阅读说明：本技术 一种基于边缘计算的刀具监测管理方法及系统 (Cutter monitoring and managing method and system based on edge calculation ) 是由 马磊 唐舒琪 于 2021-08-26 设计创作，主要内容包括：本发明提供的一种基于边缘计算的刀具监测管理系统及方法,所述管理系统包括：边缘计算管理平台、刀具监测模块,云管理中心模块,边缘云节点；所述边缘计算管理平台用于配置边缘计算应用；所述刀具监测模块用于监测数据变化情况；所述云管理中心模块与所述边缘计算管理平台连接,用于统一集中管理多个边缘计算管理平台；所述边缘云节点用于对所在区域的刀具监测设备上传的采集信息进行处理,获得处理结果数据,并将所述处理结果数据发送至所述云管理中心模块。其中边缘计算是在靠近物或者数据源头的网络边缘侧,融合网络、计算、存储、应用核心能力的分布式开放平台,就近提供边缘智能监测服务。(The invention provides a cutter monitoring and managing system and method based on edge calculation, wherein the managing system comprises: the system comprises an edge computing management platform, a cutter monitoring module, a cloud management center module and edge cloud nodes; the edge computing management platform is used for configuring an edge computing application; the cutter monitoring module is used for monitoring the data change condition; the cloud management center module is connected with the edge computing management platforms and is used for uniformly and centrally managing the edge computing management platforms; the edge cloud node is used for processing the acquired information uploaded by the cutter monitoring equipment in the area to obtain processing result data and sending the processing result data to the cloud management center module. The edge calculation is a distributed open platform which integrates network, calculation, storage and application core capabilities at the edge side of a network close to an object or a data source and provides edge intelligent monitoring service nearby.)

1. An edge-computing based tool monitoring and management system, the management system comprising: the system comprises an edge computing management platform, a cutter monitoring module, a cloud management center module and edge cloud nodes;

the edge computing management platform is used for configuring an edge computing application;

the cutter monitoring module is used for monitoring the data change condition;

the cloud management center module is connected with the edge computing management platforms and is used for uniformly and centrally managing the edge computing management platforms;

the edge cloud node is used for processing the acquired information uploaded by the cutter monitoring equipment in the area to obtain processing result data and sending the processing result data to the cloud management center module.

2. The system of claim 1, wherein the monitored data change comprises data processing, model development, training, management, deployment.

3. The system for monitoring and managing the cutting tools based on the edge computing as claimed in claim 1, wherein the cloud management center module is used for installation and arrangement of the whole cloud platform, management of a virtual host cloud and a container cloud, pushing of the edge cloud platform through an application arrangement template, automatic pulling of an application, and automatic pushing of an artificial intelligence computing model.

4. The system for monitoring and managing cutters based on edge computing as claimed in claim 1, wherein the edge cloud node is further configured to transmit structured data back to the cloud management center module, and return the computing result of the monitoring service to the upper layer application of the user.

5. A tool monitoring and managing method based on edge calculation is characterized by comprising the following steps:

the cloud management center manages a plurality of edge cloud centers in a unified mode, is responsible for unified arrangement of the application on the container platform, pushes the edge cloud platform application to be automatically pulled up through the arrangement template, and is responsible for automatic pushing of the artificial intelligence computing model;

the edge cloud center manages and collects data of cutter monitoring in the area, local intelligent analysis processing is carried out by utilizing edge node computing resources, preprocessing and pre-judging are carried out on the data, the structured data are transmitted back to the global center, and the computing result of cutter monitoring is returned to the upper application of a user.

Technical Field

The invention relates to the field of edge calculation, in particular to a cutter monitoring and managing method and system based on edge calculation.

Background

Conventional tool machining systems face a number of challenges. At present, cutter monitoring solutions such as a cutter processing counting prediction method, a sensor physical monitoring method and the like exist in the market, and the defects of marginalization of calculation and analysis, high cost, slow deployment, low accuracy and the like exist. The overall maintenance, monitoring and upgrading cost is high, and the manual support cost is high. The software and hardware updating speed is low, hardware such as a workstation and the like is replaced every 3-5 years, and the aging performance of hardware equipment at the later stage is reduced; the design software adopts a permanent license mode, and is difficult to maintain the latest version for a long time. The search for an edge intelligent platform and the provision of an efficient and reliable intelligent Internet of things system for monitoring the cutter is a problem to be solved urgently by current manufacturing enterprises.

Disclosure of Invention

In view of the above, the present invention has been developed to provide an edge-calculation-based tool monitoring management method and system that overcomes or at least partially solves the above-mentioned problems.

According to an aspect of the present invention, there is provided a tool monitoring management system based on edge calculation, the management system comprising: the system comprises an edge computing management platform, a cutter monitoring module, a cloud management center module and edge cloud nodes;

the edge computing management platform is used for configuring an edge computing application;

the cutter monitoring module is used for monitoring the data change condition;

the cloud management center module is connected with the edge computing management platforms and is used for uniformly and centrally managing the edge computing management platforms;

the edge cloud node is used for processing the acquired information uploaded by the cutter monitoring equipment in the area to obtain processing result data and sending the processing result data to the cloud management center module.

Optionally, the monitoring of the data change condition specifically includes data processing, model development, training, management, and deployment.

Optionally, the cloud management center module is used for installation and arrangement of the whole cloud platform, management of the virtual host cloud and the container cloud, pushing of the edge cloud platform through the application arrangement template, automatic application pull-up, and meanwhile, automatic pushing of the artificial intelligence computing model.

Optionally, the edge cloud node is further configured to transmit the structured data back to the cloud management center module, and return a calculation result of the monitoring service to the upper-layer application of the user.

A tool monitoring management method based on edge calculation, the management method comprising:

the cloud management center manages a plurality of edge cloud centers in a unified mode, is responsible for unified arrangement of the application on the container platform, pushes the edge cloud platform application to be automatically pulled up through the arrangement template, and is responsible for automatic pushing of the artificial intelligence computing model;

the edge cloud center manages and collects data of cutter monitoring in the area, local intelligent analysis processing is carried out by utilizing edge node computing resources, preprocessing and pre-judging are carried out on the data, the structured data are transmitted back to the global center, and the computing result of cutter monitoring is returned to the upper application of a user.

The invention provides a cutter monitoring and managing method and system based on edge calculation, wherein the edge calculation is a distributed open platform integrating network, calculation, storage and application core capabilities on the edge side of a network close to an object or a data source, and edge intelligent monitoring service is provided nearby.

The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

FIG. 1 is a diagram of an intelligent manufacturing edge calculation solution provided by an embodiment of the present invention;

FIG. 2 is a block diagram of a smart IOT system for tool monitoring based on edge cloud computing according to an embodiment of the present invention;

FIG. 3 is a diagram of AI development platform components for application developers according to an embodiment of the present invention;

fig. 4 is an edge cloud computing node component diagram of an intelligent tool monitoring internet of things system based on edge cloud computing according to an embodiment of the present invention;

FIG. 5 is a flowchart of a technique for acquiring and analyzing edge side tool load data according to an embodiment of the present invention;

fig. 6 is a deep convolution long-time memory neural network model adopted in the tool monitoring system according to the embodiment of the present invention.

Detailed Description

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

The terms "comprises" and "comprising," and any variations thereof, in the present description and claims and drawings are intended to cover a non-exclusive inclusion, such as a list of steps or elements.

The technical solution of the present invention is further described in detail with reference to the accompanying drawings and embodiments.

As shown in fig. 1, an intelligent manufacturing edge calculation solution diagram is provided by an embodiment of the present invention. As shown in fig. 1, the platform contains edge App lifecycle management, edge PaaS lifecycle management, edge IaaS lifecycle management, edge storage lifecycle management, and so on; the edge computing platform provides elastic computing service based on cloud edge nodes and side nodes of the intelligent tool monitoring system, can meet key requirements of the tool monitoring system in the aspects of response, near data processing, intelligent data analysis, rapid application deployment and the like in business real time, and can flexibly configure and manage large-scale edge computing application.

Fig. 2 is a system architecture diagram of a tool monitoring smart internet of things system based on edge cloud computing according to an embodiment of the present invention. As shown in FIG. 2, the system hierarchy design includes a cloud pipe center (local or off-site) and edge cloud platforms. The cloud pipe center is used for centrally and uniformly managing a plurality of edge cloud platforms. The cloud management center is responsible for installation and arrangement of the whole cloud platform, management of the virtual host cloud and the container cloud, pushing of edge computing nodes through the application arrangement template, automatic application pull-up and automatic pushing of an artificial intelligence computing model;

the edge cloud node is used for managing and collecting cutter monitoring in the area, preprocessing and pre-judging data, transmitting the structured data back to the cloud management center, and returning the calculation result of the monitoring service to the upper application of the user.

FIG. 3 is a developer-oriented one-stop AI development platform providing mass data preprocessing and semi-automated labeling, large-scale distributed tracking, automated model generation, and end-edge-cloud model on-demand deployment capabilities for machine learning and deep learning; and the rapid development and deployment of the service are realized.

The lower side contains infrastructure management, an operating system, heterogeneous hardware. The upper-layer application analyzes the data and predicts the result, the collected data are compared by the rule engine, the calculation of a certain function is triggered when the collected data meet the conditions, and the examples of the function calculation are uniformly managed by the local function calculation and the background engine. After the development is completed, integrating basic configuration files of all components, and verifying the deployability and the security of the package; the state of the package is changed into formal online release; and deploying the formally online package to the edge computing systems of the types corresponding to different application scenes by the deployment operation service.

Fig. 4 is a diagram of an edge cloud computing node component of a smart internet of things system for tool monitoring based on edge cloud computing according to an embodiment of the present invention; as shown in fig. 4, the platform includes an internet of things infrastructure, edge cloud computing nodes, and a cloud management center. Compared with the traditional tool monitoring system, the local computing node is increased.

The infrastructure of the internet of things comprises various internet of things devices (such as an environment sensor, an RFID label, a camera, a smart phone, an industrial billboard and the like), mainly completes the functions of collecting and reporting original data, and takes the form of an event source as the input of application service.

The edge computing node comprises an edge gateway and a service component, wherein the service component comprises local storage, task scheduling, a rule engine, log management and the like. The edge nodes strengthen the computing power of the local computing nodes, and some simple business logics directly process and return information to the user equipment at the edge nodes.

The edge computing node realizes basic service response by reasonably deploying and allocating computing and storage capacity of the network edge side and opening an API.

And the cloud management center calls operation management and equipment management through an API. The cloud management center realizes the uniform resource scheduling and the uniform resource receiving management through the orchestrator and is responsible for the mirror image distribution. The cloud management center remotely manages the local computing nodes by calling the API. The system is mainly responsible for equipment management, component management and node management of local computing nodes, automatic pushing application arrangement and self-adaptive deployment without manual monitoring and deployment.

The edge cloud center and the cloud pipe center are integrated in the form of an open API. And the local computing node and the cloud management center can be conveniently called. Whether the manual operation of the machine tool is standard or not can be remotely monitored and managed, the specific position, the state and the like of the cutter in workshop processing can be monitored and tracked in real time, all data are collected, and real-time planning and scheduling and accurate control are improved.

As shown in fig. 5, the process includes a CNC machine control system data acquisition and integration service, a tool torque monitoring and tool life correlation analysis and prediction service, a tool setting gauge & product measurement data & tool load data integration and intelligent compensation, a tool & parameter information base construction and intelligent tool selection system, and a CNC digital intelligent workshop integration service.

As shown in fig. 6. And preprocessing the acquired cutter data, and selecting a deep convolution long-time memory neural network model as a target detection model. The deep learning-based DeepConvLSTM algorithm integrates Convolution (Convolution) and LSTM operation, and can learn the spatial attribute and the time attribute of a sample. In the convolution operation, the waveform signal is filtered by multiplying the signal with a convolution kernel, preserving the high level information. In LSTM operation, the timing relationship between signals is discovered by remembering or forgetting preamble information.

Edge computing is used as a new computing mode, computing and storage resources are deployed on the side of terminal equipment, and therefore computing capacity with higher real-time performance and response capacity of services are obtained; non-critical data are processed at the edge side and do not need to be uploaded to a data center, so that network overhead and resource pressure of cloud computing are greatly reduced. The edge computing node provides the capability of edge computing based on the technologies of a rule engine, function computing and the like, a service program running on the edge computing node actively acquires data of the terminal equipment and executes partial computing tasks on the edge side so as to reduce transmission of useless data between the edge computing node and a platform. More efficient and reliable tool monitoring can be provided.

Has the advantages that:

the deep convolution long-time memory neural network model is used as a training frame of a target detection model, network structures and loss functions of a plurality of models are modified in a targeted mode, a detection scheme is designed according to detection requirements and detection effects, the target detection network is trained and classified, the advantages and disadvantages of the schemes are compared from the aspects of detection accuracy, recall rate, detection frame rate and the like, and an optimal scheme is selected. After the used model is determined, the flow shown in the block diagram 3 is adopted in the deployment stage, the model structure is optimized, the trained weight is quantized, the model reasoning is accelerated, and the detection efficiency is further improved under the condition that the precision is not lost. Through the model, parameters are modified to achieve on-line deep learning, actions are selected, then on-site feedback is carried out, production is continuously optimized, and quality loss caused by abnormal cutters is effectively avoided.

The platform comprises edge App lifecycle management, edge PaaS lifecycle management, edge IaaS lifecycle management, edge storage lifecycle management, and the like; the rapid deployment of the application is met, the large-scale edge computing application can be flexibly configured and managed, and the landing of the edge application is accelerated; the problems that the software and hardware updating speed is low and the aging performance of hardware equipment is reduced in the later period are effectively solved; the perpetual license mode may be employed to keep newer versions, depending on the actual use of the software version to which the design is to be avoided. The computing and storage resources are deployed on the side of the terminal equipment, so that the computing capability and the service response capability with higher real-time performance are obtained, and the defects of slow data transmission and complex computation are overcome; non-critical data are processed at the edge side and do not need to be uploaded to a data center, so that network overhead and resource pressure of cloud computing are greatly reduced. And the cutter monitoring efficiency can be improved by optimizing resource allocation through calculation resource scheduling. The method is easy to expand, integrates edge virtual machines, storage and edge services nearby, combines system application, and expands and contracts the edge application virtual environment.

The edge device provides the capability of edge calculation based on the technologies of a rule engine, function calculation and the like, a service program running on the edge device actively acquires data of the terminal device, and partial calculation tasks are executed on the edge side, so that transmission of useless data between the edge device and the platform is reduced. The edge data can be uniformly managed on the core cloud platform, so that management and monitoring are facilitated.

The method comprises the steps of adopting a deep convolution long-time memory neural network model as a target detection model training framework, pertinently modifying network structures and loss functions of a plurality of models, designing a detection scheme according to detection requirements and detection effects, training a target detection network and classifying, comparing the advantages and disadvantages of the schemes from the aspects of detection accuracy, recall rate, detection frame rate and the like, and selecting an optimal scheme. After the used model is determined, the model structure is optimized, the trained weight is quantized, model reasoning is accelerated, and the detection efficiency is further improved under the condition that the precision is not lost.

The above embodiments are provided to further explain the objects, technical solutions and advantages of the present invention in detail, it should be understood that the above embodiments are merely exemplary embodiments of the present invention and are not intended to limit the scope of the present invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.