阅读说明：本技术 控制系统、支持装置以及支持程序 (Control system, support device, and support program ) 是由 川合功 于 2020-02-14 设计创作，主要内容包括：实现能够将必要的异常检测配置在必要的场所的结构。用于对控制对象进行控制的控制系统包含能够执行运算处理的多个处理资源(30-1～30-6)。多个处理资源包含：收集单元(32),其收集与控制对象所包含的任意的检测对象对应的1个或多个状态值；以及异常检测单元(34),其基于根据收集到的1个或多个状态值计算出的特征量,计算表示检测对象发生了异常的可能性的值。收集单元和异常检测单元分别能够配置于多个处理资源中的任意的处理资源。(A structure capable of arranging necessary abnormality detection at a necessary place is realized. A control system for controlling a controlled object includes a plurality of processing resources (30-1 to 30-6) capable of executing arithmetic processing. The plurality of processing resources includes: a collection unit (32) that collects 1 or more state values corresponding to arbitrary detection objects included in the control object; and an abnormality detection unit (34) that calculates a value indicating the possibility of an abnormality occurring in the detection target, based on the feature amount calculated from the collected 1 or more state values. The collection unit and the abnormality detection unit may be respectively disposed in any of the plurality of processing resources.)

1. A control system for controlling a control object, wherein,

the control system has a plurality of processing resources capable of performing arithmetic processing,

the plurality of processing resources have:

a collection unit that collects 1 or more state values corresponding to arbitrary detection objects included in the control object; and

an abnormality detection means for calculating a value indicating a possibility of occurrence of an abnormality in the detection target based on the feature amount calculated from the collected 1 or more state values,

the collection unit and the abnormality detection unit may be respectively configured to any processing resource among the plurality of processing resources.

2. The control system of claim 1,

the abnormality detection means generates a determination result indicating whether or not an abnormality has occurred in the detection target, based on a value indicating a possibility of occurrence of an abnormality in the detection target.

3. The control system of claim 2,

when the collection unit and the abnormality detection unit are disposed in different processing resources, the abnormality detection unit transmits the generated determination result to the processing resource in which the collection unit is disposed.

4. The control system according to any one of claims 1 to 3,

in a case where the collection unit and the abnormality detection unit are configured in different processing resources, the collection unit transmits the collected 1 or more state values to the processing resource in which the abnormality detection unit is configured.

5. The control system according to any one of claims 1 to 4,

when the collection means and the abnormality detection means are disposed in different processing resources, the abnormality detection means transmits the calculated value indicating the possibility of occurrence of an abnormality in the detection target to the processing resource in which the collection means is disposed.

6. The control system according to any one of claims 1 to 5,

the control system further has a support device for assisting the decision of the processing resources to be the arrangement destinations of the collection unit and the abnormality detection unit.

7. The control system of claim 6,

the support device determines the arrangement destinations of the collection unit and the abnormality detection unit based on at least one of the number of state values to be collected by the collection unit, the collection destination of the state values, the specifications of the plurality of processing resources, the load factor of the internal bus, and the load factor of the network.

8. The control system according to claim 6 or 7,

the supporting apparatus transmits required data to a processing resource in which the collecting unit and the abnormality detecting unit should be configured.

9. A support device used in a control system for controlling a control object, wherein,

the control system has a plurality of processing resources capable of performing arithmetic processing,

the plurality of processing resources have:

a collection unit that collects 1 or more state values corresponding to arbitrary detection objects included in the control object; and

an abnormality detection means for calculating a value indicating a possibility of occurrence of an abnormality in the detection target based on the feature amount calculated from the collected 1 or more state values,

the support apparatus provides a user interface for assisting decision of a processing resource to be a configuration destination of the collection unit and the abnormality detection unit.

10. A support program for realizing a support apparatus for use in a control system for controlling a control object, wherein,

the control system has a plurality of processing resources capable of performing arithmetic processing,

the plurality of processing resources have:

a collection unit that collects 1 or more state values corresponding to arbitrary detection objects included in the control object; and

an abnormality detection means for calculating a value indicating a possibility of occurrence of an abnormality in the detection target based on the feature amount calculated from the collected 1 or more state values,

the support program causes a computer to realize a function of providing a user interface for assisting decision of a processing resource to be a configuration destination of the collection unit and the abnormality detection unit.

Technical Field

The present invention relates to a control system capable of detecting some abnormality that may occur in a detection target, a support device used in the control system, and a support program for implementing the support device.

Background

In various production sites, there is a need to improve the operation rate of equipment by anticipatory maintenance of machinery and equipment. The predictive maintenance is a maintenance method in which some abnormality occurring in a machine or an apparatus is detected, and if the equipment is not stopped, maintenance work such as maintenance or replacement is performed before the equipment is brought into a state.

To implement predictive maintenance, the following mechanism is required: the state values of the machine and the device are collected, and whether the machine and the device have some abnormity or not is judged based on the collected state values.

For example, japanese patent application laid-open No. 2018-097662 (patent document 1) discloses a technique capable of monitoring a phenomenon occurring in a control target at a shorter cycle.

Documents of the prior art

Patent document

Patent document 1: japanese patent laid-open publication No. 2018 and 097662

Disclosure of Invention

Problems to be solved by the invention

As disclosed in japanese patent application laid-open No. 2018 and 097662 (patent document 1), a relatively large amount of resources are required to monitor a phenomenon occurring in a controlled object in a shorter period. On the other hand, in the conventional control device, there may be a case where resources necessary for abnormality detection cannot be sufficiently secured.

An object of the present invention is to realize a configuration capable of arranging necessary abnormality detection at a necessary place.

Means for solving the problems

According to an example of the present invention, a control system for controlling a control target is provided. The control system includes a plurality of processing resources capable of executing arithmetic processing. The plurality of processing resources includes: a collection unit that collects 1 or more state values corresponding to arbitrary detection objects included in the control object; and an abnormality detection unit that calculates a value indicating a possibility of occurrence of an abnormality in the detection target based on the feature amount calculated from the collected 1 or more state values. The collection unit and the abnormality detection unit may be respectively disposed in any of the plurality of processing resources.

With this configuration, even when the processing resources are insufficient, it is possible to arrange abnormality detection in a necessary portion.

The abnormality detection means may generate a determination result indicating whether or not the detection target is abnormal, based on a value indicating a possibility that the detection target is abnormal. With this configuration, by referring to the determination result, it is possible to execute processing necessary when an abnormality occurs in the detection target.

In the case where the collecting means and the abnormality detecting means are disposed in different processing resources, the abnormality detecting means may transmit the generated determination result to the processing resource in which the collecting means is disposed. According to this configuration, the plurality of processing resources can cooperate to realize the abnormality detection processing.

In the case where the collection unit and the abnormality detection unit are disposed in different processing resources, the collection unit may transmit the collected 1 or more state values to the processing resource in which the abnormality detection unit is disposed. According to this configuration, the result of abnormality detection can be easily obtained in the processing resource that collects 1 or more state values.

When the collecting means and the abnormality detecting means are disposed in different processing resources, the abnormality detecting means may transmit the calculated value indicating the possibility of occurrence of an abnormality in the detection target to the processing resource in which the collecting means is disposed. According to this configuration, in the processing resource that collects 1 or more state values, a value indicating the possibility of occurrence of an abnormality in the detection target can be easily obtained.

The system may have support means for assisting in the decision of the processing resources to be the configuration destinations of the collection unit and the abnormality detection unit. According to this configuration, the abnormality detection process can be easily installed in the control system by using the support device.

The support apparatus may determine the arrangement destination of the collection means and the abnormality detection means based on at least one of the number of state values to be collected by the collection means, the collection destination of the state values, specifications of the plurality of processing resources, a load rate of the internal bus, and a load rate of the network. According to this structure, the abnormality detection process can be appropriately installed based on 1 or more factors.

The support apparatus may also transmit required data to a processing resource in which the collection unit and the abnormality detection unit should be configured. According to this configuration, data necessary for installing the abnormality detection process in the control system can be easily arranged.

According to another example of the present invention, there is provided a support device used in a control system for controlling a control object. The control system includes a plurality of processing resources capable of executing arithmetic processing. The plurality of processing resources includes: a collection unit that collects 1 or more state values corresponding to arbitrary detection objects included in the control object; and an abnormality detection unit that calculates a value indicating a possibility of occurrence of an abnormality in the detection target based on the feature amount calculated from the collected 1 or more state values. The support apparatus provides a user interface for assisting the decision of the processing resources to be the configuration destinations of the collection unit and the abnormality detection unit.

According to this configuration, it is possible to easily install the abnormality detection process using 1 or more processing resources included in the control system.

According to still another example of the present invention, there is provided a support program for implementing a support apparatus used in a control system for controlling a control object. The control system includes a plurality of processing resources capable of executing arithmetic processing. The plurality of processing resources includes: a collection unit that collects 1 or more state values corresponding to arbitrary detection objects included in the control object; and an abnormality detection unit that calculates a value indicating a possibility of occurrence of an abnormality in the detection target based on the feature amount calculated from the collected 1 or more state values. The support program causes the computer to realize a function of providing a user interface for assisting decision of a processing resource to be a configuration destination of the collection unit and the abnormality detection unit.

According to this configuration, it is possible to easily install the abnormality detection process using 1 or more processing resources included in the control system.

ADVANTAGEOUS EFFECTS OF INVENTION

According to the present invention, the necessary abnormality detection can be arranged at a necessary place.

Drawings

Fig. 1 is a schematic diagram showing a main part of a control system according to the present embodiment.

Fig. 2 is a schematic diagram showing a configuration example of the control system of the present embodiment.

Fig. 3 is a block diagram showing an example of a hardware configuration of a control unit constituting the control device of the present embodiment.

Fig. 4 is a block diagram showing an example of a hardware configuration of an auxiliary processing unit constituting the control device of the present embodiment.

Fig. 5 is a block diagram showing an example of the hardware configuration of the support device used in the control system of the present embodiment.

Fig. 6 is a block diagram showing an example of a configuration for realizing the abnormality detection function in the control system of the present embodiment.

Fig. 7 is a timing chart for explaining an execution cycle of control arithmetic in the control device of the present embodiment.

Fig. 8 is a schematic diagram showing an example of mounting of the abnormality detection function in the control system of the present embodiment.

Fig. 9 is a schematic diagram showing another example of the installation of the abnormality detection function in the control system of the present embodiment.

Fig. 10 is a schematic diagram showing another example of the installation of the abnormality detection function in the control system of the present embodiment.

Fig. 11 is a flowchart showing an example of a system design process in the control system of the present embodiment.

Fig. 12 is a diagram for explaining an example of pointers for determining an example of mounting of the abnormality detection function in the control system according to the present embodiment.

Fig. 13 is a diagram showing an example of a user interface screen provided by the support device of the control system according to the present embodiment.

Fig. 14 is a diagram showing an example of a user interface screen provided by the support device of the control system according to the present embodiment.

Detailed Description

Embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the drawings, the same or corresponding portions are denoted by the same reference numerals, and description thereof will not be repeated.

Application example >

First, an example of a scenario to which the present invention is applied will be described.

An example of a functional configuration of a control system capable of executing the abnormality detection processing of the present embodiment will be described.

Fig. 1 is a schematic diagram showing a main part of a control system 1 according to the present embodiment. Referring to fig. 1, a control system 1 for controlling a control target includes a plurality of processing resources 30-1 to 30-6 capable of executing arithmetic processing, as an example.

In the present specification, a "processing resource" is a term including a main body capable of independently executing arbitrary arithmetic processing. For example, a control device such as a PLC (programmable logic controller) may be a "processing resource", and each processing unit (control unit 100, auxiliary processing unit 200, and the like described later) constituting the control device may be a "processing resource". A "processing resource" typically includes a memory for expanding commands specified in a program, and a processor that sequentially executes the commands expanded on the memory.

The plurality of processing resources 30-1 to 30-6 included in the control system 1 have a function of detecting some abnormality that may occur in any detection object included in the control object.

In the present specification, "abnormality detection" or "abnormality detection function" typically includes a case where "different from normal" or "different from normal" is detected in a detection target. The plurality of processing resources 30-1 to 30-6 are typically provided with a function of determining the presence or absence of an abnormality based on a state value collected from a control target in each control cycle.

Specifically, the control system 1 is provided with: a state value collection process 32 (a process including a state value collection 180 described later) for collecting 1 or more state values corresponding to arbitrary detection objects included in the control object; and an abnormality detection process 34 (including a process of score calculation 182 and determination 184 described later) for calculating a value (typically, a "score" described later) indicating the possibility of occurrence of an abnormality in the detection target based on the feature amount calculated from the collected 1 or more state values.

In the present specification, the term "state value" is a term including a value that can be observed in an arbitrary control target (including a detection target), and may include, for example, a physical value that can be measured by an arbitrary sensor, a state value indicating on/off of a relay, a switch, or the like, a command value such as a position, a speed, and a torque given to a servo driver by a PLC, a variable value used by the PLC for calculation, and the like. Physical values that can be measured by an arbitrary sensor, state values indicating on/off of a relay, a switch, or the like may also be said to be "raw data".

Fig. 1 (a) shows an example in which the state value collection processing 32 and the abnormality detection processing 34 are disposed in the processing resource 30-1. Fig. 1 (B) shows an example in which the state value collection process 32 is disposed in the processing resource 30-1, and the abnormality detection process 34 is disposed in the processing resource 30-5. In the example shown in FIG. 1 (B), the state values collected by the state value collection process 32 of the processing resource 30-1 are sent to the processing resource 30-5.

In this way, in the present embodiment, the state value collection process 32 and the abnormality detection process 34 can be disposed in any of the plurality of processing resources 30-1 to 30-6.

With such a configuration, in the control system 1 of the present embodiment, it is possible to arrange necessary abnormality detection at a necessary location using a plurality of processing resources included in the control system 1.

Structural example of control System

First, a configuration example of the control system 1 of the present embodiment will be explained.

(b 1: monolithic Structure)

Fig. 2 is a schematic diagram showing a configuration example of the control system 1 according to the present embodiment. Referring to fig. 2, the control system 1 is configured to control a control target, and includes a plurality of control devices 10-1, 10-2, and 10-3 (hereinafter, also collectively referred to as "control devices 10") as processing resources.

The control device 10 is capable of executing arithmetic processing to periodically execute control arithmetic for controlling the equipment and the machine. The control device 10 collects, as input values, state values of a manufacturing apparatus and a production line, which are objects to be controlled, and respective sensing devices (hereinafter, also collectively referred to as "on-site"), and outputs a command value (hereinafter, also referred to as an "output value") calculated by a control operation based on the input values to the manufacturing apparatus and the production line, which are objects to be controlled, and respective actuators.

Typically, the input values and output values exchanged between the control device 10 and the site are transmitted via the site network 4 and the network 6. As the field networks 4 and 6, a network that performs constant cycle communication that guarantees the arrival time of data is preferably used. As a network for performing such constant cycle communication, EtherCAT (registered trademark) and the like are known.

In the configuration example shown in fig. 2, the control system 1 includes a remote I/O device 12 connected to the control device 10-1 via the field network 4 and a remote I/O device 12 connected to the control device 10-2 via the field network 6. To the remote I/O device 12, 1 or more relays 14 for exchanging signals with the field are connected.

The control devices 10-1, 10-2, 10-3 are connected via the control network 8 in such a manner as to be able to communicate data with each other. Any data can be exchanged between the control devices 10-1, 10-2, 10-3.

Each control device 10 includes the control unit 100, and may be provided with an auxiliary processing unit 200 as needed. Hereinafter, a hardware configuration example of the support apparatus 300 will be described in addition to the control unit 100 and the auxiliary processing unit 200.

(b2 hardware configuration example of control Unit 100)

Fig. 3 is a block diagram showing an example of the hardware configuration of the control unit 100 constituting the control device 10 according to the present embodiment. Referring to fig. 3, the control Unit 100 includes, as main components, a processor 102 such as a CPU (Central Processing Unit) or a GPU (graphic Processing Unit), a chipset 104, a main memory 106, a secondary memory 108, a USB (Universal Serial Bus) controller 112, a memory card interface 114, network controllers 116 and 118, and an internal Bus controller 120.

The processor 102 reads various programs stored in the secondary storage device 108, expands and executes the programs in the main storage device 106, and thereby realizes control operations for standard control and various processes described below. The chipset 104 mediates data exchange between the processor 102 and each component, thereby realizing processing of the control unit 100 as a whole.

The main Memory device 106 is a volatile Memory device such as a DRAM (Dynamic Random Access Memory) or an SRAM (Static Random Access Memory). The secondary storage device 108 is configured by a nonvolatile storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive).

In addition to the system program, the secondary storage device 108 stores a control program that operates in an execution environment provided by the system program.

The USB controller 112 is responsible for data exchange with an arbitrary information processing apparatus via a USB connection.

The memory card interface 114 is configured to be able to attach and detach the memory card 115, and is capable of writing data such as a control program and various settings to the memory card 115, and reading data such as a control program and various settings from the memory card 115.

The network controller 116 mediates data exchange with other control apparatuses 10 via the control network 8 (see fig. 2). The network controller 118 mediates data exchange with any device such as the remote I/O device 12 (both refer to fig. 2) via the field networks 4 and 6.

The internal bus controller 120 mediates data exchange with other units constituting the control device 10 via the internal bus. The internal bus may use a communication protocol unique to a manufacturer, or may use a communication protocol identical to or conforming to an arbitrary industrial network protocol.

Fig. 3 shows an example of a configuration in which the processor 102 executes a program to provide a desired function, but a part or all of the provided functions may be installed using a dedicated hardware Circuit (e.g., an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array)). Alternatively, the main part of the control unit 100 may also be implemented using hardware following a general-purpose architecture (e.g., a general-purpose personal computer-based industrial personal computer). In this case, a plurality of OSs (Operating systems) having different applications may be executed in parallel using virtualization technology, and a desired application may be executed on each OS.

(b3 hardware configuration example of auxiliary processing Unit 200)

Fig. 4 is a block diagram showing an example of the hardware configuration of the auxiliary processing unit 200 constituting the control device 10 according to the present embodiment. Referring to fig. 4, the auxiliary processing unit 200 includes, as main components, a processor 202 such as a CPU or GPU, a chipset 204, a main storage device 206, a secondary storage device 208, a memory card interface 214, and an internal bus controller 220.

The processor 202 reads various programs stored in the secondary storage device 208, and develops and executes them in the main storage device 206, thereby realizing arbitrary arithmetic processing accompanying control arithmetic. The chipset 204 mediates data exchange between the processor 202 and the components, thereby realizing processing as the whole of the auxiliary processing unit 200.

In addition to the system program, the secondary storage device 208 stores a security program that operates in an execution environment provided by the system program.

The memory card interface 214 is configured to be attachable to and detachable from the memory card 215, and is capable of writing data such as security programs and various settings to the memory card 215, and reading data such as security programs and various settings from the memory card 215.

The internal bus controller 220 is responsible for data exchange with the control unit 100 via the internal bus.

In fig. 4, an example of a configuration in which a program is executed by the processor 202 to provide a desired function is shown, but a part or all of the provided functions may be installed using a dedicated hardware circuit (e.g., an ASIC or an FPGA). Alternatively, the main portion of the auxiliary processing unit 200 may be implemented using hardware conforming to a general-purpose architecture (e.g., a general-purpose personal computer-based industrial personal computer). In this case, a plurality of OSs having different purposes may be executed in parallel using virtualization technology, and a desired application may be executed on each OS.

(b4 hardware configuration example of the support apparatus 300)

Fig. 5 is a block diagram showing an example of the hardware configuration of the support apparatus 300 used in the control system 1 according to the present embodiment. The support apparatus 300 is realized by executing a program using hardware conforming to a general-purpose architecture (for example, a general-purpose personal computer), as an example.

Referring to fig. 5, the support apparatus 300 includes a processor 302 such as a CPU or MPU, a drive 304, a main storage 306, a secondary storage 308, a USB controller 312, a local area network controller 314, an input unit 316, and a display unit 318. These components are connected via a bus 320.

The processor 302 reads various programs stored in the secondary storage device 308, and develops and executes the programs in the main storage device 306, thereby realizing various processes described later.

The secondary storage device 308 is configured by, for example, an HDD (Hard Disk Drive), an SSD (Solid State Drive), or the like. The secondary storage device 308 stores a support program 340 for realizing the support device 300. More specifically, various programs are stored in the secondary storage device 308, and include: a development tool 350 for creating a user program executed in the support apparatus 300, debugging the created program, defining a system configuration, setting various parameters, and the like; and a setting tool 360. The support program 340 includes a development tool 350 and a setup tool 360. In the secondary storage device 308, an OS and other necessary programs may also be stored.

The drive 304 is capable of writing data to the storage medium 305 and reading various data (user program, trace data, time-series data, or the like) from the storage medium 305. The storage medium 305 includes, for example, a storage medium 305 (for example, an optical storage medium such as a DVD (Digital Versatile Disc)) which stores a computer-readable program (typically, the support program 340) in a non-transitory manner.

Various programs (typically, the support program 340) executed by the support apparatus 300 and the like may be installed via the computer-readable storage medium 305, but may be installed in a form downloaded from a server apparatus or the like on a network. The functions provided by the support apparatus 300 are sometimes implemented in the form of utilizing a part of the modules provided by the OS.

The USB controller 312 mediates data exchange with the control unit 100 or the auxiliary processing unit 200 via a USB connection. The local network controller 314 controls data exchange with other devices via an arbitrary network.

The input unit 316 is configured by a keyboard, a mouse, and the like, and receives a user operation. The display unit 318 is configured by a display, various indicators, and the like, and outputs a processing result from the processor 302. A printer may be connected to the support apparatus 300.

In fig. 5, an example of a configuration in which a program is executed by the processor 302 to provide a desired function is shown, but a part or all of the provided functions may be installed using a dedicated hardware circuit (for example, ASIC or FPGA).

< C. Exception detection function >

Next, the abnormality detection function incorporated in the control system 1 of the present embodiment will be described.

Fig. 6 is a block diagram showing a configuration example for realizing the abnormality detection function in the control system 1 of the present embodiment. Referring to fig. 6, the control device 10 includes a variable management unit 160, a feature extraction unit 140, a machine learning processing unit 144, and a result determination unit 146 as an abnormality detection function.

The variable management unit 160 collects state values (input values) appearing in the controlled object at each predetermined control cycle, and updates the value of the plant variable 162, which is an internal state value. The present invention is not limited to the method of referring to a value using a "variable", and can be applied to a method of directly designating and referring to a physical address of a memory storing each value, or the like.

The feature extraction unit 140 calculates 1 or more feature amounts 150 from 1 or more state values (input values) corresponding to the detection target. More specifically, the feature extraction unit 140 calculates, according to the preset setting information 158, 1 or more feature quantities 150 (for example, an average value, a maximum value, a minimum value, and the like over a predetermined time) periodically or for each event according to a predetermined calculation procedure based on a value (or a temporal change in value in a unit interval) indicated by the designated 1 or more device variables 162 (state values). The unit section used by feature extraction section 140 to calculate feature amount 150 is also sometimes referred to as a "frame". The unit section (frame) is arbitrarily set according to the operation of the detection target and the like.

The machine learning processing unit 144 refers to the learning model 152, and calculates a value score 154 indicating the possibility of occurrence of some abnormality in the detection target based on the 1 or more feature amounts 150 calculated by the feature extraction unit 140.

As an example, as an algorithm for abnormality detection, the machine learning processing unit 144 may employ a method of calculating a score corresponding to an input value based on the degree of deviation of the input value from a value group in a hyperspace. As a method of detecting an abnormality based on the degree of deviation, a method of detecting an abnormality based on the shortest distance from each point to a value group (k-neighborhood method), a local deviation factor (LoF) method of evaluating a distance by a cluster including a value group, an iForest (isolation forest) method using a score calculated from a path length, and the like are known.

In the case where the learning model 152 is configured by the feature amount in the normal state, the greater the degree of deviation (i.e., score) from the learning model 152, the higher the possibility that some abnormality has occurred in the detection target can be determined. On the other hand, in the case where the learning model 152 is configured by the feature amount at the time of abnormality, the smaller the degree of deviation (i.e., score) from the learning model 152, the higher the possibility that some abnormality has occurred in the detection target can be determined. The learning model 152 can be determined by a known data mining method.

The result determination unit 146 generates a determination result 170 indicating whether or not some abnormality has occurred in the detection target, based on the score 154 calculated by the machine learning processing unit 144. The determination condition 156 may be set in advance by the support apparatus 300. Typically, the determination condition 156 includes a threshold range or the like set for the score 154 and indicating that there is a high possibility that some abnormality occurs in the detection target. The determination result 170 may be any one of "OK" which is a state in which no abnormality occurs in the detection target, and "NG" which is a state in which some abnormality occurs in the detection target.

With the above configuration, it is possible to detect the occurrence of some abnormality that may occur with respect to any detection target included in the control target. The feature extraction unit 140, the machine learning processing unit 144, and the result determination unit 146 may be formed as a library. In this case, by setting the setting information 158, the learning model 152, and the determination condition 156 in the library, even a user who lacks expert knowledge can easily use the abnormality detection function.

< D. ensuring of processing resources and topic

Next, the cycle execution of the control operation in the control device 10 (control unit 100) will be described. Fig. 7 is a timing chart for explaining an execution cycle of control calculation in the control device 10 according to the present embodiment. Fig. 7 (a) shows an example of the abnormality detection processing executed by the control device 10 having a surplus of processing resources. As shown in fig. 7 a, the control device 10 executes a control task 50 (control calculation) for each predetermined control cycle. If the time required for the execution of the control tasks 50 is sufficiently short with respect to the length of the control cycle, the total time thereof may become equal to or less than the length of the control cycle even if the abnormality detection task 52 (abnormality detection processing) is further executed. In the execution state shown in fig. 7 (a), both the control task 50 and the abnormality detection task 52 can be executed for each control cycle.

In contrast, fig. 7 (B) shows an example in which the control device 10 having no surplus of processing resources executes the abnormality detection processing. As shown in fig. 7B, if the time required for the execution of the control task 50 is relatively long with respect to the length of the control cycle, the total time thereof may exceed the length of the control cycle when the abnormality detection task 52 (abnormality detection processing) is further executed. In the execution state shown in fig. 7 (B), both the control task 50 and the abnormality detection task 52 cannot be executed in each control cycle.

In this way, when the abnormality detection function is introduced for the purpose of preventive maintenance or the like, the processing resources required for executing the abnormality detection are relatively large, and therefore, there is a possibility that the control calculation cannot be executed at a constant cycle. In particular, it is sometimes difficult to introduce an abnormality detection function into a control target that requires high responsiveness or the like.

(E. dispersed configuration)

As described above, in an environment where sufficient processing resources cannot be secured, it may be difficult to introduce the abnormality detection function. The control system 1 of the present embodiment can provide a solution to such a problem.

Fig. 8 is a schematic diagram showing an example of mounting of the abnormality detection function in the control system 1 according to the present embodiment. Fig. 8 shows an example in which the abnormality detection function is incorporated in the control devices 10-1 and 10-2.

More specifically, each of the control devices 10-1 and 10-2 performs the processing of the state value collection 180, the score calculation 182, and the determination 184. The state value collection 180 corresponds to a process in which the feature extraction unit 140 (see fig. 6) collects 1 or more state values (input values) corresponding to the detection target. The score calculation 182 corresponds to the process of calculating the score 154 by the feature extraction unit 140 and the machine learning processing unit 144 (both refer to fig. 6). The judgment 184 corresponds to a process of generating the judgment result 170 indicating whether or not some abnormality has occurred in the detection target by the result judgment unit 146 (see fig. 6).

The state value collection process 32 (see fig. 1) includes a process of state value collection 180, and the state value collection 180 is used to collect 1 or more state values corresponding to arbitrary detection objects included in the control object. The abnormality detection process 34 (see fig. 1) includes: a process of calculating a score 182 for calculating a value (score) indicating a possibility of occurrence of an abnormality in the detection target based on the feature amount calculated from the collected 1 or more state values; and a process of determining 184 for generating a determination result 170 indicating whether or not an abnormality has occurred in the detection target, based on the value (score) indicating the possibility of occurrence of an abnormality in the detection target.

In the installation example shown in fig. 8, the abnormality detection function is executed independently of each other in each of the control devices 10-1 and 10-2. The control devices 10-1 and 10-2 can also output the calculated determination result and score to the other control device 10.

Fig. 9 is a schematic diagram showing another example of the installation of the abnormality detection function in the control system 1 according to the present embodiment. Fig. 9 shows an example in which the abnormality detection function is installed by the cooperation of the control devices 10-2 and 10-3.

More specifically, the control device 10-2 executes the process of collecting the state value 180, and collects 1 or more state values corresponding to the detection target. The collected state values (raw data) are transmitted to the control device 10-3. In control device 10-3, the processes of score calculation 182 and determination 184 are executed based on the state value from control device 10-2. The determination result and the score calculated as the results of these processes are returned from the control device 10-3 to the control device 10-2. The control device 10-2 can detect that some abnormality has occurred in the detection target based on the determination result and the score from the control device 10-3, and can execute necessary processing based on the detection result.

As shown in fig. 9, when the state value collection processing 32 (state value collection 180) and the abnormality detection processing 34 (score calculation 182 and determination 184) are disposed in different processing resources (control device 10-2), the state value collection processing 32 (state value collection 180) may transmit the collected 1 or more state values to the processing resources (control device 10-3) in which the abnormality detection processing 34 (score calculation 182 and determination 184) is disposed.

In addition, when the state value collection processing 32 (state value collection 180) and the abnormality detection processing 34 (score calculation 182 and determination 184) are disposed in different processing resources (control device 10-2), the abnormality detection processing 34 (score calculation 182 and determination 184) may transmit the generated determination result 170 to the processing resource (control device 10-3) in which the state value collection processing 32 (state value collection 180) is disposed. At this time, the abnormality detection process 34 (score calculation 182 and determination 184) may transmit the value (score 154) indicating the calculated possibility of occurrence of an abnormality in the detection target to the processing resource (control device 10-3) in which the state value collection process 32 (state value collection 180) is arranged.

In the configuration example shown in fig. 9, since the control device 10-2 executes only the processing of the state value collection 180, the abnormality detection function can be installed even in a situation where there is no excess of processing resources as shown in fig. 7 (B).

As shown in fig. 9, the abnormality detection function can be installed by cooperating a plurality of control apparatuses 10, and for example, even a control apparatus 10 that does not have a library that provides abnormality detection can easily utilize the abnormality detection function. Further, since it is sufficient to collect 1 or more state values corresponding to the detection target and transmit the state values to the other control apparatuses 10, it is possible to minimize changes to the user program or the like executed by the control apparatuses 10.

Fig. 9 shows an example of a configuration in which 2 control devices 10 cooperate to realize the abnormality detection function, but the present invention is not limited to this, and 3 or more control devices 10 may cooperate to realize the abnormality detection function.

Fig. 8 and 9 show an example of a configuration in which a plurality of control devices 10 cooperate to realize an abnormality detection function, but a single control device 10 may have a plurality of processing resources independent of each other. In such a case, the plurality of processing resources may cooperate to implement the anomaly detection function.

Fig. 10 is a schematic diagram showing still another example of the installation of the abnormality detection function in the control system 1 according to the present embodiment. Fig. 10 (a) and (B) show a control device 10 configured by a control unit 100, an auxiliary processing unit 200, and 1 or more I/O units 250. The control unit 100 and the auxiliary processing unit 200 each have a processor capable of executing processing for realizing an abnormality detection function.

The control unit 100 and the I/O unit 250 are connected via the internal bus 122, and the control unit 100 can collect any state value (input value) collected by the I/O unit 250.

Fig. 10 (a) shows an example in which the abnormality detection function is incorporated in the control unit 100 of the control device 10. More specifically, control section 100 executes the processing of state value collection 180, score calculation 182, and determination 184.

In contrast, fig. 10 (B) shows an example in which the abnormality detection function is installed by the control unit 100 and the auxiliary processing unit 200 of the control device 10 in cooperation. More specifically, in the control unit 100, the process of the state value collection 180 is performed to collect 1 or more state values corresponding to the detection object. The collected state values (raw data) are sent to the auxiliary processing unit 200. In the auxiliary processing unit 200, the processes of score calculation 182 and determination 184 are executed based on the state value from the control unit 100. The determination result and the score calculated as the results of these processes are returned from the auxiliary processing unit 200 to the control unit 100. Control section 100 can detect that some abnormality has occurred in the detection target based on the determination result and the score from auxiliary processing section 200, and can execute necessary processing based on the detection result.

As shown in fig. 8 to 10, the control system 1 according to the present embodiment can realize the abnormality detection function using a single processing resource, and can also realize the abnormality detection function using a plurality of processing resources. That is, the state value collection 180, the score calculation 182, and the determination 184 can be arranged in any of the plurality of processing resources.

By adopting such a flexible installation method, it is possible to provide a necessary abnormality detection function to the necessary control device 10 without greatly changing the existing mechanism (including the user program) for controlling the control target. By adopting such an installation method, the influence on the execution of the existing user program can be limited.

In addition, the abnormality detection function can be realized by allocating a necessary function to an arbitrary processing resource according to the degree of redundancy or the like without performing special system design or the like in order to install the abnormality detection function.

< F. System design Process >

Next, a system design procedure when the above-described abnormality detection function is arranged will be described.

Fig. 11 is a flowchart showing an example of a system design process in the control system 1 according to the present embodiment. The processing shown in fig. 11 is basically performed in the support apparatus 300. That is, the support apparatus 300 assists in determining the processing resources to be the placement destinations of the state value collection processing 32 (state value collection 180) and the abnormality detection processing 34 (score calculation 182 and determination 184).

Referring to fig. 11, the user creates a user program for controlling a control object, which is executed by an arbitrary control device 10, in accordance with a predetermined design specification (step S1). When the creation of the user program is completed, the user sets a necessary abnormality detection function for the control device 10 (step S2). Next, when the user instructs to execute the simulation, the support apparatus 300 executes the simulation based on the model information of the control apparatus 10, the content of the created user program, the set abnormality detection function, and the like, and calculates the time required for the execution of the control task 50 and the abnormality detection task 52 (step S3). The calculated time is prompted to the user.

The user confirms the presented time required for execution and specifies whether or not the current setting is applied (step S4). When the application of the current setting is designated (yes in step S4), the support apparatus 300 transfers the setting data, the user program, and the like corresponding to the current setting to the target control apparatus 10 (step S5). Then, the process ends.

When the change of the current setting is designated (no in step S4), the support apparatus 300 creates a candidate for the distributed configuration abnormality detection function and presents it to the user (step S6). When the user selects an arbitrary candidate (step S7), the support apparatus 300 transmits the setting data and the user program and the like corresponding to the selected candidate to the target one or more control apparatuses 10 (step S8). Then, the process ends.

As described above, the control system 1 according to the present embodiment provides a function for assisting the design for installing the abnormality detection function.

When the time required for the execution of the control task 50 and the abnormality detection task 52 exceeds the control cycle, the control device 10, the control unit 100, the auxiliary processing unit 200, and the like that are responsible for the abnormality detection function are determined in consideration of the following factors.

Number of state values required for anomaly detection

Collection destination (device) of status values required for abnormality detection

Specification of the control device 10 (control unit 100 and auxiliary processing unit 200)

Load factor of internal bus of control device 10

Load factor of field network connected to the control device 10

Load factor of control network connected to the control device 10

In this way, the support apparatus 300 determines the arrangement destinations of the state value collection processing 32 (state value collection 180) and the abnormality detection processing 34 (score calculation 182 and determination 184) based on at least one of the number of state values to be collected, the collection destinations of the state values, the specifications of the plurality of processing resources, the load ratios of the internal buses, and the load ratios of the networks.

Basically, it is preferable that the processing resource responsible for the abnormality detection function preferentially adopts a processing resource located at a position closer to the generation source of 1 or more state values (input values) corresponding to the detection target. Further, in the case where the processing resource has a plurality of cores, the core responsible for the constant cycle processing may be preferentially used. Only when these cannot be employed, the processing resources may be used via the network.

For example, the processing of a state value having a relatively long task cycle among 1 or more state values (input values) corresponding to the detection target may be handed over to another control device 10. Further, the processing of the state value having a low priority may be transferred to another control device 10 based on the priority set by the user in advance.

Fig. 12 is a diagram for explaining an example of pointers for determining an example of mounting of the abnormality detection function in the control system 1 according to the present embodiment. Referring to fig. 12, there are two ideas, quality and robustness, as guidelines. The priority of quality is important to ensure speed and periodicity, and the priority of robustness is important to ensure stability.

From the viewpoint of the control cycle, the emphasis on quality can be performed with a shorter control cycle than the case where the emphasis on robustness is placed. In addition, from the viewpoint of data volume, robustness is emphasized to handle a smaller data volume than the case where quality is emphasized.

The user may first select which of the quality and the robustness is to be emphasized in the user interface screen or the like provided by the support apparatus 300.

For example, when the importance level is selected, it is preferable to preferentially select a processing resource capable of collecting a state value via an internal bus. In addition, it is preferable to select the processing resource that is responsible for the control task so as to also be responsible for the abnormality detection task. By executing the control task and the abnormality detection task with the same processing resource, it is possible to reliably collect necessary information in the same control cycle.

On the other hand, when importance is placed on robustness, the execution cycle of the abnormality detection task may be relatively long, but the abnormality detection processing can be reliably executed even if the target data amount is large. When importance is attached to robustness, it is also possible to select a processing resource connected via a control network or the like. However, since a large amount of data is transmitted over a network, a prior investigation or the like is also required for a network load or the like. In addition, the user may be notified of the increase in the network load and may be configured to receive a clear permission from the user.

In addition, when importance is attached to robustness, processing resources other than the processing resource responsible for the control task may be used. The control system as a whole is preferably configured such that the processing load is uniformly distributed.

As will be exemplified below, the support apparatus 300 provides a user interface for assisting the determination of the processing resources to be the placement destinations of the state value collection processing 32 (state value collection 180) and the abnormality detection processing 34 (score calculation 182 and determination 184).

Fig. 13 is a diagram showing an example of a user interface screen 400 provided by the support device 300 of the control system 1 according to the present embodiment.

Referring to fig. 13, the user interface screen 400 includes information indicating a resource monitor indicating the status of a processing resource in the control system 1 to be set and a control cycle of the processing resource. More specifically, the system configuration 410 of the control system 1 to be set and the usage rates 412, 414, and 416 corresponding to the load rates of the respective processing resources included in the respective control devices 10 are displayed. The usage rates 412, 414, and 416 may be values calculated by simulation, or may be actual measurement values when the system can be connected to a real control system.

Here, an icon 418 indicating an abnormality detection function preset by the user is also displayed on the user interface screen 400. In the example shown in fig. 13, a case is shown in which it is executed by the control unit of "PLC 2".

The execution time of each task is shown in accordance with the setting of the processing resource for executing the abnormality detection function. Specifically, the user interface screen 400 includes tasks to be executed by the respective control units of the respective control apparatuses, and an execution time display 420 indicating a time required for execution of the respective tasks. The display content of the execution time display 420 may be a value calculated by simulation, or may be an actual measurement value when the control system can be connected to a real control system.

The user can appropriately adjust the processing resources responsible for the abnormality detection function while referring to the user interface screen 400 shown in fig. 13. However, even for users lacking expert knowledge, the support apparatus 300 can present appropriate "recommended" settings so that the abnormality detection function can be appropriately installed.

Fig. 14 is a diagram showing an example of a user interface screen 402 provided by the support device 300 of the control system 1 according to the present embodiment.

Referring to fig. 14, on user interface screen 402, compared with the state shown in fig. 13, "recommended setting" is displayed. In the user interface screen 402 shown in fig. 14, an icon 418 indicating an abnormality detection function corresponds to "PLC 3", which indicates that the control unit 100 of "PLC 3" executes a process of abnormality detection.

As the processing resources responsible for the abnormality detection processing are changed, the display content is updated for the execution time display 420.

An "OK" button 430 or an "NG" button 432 is displayed on the lower portion of the user interface screen 402. When the user selects the button 430 of "OK", the support apparatus 300 reflects the display contents of the user interface screen 402. More specifically, the support apparatus 300 generates an execution file or various settings to be written to the target control apparatus 10 (the control unit 100 and the auxiliary processing unit 200) in accordance with the reflected content, and sequentially writes the files. The setting information 158, the learning model 152, and the determination condition 156 (all refer to fig. 6) are also transmitted from the support apparatus 300 to the target control apparatus 10. When the capacity of the data to be transmitted is large, the optimum transmission method may be determined in consideration of the load factor of the network.

In this way, the support apparatus 300 transmits necessary data to the processing resources of the arrangement state value collection processing 32 (state value collection 180) and the abnormality detection processing 34 (score calculation 182 and determination 184).

In this case, the variable transmitted to another control device 10 or the like via the control network or the like may be automatically set as the public variable.

The user may be presented with the setting contents automatically determined in consideration of the above-described factors, and the user may further change the presented setting contents.

Further, if the importance-added quality is selected, when it is set that the state value is transmitted from a certain control device 10 to another control device 10, it is preferable that the return determination result 170 and the score 154 to the control device 10 of the transmission source are set as default. On the other hand, if importance is placed on robustness, it is preferable to transmit the determination result 170 and the score 154 to the control device 10 located on the upper side as default. However, the setting may be arbitrarily selected by the user.

Further, there is a need to detect an abnormality that may occur in a control target using a combination of a plurality of determination results 170. In such a case, the cycles of the plurality of abnormality detection tasks may be adjusted to match each other.

Further, it is also possible to confirm the state of each processing resource after setting the real control system 1 in accordance with the setting determined by the simulation. The reset may be performed when the state of each processing resource in the actual control system 1 is different from the state calculated by the simulation.

< G, attached notes >

The present embodiment described above includes the following technical ideas.

[ Structure 1]

A control system (1) for controlling a control object, wherein,

the control system (1) has a plurality of processing resources (10; 100, 200) capable of performing arithmetic processing,

the plurality of processing resources have:

a collection unit (32; 180) that collects 1 or more state values corresponding to arbitrary detection objects included in the control object; and

abnormality detection means (34; 184, 186) for calculating a value (154) indicating a possibility of occurrence of an abnormality in the detection target on the basis of the feature amount calculated from the collected 1 or more state values,

the collection unit and the abnormality detection unit may be respectively configured to any processing resource among the plurality of processing resources.

[ Structure 2]

In the control system described in the structure 1,

the abnormality detection means generates a determination result (170) indicating whether or not an abnormality has occurred in the detection target, based on a value indicating the possibility of occurrence of an abnormality in the detection target.

[ Structure 3]

In the control system described in the structure 2,

when the collection unit and the abnormality detection unit are disposed in different processing resources, the abnormality detection unit transmits the generated determination result to the processing resource in which the collection unit is disposed.

[ Structure 4]

In the control system according to any one of configurations 1 to 3,

in a case where the collection unit and the abnormality detection unit are configured in different processing resources, the collection unit transmits the collected 1 or more state values to the processing resource in which the abnormality detection unit is configured.

[ Structure 5]

In the control system according to any one of configurations 1 to 4,

when the collection means and the abnormality detection means are disposed in different processing resources, the abnormality detection means transmits the calculated value indicating the possibility of occurrence of an abnormality in the detection target to the processing resource in which the collection means is disposed.

[ Structure 6]

In the control system according to any one of configurations 1 to 5,

the control system further has a support device (300), and the support device (300) is configured to assist in the determination of the processing resources to be the allocation destinations of the collection unit and the abnormality detection unit.

[ Structure 7]

In the control system according to the structure 6,

the support device determines the arrangement destinations of the collection unit and the abnormality detection unit based on at least one of the number of state values to be collected by the collection unit, the collection destination of the state values, the specifications of the plurality of processing resources, the load factor of the internal bus, and the load factor of the network.

[ Structure 8]

In the control system described in the structure 6 or 7,

the supporting apparatus transmits required data to a processing resource in which the collecting unit and the abnormality detecting unit should be configured.

[ Structure 9]

A support device (300) used in a control system (1) for controlling a control object, wherein,

the control system has a plurality of processing resources (10; 100, 200) capable of performing arithmetic processing,

the plurality of processing resources have:

a collection unit (32; 180) that collects 1 or more state values corresponding to arbitrary detection objects included in the control object; and

abnormality detection means (34; 184, 186) for calculating a value (154) indicating a possibility of occurrence of an abnormality in the detection target on the basis of the feature amount calculated from the collected 1 or more state values,

the support apparatus provides a user interface (400, 402), the user interface (400, 402) being used for assisting the decision of the processing resource to be the configuration destination of the collection unit and the abnormality detection unit.

[ Structure 10]

A support program (340) for implementing a support apparatus (300), the support apparatus (300) being used in a control system (1) for controlling a control object, wherein,

the control system has a plurality of processing resources (10; 100, 200) capable of performing arithmetic processing,

the plurality of processing resources have:

a collection unit (32; 180) that collects 1 or more state values corresponding to arbitrary detection objects included in the control object; and

abnormality detection means (34; 184, 186) for calculating a value (154) indicating a possibility of occurrence of an abnormality in the detection target on the basis of the feature amount calculated from the collected 1 or more state values,

the support program causes a computer to realize a function of providing a user interface (400, 402), the user interface (400, 402) being used to assist in decision of a processing resource that becomes a configuration destination of the collection unit and the abnormality detection unit.

< H. advantage >

In the control system 1 of the present embodiment, the abnormality detection function can be installed in an appropriate processing resource in accordance with the degree of redundancy of each processing resource (including the load factor of the network). Therefore, the required abnormality detection function can be installed at a required position without interfering with execution of a user program for controlling a control target. In addition, since the abnormality detection function can be installed in the redundant processing resources, it is possible to avoid a situation in which the task execution exceeds the control cycle. That is, by installing the abnormality detection function at an appropriate position, the constant periodicity of task execution in each processing resource can be maintained as much as possible.

In the control system 1 of the present embodiment, since the abnormality detection function can be installed at an appropriate position, it is not necessary to increase the specification of the specific control device 10 (the control unit 100 and the auxiliary processing unit 200), and the setting restriction of the processing resources at the time of the initial setting can be reduced. Further, even when the abnormality detection function is installed in the conventional control system 1, it is possible to use a processing resource that is redundant, and therefore, it is not necessary to newly introduce a processing resource or replace it with a processing resource of a higher specification. As a result, the degree of freedom in design of the user can be improved.

In the control system 1 of the present embodiment, the support device 300 automatically suggests a setting suitable for the installation of the abnormality detection function, and therefore, even a user lacking expert knowledge can install an arbitrary abnormality detection function.

The embodiments disclosed herein are to be considered in all respects as illustrative and not restrictive. The scope of the present invention is defined by the claims, not by the description above, and is intended to include all modifications equivalent in meaning and scope to the claims.

Description of the reference symbols

1: a control system; 2: a PLC; 4. 6: a field network; 8: a control network; 10: a control device; 12: a remote I/O device; 14: a relay; 30: processing resources; 32: collecting and processing state values; 34: abnormality detection processing; 50: controlling a task; 52: an anomaly detection task; 100: a control unit; 102. 202, 302: a processor; 104. 204: a chipset; 106. 206, 306: a main storage device; 108. 208, 308: a secondary storage device; 112. 312: a USB controller; 114. 214: a memory card interface; 115. 215: a memory card; 116. 118: a network controller; 120. 220, and (2) a step of: an internal bus controller; 122: an internal bus; 140: a feature extraction unit; 144: a machine learning processing unit; 146: a result determination unit; 150: a characteristic amount; 152: learning a model; 154: scoring; 156: judging a condition; 158: setting information; 160: a variable management unit; 162: a device variable; 170: judging a result; 180: collecting a state value; 182: calculating a score; 184: judging; 200: an auxiliary processing unit; 250: an I/O unit; 300: a support device; 304: a driver; 305: a storage medium; 314: a local area network controller; 316: an input section; 318: a display unit; 320: a bus; 340: a support program; 350: a development tool; 360: setting a tool; 400. 402, a step of: a user interface screen; 410: a system architecture; 412. 414, 416: the utilization rate; 418: an icon; 420: displaying the execution time; 430. 432: a button.