angle grinder. Continuing the above example, the selected machine learning model may be a feed-forward Artificial Neural Network (ANN) model that predicts whether an operator of the identified abrasive product is performing an unsafe abrasive operation. However, any of the machine learning models described with respect to fig. 12 may be utilized.

At block 1510, the machine learning system 1210 may execute the selected machine learning model using the input feature vectors to predict one or more outputs. According to the present example, the ANN model may map the non-normalized output of the ANN to a probability distribution of predicted outputs using a softmax function. For example, the predicted output may include operator conditions such as "OK", "approximate CRITICAL", and "CRITICAL". At block 1512, the predicted probabilities may be ranked and the highest probability output or any probability output greater than a predetermined threshold may be provided to the wearable device. Further, based on the prediction output, the machine learning system 1210 may be configured to provide suggestions to the wearable device to address any harmful conditions predicted. Other prediction outputs are also possible.

At block 1514, the wearable device may receive a notification with one or more recommendations from the machine learning system 1210 for provision to the operator. For example, the wearable device may include a graphical user interface for visual notifications to be displayed. The indication may be a flashing warning indicating that the operator is performing a hazardous operation using the abrasive apparatus. In response to the indication, the operator may change the position of the angle grinder. Such a change in positioning may result in another prediction by the machine learning system 1210 to determine whether the changed location has resolved the hazard. It is noted that other types of notification and display mechanisms described with respect to FIG. 15 are also possible.

Fig. 16 is a flow chart of a method 1600 according to an example embodiment. Method 1600 may begin at block 1602, where an automated abrasive apparatus (e.g., automated abrasive apparatus 1330) may collect sensor data from an abrasive operation. For example, an abrasive operation may include an automated abrasive device operating on a workpiece (e.g., a piece of sheet metal). During a lapping operation, the automated abrasive device can determine an RPM of a lapping wheel of the automated abrasive device based on the embedded spark constancy sensor. For example, the spark invariant sensor may be configured to detect rotation of a sensing target attached to the grinding wheel. Notably, RPM data is merely an example of sensor data that may be collected at block 1602. In fact, any of the sensor data described with respect to fig. 13 and 14 may be utilized.

At block 1604, machine learning system 1210 can receive sensor data from one or more server devices configured to aggregate sensor data from automated abrasive equipment (e.g., server device 1340) or directly from the one or more sensors of automated abrasive equipment. Machine learning system 1210 may then convert the received sensor data into feature vectors for input into one or more machine learning models. For example, machine learning system 1210 may be configured to create a feature vector (with an associated tag) for sensor data collected at a given timestamp. In other words, machine learning system 1210 may create feature vector V _1 to describe the sensor data collected at time T _1 and may create feature vector V _2 to describe the data collected at time T _ 2. Machine learning system 1210 may also convert classification data (e.g., product type) collected from sensors into a one-hot code. At block 1606, the machine learning system 1210 may pre-process the feature vectors to achieve more accurate predictions. For example, the preprocessing may include normalization techniques (scaling to unit norm), discretization of continuous features (e.g., binning), and/or other known methods.

At block 1608, the machine learning system 1210 may assign the feature vectors to the trained machine learning model for execution. As described above, machine learning system 1210 may include a plurality of trained machine learning models, each trained machine learning model being trained to make predictions for a particular subset of abrasive products. For example, machine learning system 1210 can use the unique identifier to recognize that the input feature vector corresponds to an automated abrasive device. Thus, machine learning system 1210 can select a machine learning model that has been trained with sensor data from a comparable automated abrasive apparatus. Continuing with the above example, the selected machine learning model may be a Support Vector Machine (SVM) that predicts whether an automated abrasive apparatus that is operating is malfunctioning or defective. However, any of the machine learning models described with respect to fig. 12 may be utilized.

At block 1610, the machine learning system 1210 may execute the selected machine learning model using the input feature vectors to predict one or more outputs. According to the present example, the SVM model may be a multi-class SVM that utilizes a one-to-one training mechanism (e.g., each pair of classification/prediction outputs is assigned to one SVM of the multi-class SVM) to determine a score for each of the prediction outputs. For example, the prediction output may include equipment conditions such as "OK", "malringing", "OVERHEATING", and "DANGER". At block 1612, the scores may be ranked and the predicted output with the highest score or any predicted output with a score greater than a predetermined threshold may be provided to the automated abrasive device. Further, based on the prediction output, machine learning system 1210 may be configured to provide control instructions to the automated abrasive apparatus to address any detrimental conditions. Other prediction outputs are also possible.

At block 1614, the automated abrasive apparatus may receive a notification from the machine learning system 1210 having one or more control instructions. For example, an automated abrasive device can include an embedded computing device that can control the operation of the automated abrasive device. The embedded computing device may receive the notification and adjust the rotational speed, provide the notification, turn the tool on, or turn the tool off. Alternatively and/or additionally, machine learning system 1210, upon predicting failure of the automated abrasive apparatus, may be configured to make an automation request to order a new part to repair the automated abrasive apparatus. It is noted that other types of notification and display mechanisms described with respect to FIG. 16 are also possible.

Fig. 17 is a flow diagram of a method 1700 according to an example embodiment. Method 1700 may begin at block 1702 where machine learning system 1210 outputs one or more predictions corresponding to sensor data received from one or more enterprises. For example, the prediction may include an average estimated idle time of an operator performing an abrasive operation in the enterprise. Other predictions as described herein are possible.

At block 1704, the predictions from block 1702 may be aggregated together to compute a global average prediction. For example, an average estimated idle time of an operator performing an abrasive operation may be aggregated across multiple enterprises. At block 1706, the one or more global average predictions may be used as input into a simulation engine that simulates abrasive operation of the human enterprise over a predetermined period of time (e.g., 10 years, 20 years). Accordingly, the simulation engine may rely on the one or more global mean predictions as parameters to guide the simulation and generate synthetic abrasive operation data.

At block 1708, output statistics from the simulation determined from the synthetic abrasive operation data, such as number of victims, total work time, and/or total cost, are provided. These output statistics may help users of the analysis platform 1440 make decisions such as whether to order more abrasive devices, whether to hire more operators, or whether to create new supply chain policies. Other options are also possible.

Xi. exemplary grinding wheel

During the assembly process, the abrasive product manufacturer may couple an identification mark (or colloquially referred to as a "mark") to the grinding wheel. From these markings, information unique to each grinding wheel can be determined. For example, the indicia may provide details regarding the manufacturing date of the grinding wheel, the surface material, the GPS location within the environment, the size, and/or the expiration date. Suitably, the abrasive product manufacturer and/or the abrasive product manufacturing customer may utilize the indicia to make informed decisions.

In some aspects, the markers may be used to track the movement of the grinding wheel within the enterprise environment. For example, an enterprise environment may be equipped with several indicia readers (or colloquially referred to as "readers") within a given environment. These readers may communicate with the tags to determine which grinding wheels are entering a given area of the environment and which grinding wheels are exiting such area.

In some aspects, the marking may improve the efficiency of the placement of the abrasive apparatus. For example, upon attaching the grinding wheel to the abrasive apparatus, the abrasive apparatus may automatically communicate with the signature of the grinding wheel to determine the characteristics (e.g., diameter, structure, material) of the grinding wheel. Using this information, the abrasive apparatus may determine optimal operating parameters (e.g., RPM speed, applied pressure, feed rate, depth of cut, traverse rate, coolant factors, dressing factors, etc.) for properly operating the grinding wheel.

In some aspects, the marking may increase the safety of the abrasive device. For example, upon attaching the grinding wheel to the abrasive apparatus, the abrasive apparatus may automatically communicate with the signature of the grinding wheel to determine the level of deterioration of the grinding wheel (e.g., a reduction in the grinding wheel outer coating). If the degradation exceeds a threshold level, the abrasive apparatus may notify a supervisor or apparatus operator that the degraded grinding wheel should be replaced with a new grinding wheel.

In some aspects, the tagging may improve the logistics aspects of the enterprise. For example, the enterprise may use the marking to dynamically maintain part inventories and determine if and/or when new grinding wheels should be ordered.

In some aspects, the markup may be combined with a software application that is capable of displaying markup information in a user-friendly format. Further, in response to obtaining the marking information, the software application may provide the user with the ability to perform an action, such as ordering a new grinding wheel.

As an exemplary operation, an abrasive product manufacturer may wish to track the usage of a grinding wheel within a customer environment. To this end, the abrasive product manufacturer may couple the mark to the grinding wheel and may provide a software application to the customer to communicate with the mark. Then, when the customer uses the grinding wheel, the indicia may be updated via the software application to reflect the number of workpieces on which the grinding wheel is operating, how long the grinding wheel has been in use, etc. Such usage information may be provided by the customer to the abrasive product manufacturer. In such scenarios, if the grinding wheel is shipped back to the abrasive product manufacturer for repair and/or refurbishment, the abrasive product manufacturer may know the history of the grinding wheel and may then determine an optimal repair strategy to address the use history and specific use cases of the grinding wheel.

Other features, functions, and benefits of the tag may exist and will be appreciated and understood from the following discussion.

Fig. 18A and 18B illustrate a grinding wheel 1800 according to an example embodiment. The grinding wheel 1800 may have all or some of the characteristics of the grinding wheels or grinding wheels previously described herein. In an example, the grinding wheel 1800 includes a marker 1810 and a connection mechanism 1812. The grinding wheel 1800 may be used as an abrasive component of an abrasive apparatus or abrasive product (such as a manual abrasive apparatus 1310 or an automated abrasive apparatus 1330), and may be physically connected to the abrasive apparatus, perhaps via a connection mechanism 1812. In some embodiments, marker 1810 may include a Quick Response (QR) code, a barcode, a Radio Frequency Identification (RFID) marker (active and passive), a Near Field Communication (NFC) marker, a Bluetooth Low Energy (BLE) device, or other type of marker. In an example, the marker 1810 can include information about the grinding wheel 1800, and/or can include a unique identifier, such as a Universally Unique Identifier (UUID), that can be used as a pointer reference. The pointer reference may direct the computing device to information about the grinding wheel 1800, which is stored in a database device or elsewhere. While an abrasive wheel 1800 is presently described, it should be understood that other types of abrasive products (such as bonded abrasives, coated abrasives, nonwoven abrasives, thin wheels, cutoff wheels, reinforced abrasive articles, superabrasive materials, single layer abrasive articles, and multi-layer abrasive articles) are possible and contemplated herein. Any of these other types of abrasive products may also include an identifier and may be utilized as described below.

Fig. 19 illustrates components of marker 1810, according to an example embodiment. In particular, marker 1810 is shown to include one or more sensors 1810A, a microcontroller 1810B, RFID Integrated Circuit (IC)1810C, and an antenna 1810D. In some examples, marker 1810 may have more, fewer, and/or different types of components than shown in fig. 19.

Sensor 1810A may have all or some of the characteristics of the sensors previously described herein. In some embodiments, the sensor 1810A may be physically outside of the scope of the marker 1810, but may be communicatively coupled to components within the marker 1810. In other embodiments, sensor 1810A may be physically within range of marker 1810, as depicted in fig. 19.

In some embodiments, the sensor 1810A can include a magnetometer configured to sense an ambient magnetic field of a workpiece being handled by the grinding wheel 1800. The magnetic field may be converted to an analog or digital electronic signal and transmitted to microcontroller 1810B, which may be configured to convert the magnetic field data to equivalent orientation data.

In some embodiments, the sensors 1810A may comprise temperature and humidity sensors configured to provide information about the ambient temperature and humidity level around the grinding wheel 1800. This reading may be converted to an equivalent analog or digital electronic signal and transmitted to microcontroller 1810B.

In some embodiments, sensor 1810A may comprise an accelerometer configured to measure vibration, orientation, surface sound level, Revolutions Per Minute (RPM), and/or angular acceleration of grinding wheel 1800. These measurements may be converted to analog or digital electronic signals and transmitted to microcontroller 1810B.

In some embodiments, the sensor 1810A can include a capacitive input interface capable of measuring changes in material density or potential damage associated with the grinding wheel 1800 via capacitance fluctuations between capacitive plates and/or wires. These measurements may be converted into digital electronic signals and transmitted to microcontroller 1810B.

Microcontroller 1810B may have all or some of the characteristics of computing device 1000. In some embodiments, microcontroller 1810B may be physically outside of the range of marker 1810, but may be communicatively connected to components within marker 1810. In an example, microcontroller 1810B can be configured to receive a digital electronic signal from sensor 1810A and write the received signal as data into a memory of RFID IC 1810C. In some embodiments, microcontroller 1810B may be physically outside of the range of marker 1810, but may be communicatively coupled to components within marker 1810. In other embodiments, microcontroller 1810B may be physically within the range of marker 1810, as depicted in fig. 19.

The RFIDIC1810C may be an integrated circuit that stores and processes information as well as modulates/demodulates signals. The RFID IC1810C may have all or some of the characteristics of the computing device 1000. In an example, the RFIC IC1810C can include information about a unique tag identifier, a unique tag serial number, a password, or information that may be related to the product (such as stock number, lot number, date of manufacture) or other specific information related to the grinding wheel 1800 additionally, the RFID IC1810C can be capable of storing information provided by the microcontroller 1810B, such as data collected from the sensor 1810A. The power to operate the RFID IC1810C may come from a battery pack attached to the tag 1810 or may be obtained from operation of the antenna 1810D. In some embodiments, the RFID IC1810C may be physically outside of the range of the marker 1810, but may be communicatively coupled to components within the marker 1810. In other embodiments, the RFID IC1810C may be physically within range of the marker 1810, as depicted in fig. 19.

The antenna 1810D can include an antenna structure and associated circuitry used by the tag 1810 that is configured to receive and transmit signals. In an example, the antenna 1810D can be communicatively coupled to the RFID IC 1810C. In some embodiments, antenna 1810D can comprise an inductive antenna coil that provides power to components of marker 1810. The antenna 1810D can be made from a variety of materials and can be printed, etched, stamped, or vapor deposited onto the marker 1810. In some embodiments, the antenna 1810D may be physically out of range of the marker 1810, but may be communicatively coupled to components within the marker 1810. In other embodiments, the antenna 1810D may be physically within range of the marker 1810, as depicted in fig. 19.

Fig. 20 illustrates a communication environment 2000 between a marker 1810 and a reader 2022, according to an example embodiment. Communication between the marker 1810 and the reader 2022 may occur via a communication medium 2010, which may include RFID, NFC, and/or BLE communication that communicates at ultra-high (e.g., at or near 900 megahertz), high (e.g., at or near 14 megahertz), or low (e.g., at or near 130 kilohertz) frequencies, wherein the physical distance during communication between the marker 1810 and the reader 2022 may vary based on the frequency and type of the communication medium 2010. The data received by the reader 2022 may be information related to the grinding wheel 1800 and/or the unique identifier of the grinding wheel 1800.

In some examples, the reader 2022 may take the form of a portable wireless reader system. In operation, the reader 2022 may receive information from the tag 1810 and immediately transmit the information via a wireless protocol (such as bluetooth or Wi-Fi) to the controller 2020 or to the portable mobile device.

In some examples, the reader 2022 may take the form of a portable wireless reader that is physically connected to the mobile device. In operation, the reader 2022 can receive information from the marker 1810 and immediately transmit the information to the mobile device via a USB connection, micro-USB connection, or similar physical connection mechanism.

In some examples, the reader 2022 may take the form of a stationary reader system equipped with an antenna that may communicate with the marker 1810. In operation, the reader 2022 may receive information from the tag 1810 and then transmit the information to the mobile device or controller 2020 via the wireless protocol discussed above.

Fig. 21 illustrates a communication environment 2100 between a client network 2120 and a remote network 2110 according to an example embodiment. Client network 2120 may have all or some of the characteristics of enterprise 1120, as shown and described with reference to FIG. 11. The remote network 2110 may have all or part of the features of the machine learning platform 1110, as shown and described with reference to fig. 11. In an example, client network 2120 may be a computer network used by entities to manage abrasive operations. Client network 2120 may include controller 2020, data storage 2122 (e.g., a database device, a file system), server device 2124 (e.g., a remotely hosted server device, a local server device, a virtual machine, etc.), abrasive device 2126 (e.g., a handheld abrasive device, an automated abrasive device, including manual abrasive device 1310 or automated abrasive device 1330), and user device 2128 (e.g., mobile device and/or wearable device 1320 shown and described with reference to fig. 13). Client network 2120 may be communicatively coupled to remote network 2110, which may be configured to manage aspects of client network 2120.

It should be noted that any components on the remote network 2110 and/or the client network 2120 may be replicated across multiple computing devices and/or hosted by a third party network (e.g., a cloud network) to provide data replication and increase the capacity of the service. Duplicate components may be located at various computing locations to ensure high availability in the event of a power failure at one computing location. In some cases, the remote network 2110 and/or the client network 2120 may be composed of a small number of devices and a small number of components. In other deployments, remote network 2110 and/or client network 2120 may span multiple physical locations and may include hundreds, thousands or more devices and other components. In some cases, certain components on client network 2120 may be managed by remote network 2110. For example, one or more server devices 2124 or controllers 2020 may be mounted on a client network 2120 via a remote network 2110 in order to support various embodiments herein.

Fig. 22 illustrates a method 2200 in accordance with an example embodiment. In particular, the method 2200 may represent a particular sequence or series of actions that, when executed, allow the remote network 2110 to collect tag information from the client network 2120 and provide updated state information to the user device 2128. By way of example, the method 2200 may utilize the remote network 2110, the client network 2120, and the user device 2128 during operation. However, additional components, steps, or blocks may also be added to the method 2200. For example, the steps described with respect to remote network 2110 may occur in whole or in part on a server device (e.g., server device 2124) located on client network 2120 or a third party network operated by client network 2120 (e.g., AMAZON WEB SERVICES) ^TM) The above. Such a scenario may be referred to as a "locally hosted solution" and may allow client network 2102 to perform the operations of method 2200 while at the same time by having all or part of the entities at client network 2Method 2200 is performed within 120 to maintain a high level of security and flexibility.

Method 2200 may begin at step 2202, when client network 2120 collects data from one or more markers (e.g., marker 1810). As described above, several ways may be utilized to collect marking data from the grinding wheel. For example, a portable wireless reader may be used to collect and transmit the indicia data to the controller 2020. As another example, a stationary reader system may be used to collect the tag data. In another example, a user of the mobile device may manually enter marking data after physically inspecting the characteristics of the grinding wheel. Other possibilities are also possible. Further, step 2202 can instantiate a continuous process occurring on client network 2120 in which one or more readers (e.g., reader 2022) continuously operate to collect data from one or more tags within client network 2120.

At step 2204, client network 2120 may transmit the marking data to remote network 2110. Step 2204 may be facilitated by several entities within the client network 2102. For example, the controller 2020 may transmit data to the remote network 2110, the user device 2128 may transmit data to the remote network 2110, and/or the reader 2022 may transmit the tag data to the remote network 2110. In an example, data transmission from the client network 2120 to the remote network 2110 may occur based on a predetermined frequency interval (e.g., every 1ms or 1s) or based on events occurring on the client network 2120 (e.g., any time new marker data is collected).

At step 2206, the remote network 2110 may update internal information to reflect the marker data received from step 2204. Thus, step 2206 may include updating database records, file entries, software parameters, and the like. Subsequently, at step 2208, the remote network 2101 may provide the updated user interface to the user device 2128. The user interface may be represented by means of a web page or series of web pages hosted by the remote network 2110 and provided to the user device 2128. For example, in some embodiments, the user interface presented to the user device 2128 may provide an overview of all of the grinding wheels currently in use within the client network 2120 as a series of selectable options. In some such scenarios, the selectable option may be configured to allow a user to view the marking data collected for each respective grinding wheel. In addition, the user interface may provide Key Performance Indicators (KPIs), such as overall inventory levels and actions that the user may perform.

At step 2210, upon the user selecting or otherwise selecting at least one actionable action, the at least one action may be transmitted by the user device 2128 to the remote network 2110. In response, at step 2212, the at least one action transmitted at step 2210 may be performed by remote network 2110.

In some cases, performing the at least one action may include one or more interactions with abrasive devices 2126 on client network 2120. For example, the remote network 2110 can establish communication with the abrasive devices 2126 on the client network 2120 to send requests to stop operation of certain abrasive devices, resume operation of abrasive devices, and the like.

In some cases, performing the at least one action may include interacting with a third party supplier (e.g., external supplier 1130) to order replacement parts or to repair/refurbish a malfunctioning or used grinding wheel. In some cases, for example, where a stock shortage is determined by the client network 2120, performing the at least one action may involve the remote network 2110 making a request to ship a new grinding wheel.

In some cases, performing the at least one action may involve providing an alert to other user devices 2128 on the client network 2120 in the form of a text message, an email alert, or another other type of communication. For example, an administrator from client network 2120 may transmit a message to all users of a certain product line to stop operation due to potential production problems.

Notably, the acts discussed above with respect to 2212 are not intended to be limiting. In practice, other actions are also possible.

FIG. 23 illustrates a state diagram 2320 according to an example embodiment. In particular, the state diagram 2320 conceptually illustrates how the grinding wheel 1800 may be moved between various operating states during its useful life. As used herein, an operating state may refer to a current position or function performed by or on the grinding wheel 1800. Some operating states may be accessed zero or more than once. Also, some operating states may have more than one possible next state, thus representing a decision to be made based on the information within marker 1810. It should be noted that upon entering a new state within the state diagram 2320, the marker 1810 may be "marked" to reflect this new state of the grinding wheel 1800.

The grinding wheel 1800 may begin at a manufacturer state 2322, which may represent the physical environment associated with the remote network 2110. Here, the grinding wheel 1800 may be configured and coupled with the marker 1810.

The grinding wheel 1800 may be moved from the manufacturer state 2322 to a gate state 2324, which may represent the grinding wheel 1800 reaching a physical environment associated with the client network 2120. At the door state 2324, one or more readers (e.g., reader 2222) may be used to transmit and obtain information from the marker 1810 regarding the unique identifier of the grinding wheel 1800 and/or the order number associated with the grinding wheel 1800. If the unique identifier and order number match the records of the client network 2120 (e.g., the client network 2120 has a record that they actually purchased the grinding wheel 1800 from the remote network 2310), the grinding wheel 1800 may proceed to the workplace state 2326. In some examples, the grinding wheel 1800 may be unmarked from the door status 2324 and returned to the manufacturer status 2322.

The workplace state 2326 may represent that the grinding wheel 1800 reached a grinding environment associated with the client network 2120. Once in the workplace state 2326, one or more readers (e.g., reader 2222) may be used to communicate and obtain information from the marker 1810 regarding the size and proper operating speed of the grinding wheel 1800. If the dimensions and appropriate operating speed match the records of the client network 2120 (e.g., the client network 2120 determines that the dimensions and appropriate operating speed of the grinding wheel 1800 match acceptable parameters of at least one abrasive apparatus operating within the grinding environment associated with the client network 2120), the grinding wheel 1800 may proceed to a wheel balance state 2328. Otherwise, the grinding wheel 1800 may be unmarked from the workplace state 2326 and returned to the manufacturer state 2322.

At the wheel balance state 2328, the grinding wheel 1800 may be inspected for imbalance problems that could potentially injure an abrasive equipment operator/damage the workpiece. If the imbalance problem exceeds a predetermined tolerance level, the grinding wheel 1800 may be unmarked from the wheel balance status 2328 and sent back to the workplace status 2326 for repair. Otherwise, the grinding wheel 1800 may continue to operate 2330.

In operation 2330, the grinding wheel 1800 may be attached to an abrasive apparatus (e.g., manual abrasive apparatus 1310 or automated abrasive apparatus 1330) to perform an abrasive operation. During the operation state 2330, the marker 1810 may be updated to reflect the number of workpieces on which the grinding wheel 1800 is operating, how long the grinding wheel 1800 is in use, and the like. After the grinding wheel 1800 begins to show deterioration, one or more readers may be used to communicate with the marker 1810 and obtain information from the marker regarding the expiration date and the manufacturing date of the grinding wheel 1800. In some embodiments, if the grinding wheel 1800 is near the end of its useful life, the grinding wheel 1800 may be unmarked from the operating state 2330 and may enter a discard state 2332, which may represent the end of useful life state of the grinding wheel 1800. In other examples, the grinding wheel 1800 may be unmarked from the operation 2330 state and entered the manufacturer 2322 state for servicing.

Xii. exemplary mobile device

Fig. 24A illustrates a view of a mobile device 2400 according to an example embodiment. In an example, mobile device 2400 may be configured with an authentication mechanism that may include a password, two-factor authentication, fingerprint recognition, facial recognition, or verification of other biometric information. Such authentication mechanisms may provide different levels or types of user access. For example, the user access levels may include access levels from users of the client network 2120, from administrators of the client network 2120, and/or from non-administrators of the client network 2120. Based on the current user's access level, mobile device 2400 may display different arrangements of information, provide access to different types of information, and/or provide various functions.

Information about mobile device 2400 may include maintenance information, information related to the state of client network 2120 (e.g., the number of grinding wheels available, the service life of the grinding wheels, etc.), and the like. As described above, mobile device 2400 may also include selectable options to perform actions. As an example, the action may include viewing product information 2402 and/or viewing current inventory information 2404. It should be understood that mobile device 2400 may include a smartphone, a tablet, a laptop, or another type of computing device. Still further, the mobile device 2400 may include, for example, a Head Mounted Display (HMD), a Heads Up Display (HUD), or another type of portable computing device with or without a user interface.

Such embodiments may provide a simpler way for a customer to reorder abrasive products. In such scenarios, upon receiving information that the end of life of a given abrasive product is imminent, mobile device 2400 may provide a notification of such information to a user. Additionally or alternatively, mobile device 2400 may be configured to display a product catalog and/or may functionally replace a product that is about to reach the end of its useful life for abrasive products. Thus, in such instances, a user may interact with the mobile device 2400 to request a given abrasive article, amount, type of shipment, desired delivery date, and so forth. Mobile device 240 may forward the request to client network 2120 and/or remote network 2110. Such systems and methods may reduce manual ordering efforts and/or paperwork, which may be more environmentally friendly. A reorder request made via mobile device 2400 may be authorized by a user (e.g., an administrator, manager, and/or sales person) with appropriate authority.

Fig. 24B illustrates another view of the mobile device 2400 according to an example embodiment. In particular, fig. 24B illustrates how mobile device 2400 can include multiple navigation menus to assist a user in performing various functions. For example, the inventory menu 2406 may provide an interface that displays the following information: information about all grinding wheels that have been purchased by a particular user. The operations menu 2408 may provide an interface that displays data regarding the operation of the grinding wheel, perhaps data collected by the sensor 1810A. Door entry 2410 may provide an interface that enables a user to interact with a marking coupled to a grinding wheel, perhaps by enabling a reader embedded on mobile device 2400 or communicatively connected to mobile device 2400. Workplace menu 2412 may provide an interface that displays information regarding which grinding wheels are currently in a workplace state or operating state, as previously discussed in connection with state diagram 2320. The discard menu 2414 may provide an interface that allows a user to unmark a grinding wheel, perhaps in conjunction with the discard status 2332 in the status diagram 2320. It is noted that other navigation menus are possible.

Fig. 24C illustrates yet another view of the mobile device 2400 according to an example embodiment. In particular, fig. 24C illustrates several non-limiting examples of information items that may be determined with respect to a particular grinding wheel. For example, FIG. 24C illustrates that information about: a product ID of the grinding wheel, an order ID, a product name, a product type, a size, a manufacturing date, a expiration date, a customer ID, a customer name, an operating speed, an inventory time, an inventory storage cost, a storage wait time, and a storage cost. It is noted that other items of information (e.g., GPS position of the grinding wheel) are possible.

XIII example network applications and data models

As described above, the web application may be configured to display information related to remote sensors, wearable devices, abrasive device operators, and the like. This may be accomplished by way of a web page or series of web pages hosted by the cloud computing device and provided to the user upon request. The layout and compilation of information in these web pages may enable efficient viewing of relevant information related to remote sensors, wearable devices, abrasive device operators, and the like. In addition, web pages may organize and arrange information using graphics with intuitive visual effects and easily understandable metrics.

As an additional feature, the web application may allow a user to establish an association between an abrasive device, a wearable device, an abrasive device operator, and a plant (e.g., an environment in which an abrasive operation is being performed). For example, a user may associate a plant P1 with an abrasive apparatus AT1 to indicate that an abrasive apparatus AT1 is operating within the plant P1. The user may then associate abrasive device AT1 with wearable device WD2 to indicate that the data collected by wearable device WD2 is related to the operation of abrasive device AT 1. Finally, the user may associate wearable device WD1 with operator O1 to indicate that operator O1 is wearing wearable device WD 1. In this way, the abrasive device, the wearable device, the abrasive device operator, and the plant become distinct logical entities on the web application that can be mixed together in match with each other.

Having different logical entities may bring many benefits. For example, assume that wearable device WD1 is permanently associated with operator O1. If the operator O1 suddenly becomes unavailable, no data can be collected from the wearable device WD1 during the unavailable period. On the other hand, assume that wearable device WD1 is a different logical entity than operator O1. If operator O1 becomes unavailable, wearable device WD1 may quickly associate with operator O3 and still collect data for wearable device WD 1. Advantageously, data may be collected from the wearable device WD1, whether by operator O1 or operator O3. Other advantages are also possible.

FIG. 25 illustrates a model 2500 according to an example embodiment. The model 2500 may include four basic tables-a plant table 2510, a device table 2530, a wearable table 2550, and an operator table 2550-and three linked tables-a plant device table 2520, a device wearable table 2540, and an operator wearable table 2560. These tables as a whole provide the necessary information to capture the relationship between the plant, abrasive device, wearable device and operator. In some examples, model 2500 may have more, fewer, and/or different types of tables than shown in fig. 25. In addition, the tables in the model 2500 may be truncated for clarity. In practice, however, these tables may contain more, fewer, and/or different entries.

The plant table 2510 may include entries for plants. In particular, each entry in the plant table 2510 may have a unique identifier for a plant and associated information for the plant. In some instances, a user may enter information to populate the shop table 2510, for example, through a web page or series of web pages provided by the cloud computing device.

The plant equipment table 2520 may include entries that map a given plant from the plant table 2510 to abrasive equipment from the equipment table 2530 operating in the given plant. In particular, the web application described above may provide a way to dynamically populate the entries in the plant equipment table 2520. For example, the web application may provide a series of drop down menus that allow a user to establish an association between workshops and abrasive equipment operating within those workshops.

The apparatus table 2530 may include entries for an abrasive apparatus (e.g., manual abrasive apparatus 1310). In particular, each entry in the device table 2530 may have a unique identifier for the abrasive device and associated information for the abrasive device. In some instances, a user may enter information to populate the device table 2530, for example, through a web page or series of web pages provided by the cloud computing device. In other examples, the information in device table 2530 may be populated from remote sensors and/or wearable devices as described above.

The device wearable table 2540 may include an entry that maps an abrasive device from the device table 2530 to a wearable from the wearable table 2550 that collects data associated with the abrasive device. In particular, the web application described above may provide a way to dynamically populate entries in the device wearable table 2540. For example, the web application may provide a series of drop down menus to allow a user to establish an association between the abrasive device and the wearable device. In some cases, as described above, entries in the device wearable table 2540 may be automatically populated by the reader. For example, the abrasive device can include an RFID tag, such as identification feature 1314, and the wearable device can include an RFID reader that can read the RFID tag of the abrasive device to associate the wearable device with the abrasive device.

The wearable table 2550 may include entries for wearable devices, such as the wearable device 1320. In particular, each entry in the wearable table 2550 may have a unique identifier for the wearable device and related information for the wearable device. In some instances, the user may enter information to populate the wearable watch 2550, for example, through a web page or series of web pages provided by the cloud computing device. In other examples, information in the wearable watch 2550 may be populated from remote sensors as described above.

The operator wearable table 2560 can include entries that map wearable devices from the wearable table 2550 to operators wearing wearable devices from the operator table 2570. In particular, the network application described above may provide a way to dynamically populate entries in the operator wearable table 2560. For example, the web application may provide a series of drop down menus to allow the user to establish an association between the wearable device and the operator. In some cases, as described above, entries in the operator wearable table 2560 may be automatically populated by the reader. For example, the wearable device may include an RFID tag, and the operator may have an RFID reader that may read the RFID tag of the wearable device to associate the wearable device with the operator.

The operator table 2570 may include entries for operators wearing wearable devices. In particular, each entry in operator table 2570 may have a unique identifier for the operator and associated information for the operator. In some instances, a user may enter information to populate operator table 2570, for example, through a web page or series of web pages provided by the cloud computing device.

In summary, the table of model 2500 provides information to establish: (i) which operators are associated with which wearable devices, (ii) which wearable devices are associated with which abrasive devices, and (iii) which abrasive devices are associated with which workshops. In some cases, the network application may use this information to provide metrics related to the plant, wearable device, abrasive device, and operator.

FIG. 26 illustrates a web page 2600 in accordance with an example embodiment. The web page 2600 may be provided to the user by a web application as described above. In particular, web page 2600 provides metrics related to the plant, wearable device, abrasive device, and operator.

As shown in fig. 26, a plant drop-down list 2610 allows a user to indicate a plant of a plurality of plants for which they want to receive metrics. Device drop down list 2620 allows the user to select one or more devices for which they want to receive metrics. The devices available in the device drop-down list 2620 may be based on the user's selections of the plant drop-down list 2610 and the entries in the plant device table 2520. Date range 2630 allows users to select the date range for which they want to receive metrics. After making selections for the shop drop-down list 2610, the equipment drop-down list 2620, and the date range 2630, the user may continue by pressing "search". The action may display a shop drop-down list 2610, a device drop-down list

One or more entries (e.g., entries) corresponding to information in date range 2630 and 2620

2640)。

Entry 2640 includes metrics related to the use of a device selected from device drop-down list 2620, the particular operator within a plant selected from plant drop-down list 2610, and during a time range selected from date range 2630. The particular operator may be determined based on entries in the operator wearable table 2560, the wearable table 2550, and the device wearable table 2540. Entry 2640 shows a particular operator's lapping time metric 2650, optimal lapping metric 2660, and vibration exposure metric 2670.

The grind time metric 2650 displays a bar graph of the total grind time for a particular operator during the date range 2630. In particular, embodiments relating to graphs 2900 and 3000 may be used to determine a grinding time metric 2650, as described further below.

The optimal lapping metric 2660 shows a bar graph of the time spent by a particular operator lapping within the optimal lapping parameters. In particular, the embodiments described with respect to graphs 2900 and 3000 may be used to determine the optimal lapping metric 2660. While the optimal lapping metric 2660 is illustrated as a bar graph, it should be appreciated that the amount of time when lapping occurs or the percentage or ratio of such time within the optimal lapping parameters may be represented and/or displayed in a variety of different forms. For example, the best lapping metric 2660 may be represented as a pie chart, a radar chart, a line chart, or another type of informational representation or map.

The vibration exposure metric 2670 displays a pie chart of vibration exposure times for a particular operator in three categories. In particular, the vibration exposure metric 2670 may be determined using the embodiments described with respect to graphs 2900 and 3000. While the vibration exposure metric 2670 is shown as a pie chart, it should be understood that the amount of time under respective vibration exposure conditions may be represented and/or displayed in a variety of different forms. For example, the vibration exposure metric 2670 may be represented as a pie chart, a radar chart, a line chart, or another type of information representation or information chart.

RPM graph 2680 is a plot indicating the RPM level of the devices specified in device drop down list 2620. The x-axis of the RPM graph 2680 corresponds to time values, while the y-axis of the RPM graph 2680 corresponds to RPM values.

It should be understood that web page 2600 is presented for purposes of example. In other embodiments, web page 2600 can provide other types of metrics and alternative methods of displaying such metrics.

Fig. 27 illustrates displays 2700, 2710, 2720, and 2730 of a wearable device 1320, according to an example embodiment. In particular, the display shown in fig. 27 illustrates different views that may appear on user interface components of wearable device 1320. It should be noted, however, that the display shown in FIG. 27 is not limiting; other displays are also contemplated and are possible within the scope of the present disclosure.

Display 2700 provides information about the average vibration of wearable device 1320, battery level (shown on the top left), current time (shown in the middle of the top), and whether there is a WiFi signal on wearable device 1320 (shown on the top right).

Display 2710 also depicts battery level, current time, and WiFi signal of wearable device 1320, but additionally shows a grind time metric that may be calculated, for example, using graphs 2900 and 3000 discussed in fig. 29 and 30.

Display 2720 also depicts the battery level, current time, and WiFi signal of wearable device 1320, but additionally shows an optimal grinding time metric that may be calculated, for example, using graphs 2900 and 3000 discussed in fig. 29 and 30.

Display 2730 also depicts the battery level, current time, and WiFi signal of wearable device 1320, but additionally shows a transient view of the current RPM and vibration as the operator performs the abrasive operation.

Xiv. exemplary analytical dashboard

Consistent with the discussion above, the analysis platform 1118 may be configured to display metrics associated with one or more abrasive products and/or one or more workpieces in the enterprise 1120. For example, the analysis platform 1118 may display technical orientation metrics (e.g., surface quality of the workpiece, abrasive equipment/tools near end of service life, etc.), economic orientation metrics (e.g., estimated cost per abrasive tool, estimated workplace efficiency, estimated throughput levels, etc.), and/or other types of metrics.

The analytics platform 1118 may display such metrics by means of a GUI containing one or more panes. As described herein, the term "pane" may refer to a GUI component that contains one or more locations (to display information) and/or one or more user selectable items, such as buttons or tabs. In some embodiments, the pane may be identical to or contained within a page or GUI window, although such a window may contain multiple panes. The buttons and/or tabs may be graphical control elements that display additional information within the pane.

Exemplary panes are shown in fig. 28A, 28B, and 28C below. These exemplary panes organize and arrange information using a graph with intuitive visual effects and an easily understandable diagram. As a result, a user of the analysis platform 1118 may quickly and efficiently view relevant information regarding the abrasive operations occurring at the enterprise 1120. It is noted that the following examples of panes are for illustrative purposes only and are not intended to be limiting. There may be other panes that include alternative arrangements of information.

Fig. 28A depicts a current time series pane 2810 according to an example embodiment. The current time series pane 2810 includes a machine drop down list 2812, a date range 2814, a current trace map 2820, a tool ID map 2822, and a part count map 2824. The current time series pane 2810 also includes a navigation bar 2810 that includes "current time series", "operational factors", "loop comparisons", and "performance metrics" tabs. It is noted that the "current time series" tab is shown in dashed lines to indicate that the information for that tab is currently displayed.

Machine drop down list 2812 allows a user to select one or more abrasive devices for which they want to receive metrics. The devices available in machine drop down list 2812 may include all devices operating (or already operating) in enterprise 1120. In the example shown in fig. 28A, an abrasive device "13333" has been selected.

Date range 2814 allows users to select a date range for which they want to receive metrics. In the example shown in fig. 28A, the dates range from "11/01/2019" to "11/02/2019".

Upon selection for the machine drop down list 2812 and date range 2814, the current time series pane 2810 may responsively display metrics related to the selection in the trace map 2820, tool ID map 2822, and part count map 2824.

Trace 2820 is a plot depicting the current experienced by abrasive apparatus "13333" over time. The x-axis of the trace 2820 corresponds to time values, while the y-axis corresponds to current values (in amps). As shown, the current of abrasive device "13333" varies over time. In some time periods, the current is higher. These may correspond to the time period for which the abrasive device "13333" performs an abrasive operation. In other time periods, the current is lower. These may correspond to periods of time when the abrasive apparatus "13333" is not performing abrasive operations. Further, the trace diagram 2820 shows a number of repeating patterns. For example, mode 2820A is similar to mode 2820B. These repeating patterns may correspond to periods of time during which the abrasive apparatus "13333" performs similar abrasive operations.

Consistent with the discussion above, in some embodiments, analysis platform 1118 may be configured to allow an operator to mark various traces in trace graph 2820. For example, the operational state of "lapping" may be assigned to the trace in trace 2820 having the larger peak. As another example, an "idle" operational state may be assigned to a trace in trace graph 2820 that has a steady slope. The labels may be used as labels or additional training features for training one or more machine learning models, as previously described herein.

Tool ID map 2822 is a plot depicting an abrasive tool (e.g., a grinding wheel) used by abrasive apparatus "13333" over time. The x-axis of the tool ID map 2822 corresponds to time values, while the y-axis corresponds to abrasive tool ID values. As shown, the abrasive tool used by abrasive apparatus "13333" changes over time. For example, the abrasive apparatus "13333" uses both the abrasive tool 2822A and the abrasive tool 2822B during the time period corresponding to the mode 2820A.

Part count diagram 2824 is a plot depicting a workpiece of abrasive apparatus "13333" operating over time. The x-axis of the part count map 2824 corresponds to time values, while the y-axis corresponds to workpiece ID values. As shown, the workpiece being manipulated by the abrasive apparatus "13333" changes over time. For example, during a time period corresponding to mode 2820A, the abrasive apparatus "13333" is operating on the workpiece 2824A, while during a time period corresponding to mode 2820B, the abrasive apparatus "13333" is operating on the workpiece 2824B.

In example embodiments, the trace map 2820, tool ID map 2822, and part count map 2824 may be used to understand the impact of modifying abrasive operations. For example, mode 2820A may correspond to a time period before modifying abrasive operation, while mode 2820B may correspond to a time period after modification. If mode 2820A and mode 2820B exhibit similar traces in trace 2820, it may be determined that the modification has no substantial effect on abrasive operation. On the other hand, if mode 2820A and mode 2820B exhibit different traces in trace diagram 2820, it may be determined that the modification does have a substantial effect. Exemplary modifications to the abrasive operation may include changing the operator, changing the speed of operation, and the like.

In some embodiments, an operator of the abrasive apparatus may utilize the trace map 2820, the tool ID map 2822, and the part count map 2824 to make real-time adjustments to the abrasive operation. For example, if it is determined that the current value of abrasive apparatus "13333" on trace 2820 increases each time abrasive apparatus "13333" is used with abrasive tool 2822A, this may indicate that abrasive tool 2822A is becoming dull (e.g., more energy is required to operate at the same speed). Thus, the operator may perform a dressing process on the abrasive tool 2822A to sharpen it and return it to an operating condition. As another example, if the current value of the abrasive apparatus "13333" on the trace 2820 exceeds a preset upper limit, the operator may turn off the abrasive apparatus "13333". Other examples are possible.

In some embodiments, the trace map 2820, tool ID map 2822, and part count map 2824 may be used to understand economic metrics related to abrasive operations. For example, if it is determined that the time it takes to abrade the workpiece 2824A with the abrasive tool 2822A is less than the time it takes to abrade the workpiece 2824A with the abrasive tool 2822B, it is economically advantageous to order more abrasive tools 2822A than abrasive tools 2822B, as using the abrasive tool 2822A will reduce the total abrading cycle time. Other examples of economic metrics are possible.

Fig. 28B depicts a loop comparison pane 2802 according to an example embodiment. As with the current time series pane 2800, the loop comparison pane 2802 includes a machine drop down list 2812 and a date range 2814. However, unlike the current time series pane 2800, the loop comparison pane 2802 includes a loop comparison graph 2830 and a metrics drop down list 2832. The loop comparison pane 2802 also contains a navigation bar 2810 that includes "current time series", "operational factors", "loop comparison", and "performance metrics" tabs. Notably, the "loop compare" tab is shown in dashed lines to indicate that information for that tab is currently displayed.

The cycle comparison graph 2830 includes a data graph showing the current usage of each abrasive device over a certain period of time. The x-axis of the cyclic comparison graph 2830 corresponds to time values, while the y-axis corresponds to current values (in amps). As shown in fig. 28B, cyclic comparison 2830 plots two abrasive devices simultaneously: current use cases of abrasive apparatus "43128" (orange plot) and abrasive apparatus "43131" (blue plot). The cyclic comparison graph 2830 can be beneficial because it allows a user to simultaneously view metrics associated with several abrasive devices on one data graph, enabling visual comparisons between the performance of the abrasive devices.

If the user decides to select an additional abrasive device from the machine drop-down list 2812, an additional plot of the additional abrasive device is added to the loop comparison graph 2830. In theory, the cyclic comparison graph 2830 may include data plots for tens or even hundreds of abrasive devices.

The metric drop down list 2832 may be a drop down menu that allows the user to select alternative metrics to display on the loop comparison graph 2830. Exemplary metrics may include current, vibration, feed rate or RPM, among other possibilities. In an example embodiment, the loop comparison graph 2830 may automatically update itself in response to a selection at the metrics drop down list 2832. For example, instead of displaying the current values in the y-axis of the cyclic comparison graph 2830, the values selected via the metric dropdown list 2832 may be displayed on the y-axis.

Fig. 28C depicts a performance metrics pane 2804, according to an example embodiment. As with the current time series pane 2800, the performance metrics pane 2804 includes a machine drop down list 2812 and a date range 2814. However, unlike the current time series pane 2800, the performance metrics pane 2804 includes a performance metrics map 2840. In addition, the performance metrics pane 2804 contains a navigation bar 2810 that includes "current time series", "operational factors", "loop comparisons", and "performance metrics" tabs. Notably, the "performance metrics" tab is shown in dashed lines to indicate the information that the tab is currently displayed.

The performance metric graph 2840 may include various plots that track integrated current over time for several abrasive tools. The x-axis of the performance metric map 2840 corresponds to time values, while the y-axis corresponds to current values (in amps). Similar to the cycle comparison graph 2830, the performance metric graph 2840 may be beneficial because the performance metric graph allows a user to simultaneously view metrics associated with several abrasive tools on a data graph, enabling visual comparisons between the performance of the abrasive tools.

XV. exemplary System and method for computing

As previously mentioned, the abrasive product/tool may include a sensor that detects the angular velocity (RPM) of the grinding wheel or disc. Wearable device 1320 may communicate with these sensors to receive RPM information and determine the grinding power and/or applied grinding force of the abrasive product/tool. Additionally and/or alternatively, the wearable device 1320 may use the sound data to determine the RPM of the grinding wheel or disc. In particular, wearable device 1320 may analyze the magnitude of the sound data and then map the sound magnitude to an estimated RPM value using a correlation table. The mapping between sound amplitude and estimated RPM value may vary depending on the type of abrasive product/tool.

In any of the above scenarios, the wearable device 1320 relies on communication with sensors or the type of abrasive product/tool (e.g., for mapping) to determine RPM information. However, it may be advantageous to decouple the dependencies of wearable device 1320 from the abrasive product/tool. Doing so may allow, for example, wearable device 1320 to determine the RPM of any grinding wheel or disc regardless of how the user of wearable device 1320 holds the abrasive product/tool, regardless of the type of abrasive product/tool held, and regardless of whether there are any communication sensors on the abrasive product/tool.

To independently determine the RPM, a vibration signal may be used. In particular, the vibration signal may be determined from an accelerometer of the wearable device 1320. As described above, the accelerometer collects acceleration data related to the vibration of the user's hand. Because the hand vibration is caused by the abrasive product/tool vibration, the acceleration data is indicative of the abrasive product/tool vibration. The acceleration data can then be used to calculate a gRMS value over time, thereby generating a vibration signal. Notably, the gRMS calculations may be performed on the wearable device 1320, on a remote device (such as the aforementioned cloud computing devices), or partially on the wearable device 1320 and partially on the remote device.

Fig. 29 illustrates a graph 2900 according to an example embodiment. As shown in fig. 29, graph 2900 includes a signal 2902 representative of the vibration of wearable device 1320 over time. That is, signal 2902 is generated by vibrations experienced by a user while wearing wearable device 1320 and using an abrasive product/tool. The x-axis of graph 2900 corresponds to time values, while the y-axis corresponds to vibration values (in gRMS).

It is important to recognize that because the RPM of the grinding wheel or disk contributes to signal 2902, a fourier transform (e.g., Fast Fourier Transform (FFT), Short Time Fourier Transform (STFT), etc.) may be performed on signal 2902 to determine the RPM value. For example, software embedded on wearable device 1320 may perform a fourier transform on signal 2902 from the time period between t0 to t3 to determine the RPM of the grinding wheel or disc from t0 to t 3.

In some embodiments, the RPM of the grinding wheels of the disc may vary over time. For example, a user may push the grinding wheel or disc harder into the workpiece (the friction of the workpiece and thus the rotational speed is reduced), the power level of the abrasive device/tool may be changed, and so on. To address this issue, signal 2902 may be partitioned/sampled into shorter segments, and then software embedded in wearable device 1320 may compute a fourier transform for each shorter segment. For example, fourier transform may be performed on the signal 2902 from a time period between t0 to t1, a time period between t1 to t2, and so on. The RPM for each time period may be plotted to determine a graph of RPM versus time (as shown in fig. 30).

In some embodiments, signal 2902 may consist of multiple base frequencies and/or may have aliasing frequencies/aliases. To determine the exact frequency corresponding to the RPM of the grinding wheel or disc, the frequency with the highest amplitude or the frequency with an amplitude within a predetermined range may be used. Alternatively, in scenarios where signal 2902 is divided into shorter segments, the RPM for a given time period may be determined based on a frequency having an amplitude that exhibits little deviation from a previous time period. Other methods are also possible.

In some embodiments, signal 2902 represents the vibration of wearable device 1320 with respect to a given axis (e.g., an accelerometer may be used to measure and record vibration data in three axes (x, y, and z)). In these cases, a vibration signal may be determined for each axis, and a collective/composite vibration signal for the grinding wheel or disk may be determined by weighting/combining the individual vibration signals for each axis. In some instances, the weighting/combining may be based on occupational safety standards, such as the ISO 5349 standard discussed herein. To illustrate, applying the ISO 5349 standard may involve combining the vibration signals from each axis by means of a root mean square calculation, where the weight of each axis is different in the composite vibration signal. However, other occupational safety criteria and their corresponding algorithms for determining aggregate/composite vibration signals are also contemplated herein. Wearable device 1320 may be configured to additionally and/or alternatively perform those algorithms of the ISO 5349 standard.

As discussed above, a limit may be placed on signal 2902. More specifically, an upper limit 2904 and a lower limit 2906 may be used to represent an upper limit and a lower limit of vibration, where the area between the upper limit 2904 and the lower limit 2906 is the "optimal area" for vibration of the abrasive product/tool. In some embodiments, the upper limit 2904 and the lower limit 2906 may be determined by the manufacturer of the wearable device 1320 or the manufacturer of the abrasive product/tool. In other embodiments, the upper limit 2904 and the lower limit 2906 may be based on occupational safety standards implemented today or in the future. For example, the upper limit 2904 and the lower limit 2906 may be based on standards set by the Occupational Safety and Health Administration (OSHA), the National Institute for Occupational Safety and Health (NIOSH), the european office for work safety and health (EU-OSHA), or the international organization for standardization (ISO). In some cases, the upper limit 2904 and the lower limit 2906 may be based on ISO 5349 exposure.

In some embodiments, the upper limit 2904 and the lower limit 2906 may be determined based on values in firmware installed into the wearable device 1320 at the time of manufacture or user-defined values dynamically loaded into the firmware of the wearable device 1320. In an example, the user-defined value may be communicated to the wearable device 1320 via a user interface component of the wearable device 1320, may be communicated to the wearable device 1320 via a network application (such as the network application described below), or may be communicated to the wearable device 1320 from a cloud computing device (such as the cloud computing device described above). Other possibilities also exist.

Because it may be valuable to the user to keep the vibration of the abrasive product/tool within the optimal area, the wearable device 1320 may determine deviations from the optimal area. For example, wearable device 1320 may determine exposure time 2908, which corresponds to the length of time that the vibration is in the optimal region. The exposure time 2908 may be compared to the total operating time (e.g., t3-t0) to determine the percentage of time within the optimal region. If the percentage of time within the optimal area is low enough, wearable device 1320 may provide information to increase the percentage of time, perhaps by outputting a visual, tactile, and/or audio alert that provides an improvement in operation, a recommended angle of operation, and the like.

As another example, the wearable device 1320 may determine a critical exposure time 2910, which represents a period of vibration above the upper limit 2904. Because operation beyond the critical exposure time 2910 may be harmful to the user, the wearable device 1320 may provide information to reduce the critical exposure time 2910, perhaps by outputting a visual, tactile, and/or audio alert in a manner similar to that described above.

Fig. 30 illustrates a graph 3000 according to an example embodiment. As shown in fig. 30, graph 3000 includes a signal 3002, which may be indicative of the RPM of the grinding wheel or disc over time. That is, the signal 3002 may result from a fourier transform performed on the signal 2902 from the graph 2900. The x-axis of graph 3000 corresponds to time values, while the y-axis corresponds to RPM values (in gRMS).

Similar to graph 2900, graph 3000 includes an upper limit 3004 and a lower limit 3006, representing the upper and lower limits of RPM, respectively. The area between the upper limit 3004 and the lower limit 3006 is the "optimal area" for the RPM of the grinding wheel or disc. In some embodiments, the upper limit 3004 and the lower limit 3006 may be determined by a manufacturer of the wearable device 1320 or a manufacturer of the abrasive product/tool. In other embodiments, the upper limit 3004 and the lower limit 3006 may be based on occupational safety standards implemented today or in the future.

In some embodiments, the upper limit 3004 and the lower limit 3006 may be determined based on values in firmware installed into the wearable device 1320 at the time of manufacture or user-defined values in firmware dynamically loaded into the wearable device 1320. In an example, the user-defined value may be communicated to the wearable device 1320 via a user interface component of the wearable device 1320, may be communicated to the wearable device 1320 via a network application (such as the network application described below), or may be communicated to the wearable device 1320 from a cloud computing device (such as the cloud computing device described above). Other possibilities also exist.

Much like plot 2900, it may be valuable to the user to keep the RPM within the optimal region of plot 3000. Thus, wearable device 1320 may be used to determine the deviation of the RPM from the optimal region. For example, wearable device 1320 may determine a threshold time 3008, which corresponds to a length of time that the RPM is above an upper limit 3004. Likewise, wearable device 1320 may be used to determine a low usage time 3010, which corresponds to a length of time that the RPM is below lower limit 3006. In either case, the wearable device 1320 may provide information to reduce the critical time 3008 and the low use time 3010, perhaps by outputting a visual, tactile, and/or audio alert that provides an operational improvement, a recommended angle of operation, and so forth.

In some embodiments, data from graph 2900 and/or graph 3000 may be transmitted by wearable device 1320 to a cloud computing device for storage and additional computing. For example, the cloud computing device can execute the machine learning algorithm described above to find patterns (e.g., grind time, optimal RPM time, overload time, optimal vibration time, etc.) with respect to signal 2902 and/or signal 3002. The discovered patterns may then be transmitted to a web application that provides information to the user. Additionally and/or alternatively, the network application may include a graph of the vibration of wearable device 1320 over time (e.g., graph 2900) and/or may include a graph of the RPM of wearable device 1320 over time (e.g., graph 3000). The web application may be automatically scalable — capable of viewing on tablet devices, desktop computing devices, mobile devices, and the like. Further, the web application may be configured to establish a private account for each user and may have appropriate security measures to isolate each user's data and ensure privacy. In some embodiments, for example, a cloud computing device or network application may be used to update the firmware of the wearable device 1320 by transmitting a software update to the communication interface 106 of the wearable device 1320.

It is noted that while the above embodiments are discussed with respect to vibration and RPM data, other types of data are also contemplated in the disclosure herein.

In one example, a temperature sensor/relative humidity sensor may be used to provide data regarding the ambient temperature and humidity level around wearable device 1320. In turn, the data collected by the temperature/relative humidity sensors may be used to measure the heat exposure time of the abrasive product/tool operated by the user of the wearable device 1320. For example, a temperature sensor/relative humidity sensor may calculate that the abrasive product/tool is operating in an environment at 55 ° F for 2 hours, and then operating in an environment at 105 ° F for 6 hours. The calculated heat exposure time may then be used to determine the remaining product life/productivity of the abrasive product/tool. For example, if the abrasive product/tool is often operated in a high temperature environment, the expected product life of the abrasive product/tool may be shorter than the life of the abrasive product/tool that is often operated in a medium temperature environment.

In another example, the magnetometer may be used to provide data related to the ambient magnetic field/orientation of the wearable device 1320 or a workpiece operated by a user of the wearable device 1320.

In yet another example, a capacitive sensor may be used to provide data regarding material density or potential damage associated with the wearable device 1320 or the abrasive tool.

In further examples, current measurements may be obtained from an abrasive tool and converted into power data. The power data may be used to provide grinding cycle data for the abrasive tool, and in some cases, may be compared to the aforementioned vibration and RPM data to further understand abrasive operation. Further, the data described above, as well as data from other sensors (such as inertial sensors, pressure sensors, and/or force sensors), may be graphical, transformed, displayed on the dashboard (such as displays 2700, 2710, 2720, and 2730 described above), and associated with upper and lower threshold limits, as similarly described with respect to graphs 2900 and 3000.

Xvi. exemplified examples

Embodiments of the present disclosure may relate to one of the Enumerated Example Embodiments (EEEs) listed below.

EEE 1 is a computer-implemented method comprising:

receiving, at a computing device, sensor data from one or more sensors, wherein the one or more sensors are disposed proximate to an abrasive product or a workpiece associated with the abrasive product, wherein the one or more sensors are configured to collect wear operation data associated with an abrasive operation involving the abrasive product or the workpiece;

Training, with the aid of the computing device and based on the sensor data, the machine learning system to determine product-specific information of the abrasive product or workpiece-specific information of the workpiece; and

a trained machine learning system is provided with the aid of a computing device.

EEE 2 is a computer-implemented method of EEE 1, further comprising: at least a portion of the sensor data is marked to provide marked sensor data, wherein the marked sensor data includes one or more marks, each mark identifying different product-specific information for the abrasive product.

EEE 3 is a computer-implemented method of EEE 2, wherein the one or more markers identify a pattern of wear operation data for a duration of time prior to an abrasive product related event.

EEE 4 is a computer-implemented method of EEE 3, wherein the pattern of wear operation data includes one or more phases, each phase associated with one or more sensor thresholds, wherein the one or more flags are associated with a phase based on the one or more sensor thresholds.

EEE 5 is a computer-implemented method of EEE 1, wherein training a machine learning system includes training one or more machine learning models, wherein each model is trained with sensor data from an abrasive product having a unique identifier from a shared set of identifiers.

EEE 6 is a computer-implemented method of EEE 1, wherein sensor data from one or more sensors is aggregated by a local computing device to provide aggregated sensor data, and wherein receiving the sensor data comprises receiving the aggregated sensor data from the local computing device.

EEE 7 is a computer-implemented method of EEE 1, wherein the one or more sensors are disposed in a wearable device, wherein sensor data from the one or more sensors is aggregated by the wearable device to provide aggregated sensor data, and wherein receiving the sensor data comprises receiving the aggregated sensor data from the wearable device.

EEE 8 is a computer-implemented method of EEE 7, wherein the sensor data includes information indicative of a Revolutions Per Minute (RPM) value of the abrasive product.

EEE 9 is a computer-implemented method comprising:

receiving, at a computing device, sensor data from one or more sensors, wherein the one or more sensors are disposed proximate to an abrasive product or a workpiece associated with the abrasive product, wherein the one or more sensors are configured to collect wear operation data associated with an abrasive operation involving the abrasive product or the workpiece, and wherein the computing device has access to a trained machine learning system configured to receive input sensor data and output product-specific information of the abrasive product or workpiece-specific information of the workpiece;

Determining product-specific information of the abrasive product or workpiece-specific information of the workpiece by applying a trained machine learning system to the sensor data; and

providing the product-specific information or the workpiece-specific information to one or more client devices.

EEE 10 is a computer-implemented method of EEE 9, wherein the one or more sensors are disposed in a wearable device, wherein sensor data from the one or more sensors is aggregated by the wearable device to provide aggregated sensor data, and wherein receiving the sensor data comprises receiving the aggregated sensor data from the wearable device.

EEE 11 is a computer-implemented method of EEE 10, wherein the sensor data includes information indicative of a Revolutions Per Minute (RPM) value of the abrasive product.

EEE 12 is a computer-implemented method of EEE 9, wherein the one or more client devices include at least one of: a wearable device, a mobile device, a dashboard device, a network server, an analytics processing engine, or a third party server.

EEE 13 is a computer-implemented method of EEE 9, wherein providing product-specific information comprises providing information associated with a new abrasive product or a newer abrasive product, wherein the information comprises, at least in part, instructions for constructing the new abrasive product or the newer abrasive product.

EEE 14 is a computer-implemented method of EEE 9, wherein providing product-specific information or workpiece-specific information comprises providing a notification to the one or more client devices.

The EEE 15 is a computer-implemented method of EEE 9, wherein providing product-specific information or workpiece-specific information includes providing one or more product-specific solutions to the one or more client devices for solving a problem with an abrasive product.

EEE 16 is a computer-implemented method of EEE 15, wherein the one or more client devices are configured to:

displaying the one or more product-specific solutions on a graphical user interface,

receiving, via a graphical user interface, a selection of one of the one or more product-specific solutions;

determining training data for a trained machine learning system based on the selected product-specific solution; and

the training data is transmitted to the computing device.

EEE 17 is a computer-implemented method of EEE 9, wherein the abrasive product is a hand-held abrasive product operated by a user.

EEE 18 is a computer-implemented method of EEE 17, wherein the abrasive operation data includes at least one of: vibration data associated with the handheld abrasive product or acceleration data associated with the handheld abrasive product.

EEE 19 is a computer-implemented method of EEE 17, wherein providing product-specific information or workpiece-specific information includes providing a notification to a graphical interface of a wearable device worn by a user.

EEE 20 is a computer-implemented method of EEE 17, wherein the product-specific information includes at least one of: the time taken to perform the task assigned to the user, the user's idle time, or the user's work time.

EEE 21 is a computer-implemented method of EEE 17, wherein the product-specific information comprises at least one of: the working angle of the user relative to the hand-held abrasive product, the working angle of the hand-held abrasive product relative to the workpiece, the gripping force of the user on the hand-held abrasive product, or the pressure applied by the user on the hand-held abrasive product.

The EEE 22 is a computer-implemented method of EEE 17, wherein the product-specific information includes an end-of-life estimate of the handheld abrasive product, wherein the end-of-life estimate includes an estimated amount of time a user can safely use the handheld abrasive product.

EEE 23 is a computer-implemented method of EEE 9, wherein the abrasive product is an automated abrasive product operated by a controller.

EEE 24 is a computer-implemented method of EEE 23, wherein the one or more sensors include a spark-constant sensor configured to collect an operating speed of an automated abrasive product.

EEE 25 is a computer-implemented method of EEE 23, wherein providing product-specific information comprises providing a determination that: one or more abrasive articles of the automated abrasive product are damaged or fail.

EEE 26 is a computer-implemented method of EEE 25, wherein in providing the determination, the computing device is further configured to:

identifying, by a product database, one or more replacement abrasive articles of an automated abrasive product; and

in response to identifying the one or more replacement abrasive articles, a request is made for one or more replacement articles or a refurbishment process.

EEE 27 is a computer-implemented method of EEE 23, wherein providing product-specific information comprises transmitting at least one control instruction to a controller of an automated abrasive product, wherein the at least one control instruction comprises at least one of: adjusting a rotational speed of the automated abrasive product, providing a notification to the automated abrasive product, turning the automated abrasive product on, or turning the automated abrasive product off.

EEE 28 is a computer-implemented method of EEE 9, further comprising:

providing a search interface via a graphical user interface, wherein the search interface comprises a plurality of user-selectable criteria, wherein the user-selectable criteria comprises at least one of: a location menu, an equipment menu, a date range, or a workpiece menu;

Receiving user-selected search criteria from user-selectable criteria via a graphical user interface;

determining one or more metrics based on user-selected search criteria; and

displaying the one or more metrics via a graphical user interface.

EEE 29 is a computer-implemented method of EEE 28, wherein the one or more metrics include at least one of: a grinding time metric, a best grinding metric, a vibration metric, a depth of cut, a current trace, a tool identifier, or a part count.

EEE 30 is a computer-implemented method of EEE 29, further configured to receive, via a graphical user interface, a desired metric selected from a menu of metrics, wherein displaying the one or more metrics is based on the desired metric.

EEE 31 is a computer-implemented method of EEE 9, further comprising: providing, via a graphical user interface, a cycle comparison interface, wherein the cycle comparison interface is configured to display at least a portion of the sensor data in an overlapping arrangement of a plurality of periodic time series.

EEE 32 is a computer-implemented method of EEE 9, further comprising: comparing a plurality of periodic time series of at least a portion of the sensor data, wherein determining product-specific information of the abrasive product or workpiece-specific information of the workpiece is based at least in part on the comparison.

EEE 33 is a computer-implemented method of EEE 9, wherein determining product-specific information of an abrasive product or workpiece-specific information of a workpiece includes determining one or more of: a predicted future condition of the abrasive product or a predicted future condition of the workpiece.

The EEE 34 is a computer-implemented method of the EEE 33, wherein providing product-specific information or workpiece-specific information includes providing at least one of: a predicted future condition of the abrasive product or a predicted future condition of the workpiece.

EEE 35 is a computer-implemented method of EEE 9, wherein determining product-specific information of an abrasive product or workpiece-specific information of a workpiece includes determining a normative action.

EEE 36 is a computer-implemented method of EEE 35, wherein providing product-specific information or workpiece-specific information comprises providing a normative action, wherein the normative action comprises at least one of: adjusting the operating parameters of the grinding machine, performing maintenance operations, repairing abrasive products, or replacing abrasive products.

EEE 37 is a computing system that includes:

one or more processors; and

a data store, wherein the data store has stored thereon computer-executable instructions that, when executed by the one or more processors, cause the computing system to perform operations comprising:

Receiving sensor data from one or more sensors, wherein the one or more sensors are disposed proximate to an abrasive product or a workpiece associated with the abrasive product, wherein the one or more sensors are configured to collect wear operation data associated with an abrasive operation involving the abrasive product or the workpiece;

training a machine learning system based on the sensor data to determine product-specific information of the abrasive product or workpiece-specific information of the workpiece; and

a trained machine learning system is provided.

EEE 38 is a computing system of EEE 37, wherein the operations further comprise: at least a portion of the sensor data is marked to provide marked sensor data, wherein the marked sensor data includes one or more marks, each mark identifying different product-specific information for the abrasive product.

EEE 39 is a computing system of EEE 38, wherein the one or more markers identify a pattern of wear operation data for a duration of time prior to an abrasive product related event.

EEE 40 is a computing system of EEE 39, wherein the pattern of wear operation data includes one or more phases, each phase associated with one or more sensor thresholds, wherein the one or more flags are associated with a phase based on the one or more sensor thresholds.

EEE 41 is a computing system for EEE 37, wherein training a machine learning system includes training one or more machine learning models, wherein each model is trained with sensor data from an abrasive product having a unique identifier from a shared set of identifiers.

EEE 42 is a computing system of EEE 37, wherein the sensor data is aggregated by the local computing device to provide aggregated sensor data, and wherein receiving the sensor data includes receiving the aggregated sensor data from the local computing device.

EEE 43 is a computing system of EEE 37, wherein the one or more sensors are disposed in a wearable device, wherein sensor data from the one or more sensors is aggregated by the wearable device to provide aggregated sensor data, and wherein receiving the sensor data comprises receiving the aggregated sensor data from the wearable device.

EEE 44 is a computing system for EEE 37, wherein the sensor data includes information indicative of a Revolutions Per Minute (RPM) value of the abrasive product.

EEE 45 is a computing system that includes:

a trained machine learning system configured to receive input sensor data and output product-specific information or workpiece-specific information based on the input sensor data; and

A computing device configured to:

receiving sensor data from one or more sensors, wherein the one or more sensors are disposed proximate to an abrasive product or a workpiece associated with the abrasive product, wherein the one or more sensors are configured to collect wear operation data associated with an abrasive operation involving the abrasive product or the workpiece;

determining product-specific information of the abrasive product or workpiece-specific information of the workpiece by applying a trained machine learning system to the sensor data; and

providing the product-specific information or the workpiece-specific information to one or more client devices.

EEE 46 is a computing system of EEE 45, wherein the one or more client devices include at least one of: a wearable device, a mobile device, a dashboard device, a network server, an analytics processing engine, or a third party server.

EEE 47 is a computing system of EEE 45, wherein providing product-specific information comprises providing information associated with a new abrasive product or a newer abrasive product, wherein the information comprises, at least in part, instructions for constructing the new abrasive product or the newer abrasive product.

EEE 48 is a computing system of EEE 45, wherein providing product-specific information or workpiece-specific information includes providing notifications to the one or more client devices.

The EEE 49 is a computing system of the EEE 45, wherein providing the product-specific information or the workpiece-specific information includes providing one or more product-specific solutions to the one or more client devices for solving the problem with the abrasive product.

EEE 50 is a computing system of EEE 49, wherein the one or more client devices are configured to:

displaying the one or more product-specific solutions on a graphical user interface,

receiving, via a graphical user interface, a selection of one of the one or more product-specific solutions;

determining training data for a trained machine learning system based on the selected product-specific solution; and

the training data is transmitted to the computing device.

EEE 51 is the computing system of EEE 45, wherein the abrasive product is a hand-held abrasive product operated by a user.

EEE 52 is a computing system of EEE 51, wherein the abrasive operation data includes at least one of: vibration data associated with the handheld abrasive product or acceleration data associated with the handheld abrasive product.

EEE 53 is a computing system of EEE 51, wherein providing product-specific information or workpiece-specific information includes providing a notification to a graphical interface of a wearable device worn by a user.

EEE 54 is a computing system of EEE 51, wherein the product specific information includes at least one of: the time taken to perform the task assigned to the user, the user's idle time, or the user's work time.

EEE 55 is a computing system of EEE 51, wherein the product specific information comprises at least one of: the working angle of the user relative to the hand-held abrasive product, the working angle of the hand-held abrasive product relative to the workpiece, the gripping force of the user on the hand-held abrasive product, or the pressure applied by the user on the hand-held abrasive product.

The EEE 56 is a computing system of the EEE 51, wherein the product specific information includes an end-of-life estimate of the handheld abrasive product, wherein the end-of-life estimate includes an estimated amount of time that a user can safely use the handheld abrasive product.

EEE 57 is a computing system for EEE 45, wherein the abrasive product is an automated abrasive product operated by a controller.

EEE 58 is a computing system of EEE 57, wherein the one or more sensors include a spark-constant sensor configured to gather an operating speed of the automated abrasive product.

EEE 59 is a computing system of EEE 57, wherein providing product specific information includes providing the following determinations: one or more abrasive articles of the automated abrasive product are damaged or fail.

EEE 60 is a computing system of EEE 59, wherein in providing the determination, the computing device is further configured to:

identifying, by a product database, one or more replacement abrasive articles of an automated abrasive product; and

in response to identifying the one or more replacement abrasive articles, a request is made for one or more replacement articles.

EEE 61 is a computing system of EEE 57, wherein providing product-specific information comprises transmitting at least one control instruction to a controller of an automated abrasive product, wherein the at least one control instruction comprises at least one of: adjusting a rotational speed of the automated abrasive product, providing a notification to the automated abrasive product, turning the automated abrasive product on, or turning the automated abrasive product off.

EEE 62 is a computing system of EEE 45, wherein the one or more sensors are disposed in a wearable device, wherein sensor data from the one or more sensors is aggregated by the wearable device to provide aggregated sensor data, and wherein receiving the sensor data comprises receiving the aggregated sensor data from the wearable device.

EEE 63 is a computing system for EEE 59, wherein the sensor data includes information indicative of a Revolutions Per Minute (RPM) value of the abrasive product.

EEE 64 is a computing system of EEE 45, further comprising:

a display configured to provide a graphical user interface having a search interface, wherein the search interface includes a plurality of user-selectable criteria, wherein the user-selectable criteria include at least one of: a location menu, a device menu, a date range, or a workpiece menu, wherein the graphical user interface is configured to:

receiving user-selected search criteria from user-selectable criteria

Determining one or more metrics based on user-selected search criteria; and

displaying the one or more metrics via a graphical user interface.

EEE 65 is a computing system of EEE 64, wherein the one or more metrics include at least one of: a grinding time metric, a best grinding metric, a vibration metric, a depth of cut, a current trace, a tool identifier, or a part count.

EEE 66 is a computing system of EEE 65, wherein the graphical user interface is further configured to receive a desired metric selected from a menu of metrics, wherein displaying the one or more metrics is based on the desired metric.

EEE 67 is a computing system of EEE 64, wherein the display is further configured to provide, via the graphical user interface, a cyclical comparison interface, wherein the cyclical comparison interface is configured to display at least a portion of the sensor data in an overlapping arrangement of a plurality of periodic time series.

EEE 68 is a computing system of EEE 45, wherein the computing device is further configured to:

comparing a plurality of periodic time series of at least a portion of the sensor data, wherein determining product-specific information of the abrasive product or workpiece-specific information of the workpiece is based at least in part on the comparison.

EEE 69 is a computing system of EEE 45, wherein the product-specific information or the workpiece-specific information includes information indicative of one or more operating parameters of the remote device.

EEE 70 is a computing system that includes:

a trained machine learning system configured to receive input sensor data and output product-specific information related to an abrasive product or workpiece-specific information related to a workpiece associated with the abrasive product based on the input sensor data; and

a computing device configured to:

receiving sensor data from one or more sensors, wherein the one or more sensors are disposed proximate to a plurality of abrasive products, and wherein the one or more sensors are configured to collect wear operation data associated with an abrasive operation involving the plurality of abrasive products;

Determining product-specific information for the plurality of abrasive products or workpiece-specific information for a plurality of workpieces associated with the plurality of abrasive products by applying a trained machine learning system to sensor data; and

providing the product-specific information or the workpiece-specific information to one or more client devices.

EEE 71 is a computing system of EEE 70, wherein the plurality of abrasive products are located at a plurality of enterprises.

EEE 72 is a computing system of EEE 71, wherein the computing device is configured to anonymize the sensor data in order to separate the sensor data from each of the plurality of enterprises.

The EEE 73 is a computing system of the EEE 70, wherein the sensor data includes composite sensor data, wherein the composite sensor data is generated by a digital twinning of the plurality of abrasive products.

The EEE 74 is a computing device that includes:

one or more processors; and

a data store, wherein the data store has stored thereon computer-executable instructions that, when executed by the one or more processors, cause the computing device to perform functions of a computer-implemented method comprising any of EEEs 1-36.

EEE 75 is an article of manufacture comprising one or more computer-readable media storing non-transitory computer-readable instructions that, when executed by one or more processors of a computing device, cause the computing device to perform functions of a computer-implemented method comprising any of EEEs 1-36.

The EEE 76 is a computing device that includes:

means for performing the computer-implemented method of any of the EEEs 1-36.

EEE 77 is a system comprising:

an abrasive product comprising a marking, wherein the abrasive product is associated with a Universally Unique Identifier (UUID);

a mobile device comprising a user interface and an indicia reader, wherein the mobile device is configured to perform operations comprising:

interrogating the tag using the tag reader to determine the UUID;

based on the UUID, performing at least one of:

displaying, via a user interface, inventory information regarding similar abrasive products;

displaying product information or GPS information about a similar abrasive product via a user interface; or

A reorder request for a similar abrasive product is generated and transmitted to a client network or a remote network.

EEE 78 is a method comprising:

interrogating the indicia of the abrasive product with an indicia reader of the mobile device to determine an associated Universally Unique Identifier (UUID);

based on the UUID, performing at least one of:

displaying, via a user interface of a mobile device, inventory information regarding similar abrasive products;

Displaying product information or GPS information about a similar abrasive product via a user interface; or

A reorder request for a similar abrasive product is generated and transmitted to a client network or a remote network.

EEE 79 is a method comprising:

obtaining tag information from a client network via a tag reader, wherein the tag information communicates with

Associating the marking of the abrasive product with a unique identifier (UUID);

transmitting the tagging information to a remote network;

updating at least one database with the label information;

displaying the marker information using a user interface;

selecting at least one actionable action based on user interaction with a user interface, wherein

The at least one actionable action comprises at least one of:

establishing communication with an abrasive device, the abrasive device associated with an abrasive product;

adjusting operation of an abrasive apparatus associated with the abrasive product;

notifying the supplier of the order of the one or more replacement parts;

notifying the supplier to repair or refurbish the abrasive product; or

The user of the abrasive product is notified by means of a text message or an email.

EEE 80 is a method comprising:

Interrogating an indicia of an abrasive product with an indicia reader of a mobile device to determine an associated universal uniqueness

An identifier (UUID);

determining an operating state of the abrasive product from a plurality of possible operating states based on the UUID, wherein

The plurality of possible operating states comprises at least one of:

a manufacturer status;

a door state;

a workplace state;

a wheel equilibrium state;

an operating state; and

a waste state; and

storing an operational state of the abrasive product in at least one of: a product database or a memory associated with the tag.

EEE 81 is a mobile device comprising:

an authentication system configured to determine an access level from a plurality of possible access levels based on a user identification; and

a user interface configured to provide an abrasive product based on the determined level of access

Information, wherein the abrasive product information comprises at least one of:

maintaining the information;

GPS information;

the status of the client network;

abrasive product details; or

Current inventory information.