Failure factor analysis device and failure factor analysis method

Abstract

To efficiently perform an automatic collection of information required for a failure factor analysis, and minimize dispatch to an actual place for the information collection as much as possible.SOLUTION: A failure factor analysis device 200 comprises: a failure knowledge database 400 in which knowledge about a failure of an appliance is stored; a collection information determination unit 300 that determines a content of information to be collected from the appliance based on an error code and information stored in the failure knowledge database 400; a collection information evaluation unit 500 that evaluates information collected from the appliance; and a failure mode estimation unit 600 that estimates a failure mode of the appliance from the evaluation result and information in the failure knowledge database 400. The collection information determination unit 300 enumerates inspection items required to determine the failure mode associated with the error code, and determines to collect measurement data and image information of a measurement value with required and sufficient accuracy to determine the inspection items.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a failure factor analysis device and a failure factor analysis method for analyzing the failure factors of equipment using signals obtained from an object to be analyzed. [Background technology]

[0002] In some cases, such as hydroelectric power plants built in mountainous areas, the plant's operating status is monitored by monitoring equipment in a monitoring room installed in a separate location from the plant. The amount of data communication between the plant and the monitoring equipment in the monitoring room is often limited. For this reason, when a failure occurs, the information transmitted from the plant to the monitoring room may include data reporting the occurrence of the failure, but may not include operational data or detailed information. In such cases, operators must be dispatched to the site to collect information needed to identify the cause of the failure, which requires labor and cost. Therefore, there is a need to minimize the number of on-site dispatches required for information collection. Patent Documents 1 and 2 disclose technologies for supporting plant information collection.

[0003] Patent Document 1 discloses a technology that collects and displays only the data that maintenance personnel need during maintenance work, enabling the maintenance personnel to quickly and accurately investigate and analyze elevator maintenance. According to this technology, an event identification data collection means collects error code data regarding door failures from CPU1 to CPUN of an elevator control device. An event-related data name designation means refers to an event-related data name storage means and designates a data name related to the door failure. The event-related data collection means collects only this designated event-related data from CPU1 to CPUN, and an event-related data display control means controls the display of this collected event-related data.

[0004] Patent Document 2 discloses an inspection support system that can automatically and accurately acquire the indicated value of an indicating instrument from an image of the indicating instrument. According to this technology, in the inspection support system, a readability determination unit evaluates an image of the indicating instrument captured by a surveillance camera to determine whether the indicated value of the indicating instrument can be automatically read. If the readability determination unit determines that the image is readable, the image reading unit reads the indicated value of the indicating instrument from the image of the indicating instrument. If the readability determination unit determines that the image of the indicating instrument captured by the surveillance camera under first imaging conditions is unreadable, an imaging control unit causes the surveillance camera to capture an image of the indicating instrument under second imaging conditions different from the first imaging conditions. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2003-73051 [Patent Document 2] Japanese Patent Publication No. 2022-185364 Summary of the Invention [Problem to be solved by the invention]

[0006] By utilizing the technology described in Patent Document 1, it is possible to identify data related to failure events and support information collection. In addition, by utilizing the technology described in Patent Document 2, it is possible to read indicated values ​​from images of instruments.

[0007] When checking measurement values ​​from video footage, it is sufficient to capture the video with sufficient accuracy for failure cause analysis. In other words, there are cases where it is sufficient to know whether the measurement value exceeds the threshold, and cases where specific numerical values ​​are required. In the former case, even if the camera is foggy or out of focus and the exact value cannot be read, as long as it is clear whether the measurement value exceeds the threshold, it is sufficient for failure cause analysis.

[0008] The present invention has been made in view of the above circumstances, and has as its object to efficiently and automatically collect information for failure cause analysis and to reduce as much as possible the need for on-site dispatch for information collection. [Means for solving the problem]

[0009] To solve the above problems, one aspect of the present invention provides a failure factor analysis device that analyzes the causes of failures occurring in equipment based on error codes generated from equipment measurement data and outputs information about the analysis results. This failure factor analysis device includes a failure knowledge database that stores knowledge about equipment failures, a collected information determination unit that determines the content of information to be collected from the equipment based on the error codes and the information stored in the failure knowledge database, a collected information evaluation unit that evaluates the information collected from the equipment, and a failure mode estimation unit that estimates the equipment failure mode based on the evaluation results by the collected information evaluation unit and the information stored in the failure knowledge database. The collected information determination unit lists the inspection items required to determine the failure mode associated with the error codes and determines to collect measurement data and image information of the measurement values ​​with the accuracy necessary and sufficient for determining the inspection items. [Effects of the Invention]

[0010] According to at least one aspect of the present invention, it is possible to efficiently and automatically collect information necessary and with sufficient accuracy for failure cause analysis. As a result, in this embodiment, it is possible to minimize the need for on-site dispatches to collect information for failure cause analysis. Problems, configurations, and effects other than those described above will become apparent from the following description of the preferred embodiments of the invention. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram showing an example of the configuration of a failure factor analysis device according to an embodiment of the present invention; [Figure 2] 1 is a block diagram showing an example of the hardware configuration of a computer included in a failure cause analysis device according to a first embodiment of the present invention. [Figure 3] 1 is a diagram showing an example of a meter installed in a device to which a failure factor analysis device according to an embodiment of the present invention is applied; [Figure 4] FIG. 2 is a diagram showing an example of data stored in a fault knowledge database according to an embodiment of the present invention. [Figure 5] 3 is a graph showing one aspect of the relationship between pieces of information held in a fault knowledge database according to an embodiment of the present invention. [Figure 6] 3 is a flowchart showing an example of the operation of the failure factor analysis device according to the embodiment of the present invention. [Figure 7] 10 is a flowchart illustrating an example of the operation of a collection information determination unit according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, examples of modes for carrying out the present invention (hereinafter referred to as "embodiments") will be described with reference to the accompanying drawings. In this specification and the accompanying drawings, identical or similar components are given the same reference numerals, and redundant explanations may be omitted or only explanations focusing on the differences may be given. The number of each component may be singular or plural unless otherwise specified.

[0013] In the following embodiment, various types of information are described in table format, but the various types of information may be in a data format other than a table format. Furthermore, various names such as "XX information," "XX table," "XX list," and "XX list" are interchangeable. Furthermore, when describing identification information, expressions such as "identification information," "name," and "ID" are used, but these are interchangeable.

[0014] [Configuration of failure cause analysis device] First, a configuration example of a failure factor analysis device according to an embodiment of the present invention will be described with reference to Fig. 1. The present invention can be widely used as a failure factor analysis device and a failure factor analysis method.

[0015] 1 is a block diagram showing an example of the configuration of a failure factor analysis device according to this embodiment. In this embodiment, a failure factor analysis device 200 is connected to a plant 100 to be analyzed. The failure factor analysis device 200 is also connected to a monitoring device 901 in a monitoring room 900 via a network 1000.

[0016] The plant 100 is configured with equipment 110, a control device 120, and an alarm system . The control device 120 receives measurement signals 140 such as temperature, flow rate, and pressure measured by instruments (sensors) installed in the equipment 110, and transmits operation signals 150 (an example of a control signal) for operating the equipment 110. The alarm system 130 receives the measurement signal 160 output from the control device 120, and transmits an alarm signal 170 to the control device 120 as necessary. The alarm system 130 also transmits the alarm signal 170 as a network signal 20 to the monitoring device 901 in the monitoring room 900 via the network 1000.

[0017] The failure factor analysis device 200 includes a collection information determination unit 300, a collection information evaluation unit 500, an information compression unit 700, and a failure mode estimation unit 600 as calculation units, and the operations of these calculation units are controlled by an operation control unit 800 in the failure factor analysis device 200. The operation of these calculation units will be described later with reference to the drawings.

[0018] The failure cause analysis device 200 also includes a failure knowledge database 400 as a database. As will be described in detail later, the failure knowledge database 400 stores information in which inspection items are associated with failure modes. Note that the database is abbreviated as "DB" in Fig. 1. The database stores electronic information, and the information is usually stored in a format called an electronic file (electronic data).

[0019] Furthermore, the failure factor analysis device 200 is provided with an input / output interface 210 as an interface with the outside. The failure factor analysis device 200 can send and receive network signals 20 to and from the network 1000 via the input / output interface 210. The failure factor analysis device 200 is also communicably connected to a monitoring device 901 in a monitoring room 900 via a network signal 21.

[0020] The monitoring room 900 is provided with a monitoring device 901 equipped with an external input device 910 and an image display device 940. The external input device 910 is composed of a keyboard 920 and a mouse 930. A monitor inputs information, instructions, etc. into the monitoring device 901 using the external input device 910, and operates the plant 100 and the failure cause analysis device 200. The image display device 940 displays information, etc. of the network signals 20 and 21 collected from the plant 100 and the failure cause analysis device 200.

[0021] In the failure factor analysis device 200 of this embodiment, the arithmetic device and the failure knowledge database 400 are provided inside the failure factor analysis device 200, but this example is not limiting. Some of these devices may be located outside the failure factor analysis device 200, and only data may be communicated between the devices. In this embodiment, the external input device 910 is composed of a keyboard 920 and a mouse 930, but any device for inputting data, such as a microphone for voice input or a touch panel, may be used. Furthermore, in this embodiment, the failure factor analysis device 200 is applied to a plant, but it goes without saying that the application can also be implemented to facilities other than plants.

[0022] [Computer hardware configuration] Next, the hardware configuration of the computer included in the failure factor analysis device 200 will be described with reference to FIG. Fig. 2 is a block diagram showing an example of the hardware configuration of a computer included in the failure factor analysis device 200. The computer C shown in Fig. 2 may be, for example, a personal computer or a dedicated computer.

[0023] The computer C includes a CPU (Central Processing Unit) 201, a ROM (Read Only Memory) 202, a RAM (Random Access Memory) 203, a non-volatile storage 204, and a communication interface 205. The blocks within the computer C are connected via a system bus so that they can send and receive data to and from each other.

[0024] The CPU 201, ROM 202, RAM 203, and nonvolatile storage 204 constitute a control unit. This control unit is used as an example of a computer that controls the operation of each arithmetic unit of the failure factor analysis device 200 shown in Fig. 1. The CPU 201 reads out software programs that realize the functions of each arithmetic unit from the ROM 202, and loads the programs into the RAM 203 for execution.

[0025] The ROM 202 is used as an example of a non-volatile memory (recording medium). The ROM 202 stores an OS (Operating System), various parameters, programs for operating each device, and the like. The RAM 203 is used as an example of a volatile memory. Variables, parameters, and the like generated during the arithmetic processing of the CPU 201 are temporarily written to the RAM 203. Instead of the CPU 201, another processor such as an MPU (Micro Processing Unit) may be used as the arithmetic processing device.

[0026] The nonvolatile storage 204 is an example of a recording medium and is capable of storing data used by the program and data obtained by executing the program. For example, the nonvolatile storage 204 stores data accumulated in the fault knowledge database 400, information such as thresholds and accuracy for fault determination, and the like. The nonvolatile storage 204 may also store an OS or a program executed by the CPU 201. The nonvolatile storage 204 may be a hard disk drive (HDD), a solid state drive (SSD), an optical or magnetic disk medium, a semiconductor memory card, or the like. The program may be provided to the failure factor analysis device 200 via a wired or wireless transmission medium such as a local area network (LAN), the Internet, or digital satellite broadcasting.

[0027] The communication interface 205 may be, for example, a network interface card (NIC). The communication interface 205 is configured to be capable of transmitting and receiving various data to and from an external device via a communication network such as a LAN or the Internet to which a terminal is connected, or a dedicated line. The communication means may be either wireless communication such as Wi-Fi (registered trademark) or LTE (Long Term Evolution) (registered trademark), or wired communication.

[0028] It should be noted that the monitoring device 901 and the control device 120 may also be considered to have basically the same hardware configuration as the failure cause analysis device 200. As described above, in the monitoring device 901, the external input device 910 and the image display device 940 are connected to the computer C.

[0029] [Example of an instrument] 3 is a diagram showing an example of a meter installed in a device to which the failure factor analysis device 200 according to this embodiment is applied. A flow meter 190 shown in FIG. 3 measures the flow rate in a pipe 180 connected to the device 110.

[0030] Generally, a number of measuring instruments such as flow meters, thermometers, and pressure gauges are installed in the equipment 110 of the plant 100. Measurement signals 140 output by the measuring instruments are transmitted to the control device 120, where the data is accumulated and used to calculate the operation signal 150.

[0031] Measuring instruments include those that can measure values ​​as digital signals and those that display values ​​in analog form, as shown in Figure 3. Analog-displayed measurement values ​​(indicated values) can be confirmed by patrolling the site. Also, analog-displayed measurement values ​​can be photographed with a surveillance camera or a robot equipped with a camera (including unmanned aerial vehicles such as drones), and the measurement values ​​can be confirmed from the captured image. If the captured image is not clear and the indicated value cannot be read, the inspection support system described in Patent Document 2 has the surveillance camera capture the indicating instrument under different imaging conditions.

[0032] On the other hand, analyzing the cause of a failure does not necessarily require accurate readings of the indicated values. In other words, to identify the cause of a failure, there are cases where it is sufficient to know whether the indicated value exceeds a certain threshold, and cases where specific numerical values ​​are required. In the former case, even if the camera is cloudy or out of focus and an accurate value cannot be read, knowing whether the threshold has been exceeded is sufficient for failure cause analysis. The accuracy of the information required to analyze the cause of a failure varies depending on the abnormal event in the plant.

[0033] For example, the failure factor analysis device 200 according to this embodiment determines whether an image has been captured with sufficient accuracy necessary for failure factor analysis, and does not take an image again if the image has been captured with the required accuracy. This makes it possible to collect necessary information with the minimum number of captures and effort required.

[0034] [Fault knowledge database] Next, the data stored in the fault knowledge database 400 according to this embodiment will be described with reference to FIGS. FIG. 4 is a diagram showing an example of data stored in the failure knowledge database 400 of the failure cause analysis device 200. As shown in FIG.

[0035] As shown in FIG. 4, the fault knowledge database 400 stores information such as knowledge about equipment failures, equipment, failure modes, and inspection items. Here, inspection items are the same as sensor measurement items, and failure modes that are expected when the measurement value deviates from the normal state are associated with the inspection items. Furthermore, the inspection items in the fault knowledge database 400 store information such as thresholds and accuracy for failure determination. Information on the accuracy of failure determination is information that indicates whether high-precision or low-precision information is required for failure determination. For example, when high-precision information is required, numerical values ​​that define the upper and lower limits of a certain range can be used. Such numerical values ​​that represent a predetermined range can be considered as thresholds with a range. In the case of a single numerical value, i.e., a single threshold determination, the measurement value may be low-precision information.

[0036] FIG. 5 is a graph showing one aspect of the relationship between the various pieces of information held in the fault knowledge database 400 shown in FIG. 4. In this figure, the information in the fault knowledge database 400 is expressed as a graph with a tree structure. In the figure, solid lines indicate inspection items or error codes, and dashed lines indicate failure modes. Here, the error code is information contained in the alarm signal 170 generated by the alarm system 130, and is symbolic information that indicates the type and cause of a failure occurring in the device 110. This error code is symbolic information assigned to a device failure, and can also be considered information that classifies the device failure.

[0037] In the example in Figure 5, information about equipment I is shown. "Error code 1" (root node) has "Failure mode A" and "Failure mode B" as child nodes. "Failure mode A" has "Inspection item 1, α temperature, 200°C or higher," "Inspection item 2, α pressure, 1-3 MPa," and "Inspection item 3, α flow rate, 10 t / h or less" as child nodes. "Failure mode B" has "Inspection item 2, α pressure, 1-3 MPa" and "Inspection item 3, α flow rate, 10 t / h or less" as child nodes.

[0038] 5, it is also possible to represent the information in the fault knowledge database 400 in a graph such as a tree structure, and display this information on the image display device 940. Although not shown, coordinate information of devices and instruments may be stored in the fault knowledge database 400, and image information may be collected with reference to this coordinate information.

[0039] [Operation of the failure factor analysis device] 6 is a flowchart showing an example of the operation of the failure factor analysis device 200 according to this embodiment. The processing of this flowchart is realized by the operation control unit 800 (FIG. 1) operating each arithmetic unit. The implementation of each step of the flowchart will be described below.

[0040] First, in step S1, the operation control unit 800 operates the collected information determination unit 300. The collected information determination unit 300 receives a plant signal 1 collected in the plant 100 as a plant signal 2 via the input / output interface 210. The plant signal 2 includes signals related to the plant 100, such as a measurement signal 140, an operation signal 150, and an alarm signal 170. The measurement signal 140 is not limited to data of a measurement value measured by an instrument, but may also include data of a captured image of the measurement value indicated by the instrument.

[0041] The collected information determination unit 300 transmits a plant signal 3 corresponding to the plant signal 2 to the fault knowledge database 400, and acquires a fault knowledge database signal 4 related to the alarm signal 170 included in the plant signal 3. The collected information determination unit 300 determines the content and accuracy of information to be collected based on the fault knowledge database signal 4, and outputs a collected information determination signal 5 to the input / output interface 210 and the collected information evaluation unit 500. The detailed processing content of this step S1 will be described later.

[0042] Next, in step S2, the input / output interface 210 transmits a collection information determination signal 6 to the plant 100. The plant 100 collects necessary information according to the content of the collection information determination signal 6. The collected information includes the measurement signals 140, the operation signals 150, and the alarm signals 170, as well as image information captured using a surveillance camera, a drone, a robot, or the like. For example, the collected information includes images of instruments installed in the plant 100.

[0043] Next, in step S3, the operation control unit 800 operates the collected information evaluation unit 500. The collected information evaluation unit 500 receives the collected information signal 7 including information collected in the plant 100 as the collected information signal 8 via the input / output interface 210.

[0044] Next, in step S4, the collected information evaluation unit 500 compares the collected information determination signal 5 determined by the collected information determination unit 300 with the collected information signal 8, and determines whether information has been collected with the content and accuracy required for failure mode estimation. In other words, the collected information evaluation unit 500 verifies the comprehensiveness and necessary sufficiency of the collected information. Then, the collected information evaluation unit 500 transmits a collected information evaluation signal 9 containing information obtained by evaluating the collected information, i.e., the result of the above determination, to the failure mode estimation unit 600, and transmits a collected information evaluation signal 10 to the information compression unit 700. The content of the collected information evaluation signal 9 and the collected information evaluation signal 10 can be considered to be the same.

[0045] If the determination result in step S4 is YES, the process proceeds to step S5. On the other hand, if the determination result in step S4 is NO, the process returns to step S2, and necessary information in the plant 100 is collected again.

[0046] In step S5, the operation control unit 800 operates the failure mode estimation unit 600. The failure mode estimation unit 600 transmits the collected information evaluation signal 10 to the failure knowledge database 400 as a collected information evaluation result 11, and acquires from the failure knowledge database 400 a failure knowledge database signal 12 including a failure mode that is expected to have occurred in the plant 100. The failure mode estimation unit 600 transmits the failure knowledge database signal 12 to the information compression unit 700 as a failure mode estimation result signal 13.

[0047] Next, in step S6, the operation control unit 800 operates the information compression unit 700. The information compression unit 700 compresses the information included in the collected information evaluation signal 9 and the failure mode estimation result signal 13, and transmits a compressed information signal 14 including the compressed information to the input / output interface 210. The input / output interface 210 transmits the compressed information signal 14 as a network signal 20 to the network 1000. The compressed information signal 14 is transmitted via the network 1000 as a network signal 21 to the monitoring device 901 in the monitoring room 900, and the contents of the compressed information are displayed on the image display device 940. After processing step S6, the operation control unit 800 terminates the processing of this flowchart.

[0048] [Operation of the information processing decision unit] Next, the operation of the collection information determining section 300 according to this embodiment will be described in detail with reference to FIG. 7 is a flowchart showing an example of the operation of the collected information determining section 300 according to this embodiment. 7 shows step S1 in FIG.

[0049] First, in step S11, the collection information determining unit 300 refers to the failure knowledge database 400 (FIG. 4) and lists failure modes associated with the error code of the device to be analyzed from the knowledge about failures.

[0050] Next, in step S12, the collection information determining unit 300 lists the test items associated with the failure modes listed in step S11. That is, the collection information determining unit 300 lists the test items necessary for determining the failure mode.

[0051] Next, in step S13, the collection information determining section 300 selects one test item that has not been selected so far from the test items listed in step S12.

[0052] Next, in step S14, the collection information determination unit 300 determines whether or not the fault knowledge database 400 contains information about the judgment accuracy for the test item selected in step S13. Here, the information about judgment accuracy is, for example, information about whether the test item stored in the fault knowledge database 400 is a judgment of a single threshold value (a judgment of only "greater than or equal to" or "less than or equal to"), or whether it is a judgment of a certain predetermined range (a threshold value with a width). If the judgment result is YES, the process proceeds to step S16, and if the judgment result is NO, the process proceeds to step S15.

[0053] In step S15, the collected information determination unit 300 utilizes generative artificial intelligence (AI) or the like to supplement missing information related to accuracy determination. Generative AI is also called generative AI. In this step, if there is no information on fault determination thresholds, accuracy, etc. stored in the fault knowledge database 400, the generative AI is utilized to collect additional information on these fault determination thresholds, accuracy, etc. This makes it possible to simplify and reduce the information stored in the fault knowledge database 400. After this process is completed, the process proceeds to step S16.

[0054] The generative AI is an AI model realized by applying a machine learning model, a large-scale language model (LLM), etc. When the target information is lacking, the AI ​​model is configured to access resources on the network 1000 and search for and acquire information related to the accuracy of judgment of the inspection items of the target device (such as the threshold and accuracy of fault judgment). For example, the AI ​​model searches for specification information of the target device posted on a website published by the device manufacturer and extracts information related to the accuracy of judgment of the inspection items from the specification information. The AI ​​model associates the acquired information related to the accuracy of judgment of the inspection items with the device's failure mode and stores it in the fault knowledge database 400. The AI ​​model of the generative AI can be stored in the non-volatile storage 204.

[0055] In step S16, the collection information determination unit 300 determines whether the information on the judgment accuracy confirmed in step S14 or the information on the judgment accuracy supplemented in step S15 is sufficient to be threshold judgment only. The criterion for the determination is the contents of the inspection items stored in the fault knowledge database 400. If the information on the judgment accuracy is threshold judgment only (YES determination in step S16), the process proceeds to step S18; otherwise (NO determination in step S16), the process proceeds to step S17.

[0056] In this way, if the information on the accuracy of fault judgment contained in the fault knowledge database 400 indicates high accuracy, the collected information determination unit 300 decides to collect measurement data and image information of measurement values ​​with high accuracy, and if the information on the accuracy of fault judgment indicates low accuracy, it decides to collect measurement data and image information of measurement values ​​with low accuracy.

[0057] In step S18, the collection information determining unit 300 determines that low-accuracy information collection is required for the test items selected in step S13. In addition, in step S17, the collection information determining unit 300 determines that high-accuracy information collection is required for the test items selected in step S13. That is, the collection information determining unit 300 determines that low-accuracy information is required when it is sufficient to know whether or not a measurement value exceeds a threshold, and that high-accuracy information collection is required when specific numerical values ​​must be known.

[0058] For example, if low-accuracy information is sufficient for failure cause analysis, but only high-accuracy information is available, the high-accuracy information (although sufficient) is not necessarily required. Conversely, if high-accuracy information is desired but only low-accuracy information is available, the low-accuracy information is not sufficient (the information is insufficient for cause analysis). In this embodiment, the information to be collected can be optimized by determining the information to be collected from the above-mentioned perspective.

[0059] After processing step S17 or step S18, in step S19, the collected information determination unit 300 determines whether the accuracy of information collection has been determined for all test items. If the accuracy of information collection has not been determined for all test items (NO determination in step S19), the process returns to step S13, selects a test item that has not been selected so far, and executes the processes of steps S14 to S19 as appropriate. On the other hand, if the accuracy of information collection has been determined for all test items (YES determination in step S19), the process of this flowchart ends.

[0060] [Specific examples of determining information to be collected] Next, a specific example of a failure cause analysis when the flowcharts of FIGS. 6 and 7 are operated using the knowledge related to failures in the failure knowledge database 400 shown in FIGS. 4 and 5 will be described.

[0061] When error code 1 occurs, first, in step S1, the collection information determination unit 300 extracts "failure mode A" and "failure mode B" as failure modes associated with error code 1. Next, the collection information determination unit 300 extracts "α temperature," "α pressure," and "α flow rate" as inspection items associated with these failure modes. The collection information determination unit 300 determines that low-precision information collection is required for the inspection of "α temperature" and "α flow rate," since only threshold judgment is required, and determines that high-precision information collection is required for the inspection of "α pressure," since it is for other cases.

[0062] For example, if the α temperature of inspection item 1 is 200°C or higher, inspection item 1 is associated with "failure mode A" in FIG. 4, and therefore step S5 (failure mode estimation unit 600) outputs "failure mode A" as the estimated failure mode.

[0063] Furthermore, when placing a camera later, the camera position may be determined based on the determination results in steps S17 and S18. That is, the camera may be placed near an instrument that requires highly accurate information collection (in the case of step S17). On the other hand, the camera may be placed at a position away from an instrument that requires less accurate information collection, and the camera may be rotated to capture an image of the instrument (in the case of step S18).

[0064] As described above, the failure factor analysis device 200 according to this embodiment is a failure factor analysis device that analyzes the failure factors of failures occurring in equipment based on error codes generated from measurement data of the equipment, and outputs information related to the analysis results. This failure cause analysis device 200 is configured to include a failure knowledge database 400 in which knowledge related to equipment failures is stored, a collected information determination unit 300 that determines the content of information to be collected from the equipment based on the error code and the information stored in the failure knowledge database 400, a collected information evaluation unit 500 that evaluates the information collected from the equipment, and a failure mode estimation unit 600 that estimates the equipment failure mode from the evaluation results by the collected information evaluation unit 500 and the information stored in the failure knowledge database 400. The collected information determination unit 300 lists the inspection items required to determine the failure mode associated with the error code, and determines to collect measurement data and image information of the measurement values ​​with the accuracy necessary and sufficient for determining the inspection items.

[0065] By using the failure factor analysis device 200 according to this embodiment configured as described above, it is possible to efficiently and automatically collect information necessary for failure factor analysis with sufficient accuracy. As a result, in this embodiment, it is possible to minimize the need for on-site dispatches to collect information for failure factor analysis.

[0066] As described above, the present invention is not limited to the above-described embodiments, and various other modifications and applications are possible without departing from the spirit of the invention as set forth in the claims. For example, the above-described embodiments have been described in detail and specifically to clearly explain the present invention, and are not necessarily limited to those including all of the components described. Furthermore, it is also possible to add, replace, or delete other components to or from part of the configuration of each embodiment.

[0067] Furthermore, the above-described configurations, functions, processing units, etc. may be partially or entirely realized in hardware, for example, by designing them as integrated circuits, etc. As the hardware, a broad processor device such as an FPGA (Field Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit) may be used. [Explanation of symbols]

[0068] 1-3...plant signal, 4...fault knowledge database signal, 5-6...collected information determination signal, 7-8...collected information signal, 9-11...collected information evaluation signal, 12...fault knowledge database signal, 13...failure mode estimation result signal, 14...compressed information signal, 20-21...network signal, 100...plant, 110...equipment, 120...control device, 130...alarm system, 140...measurement signal, 150...operation signal, 160...measurement signal, 170...alarm signal, 190...instrument, 200...failure cause analysis device, 210...input / output interface, 300...collected information determination unit, 400...fault knowledge database, 500...collected information evaluation unit, 600...failure mode estimation unit, 700...information compression unit, 800...equipment operation control unit, 900...monitoring room, 901...monitoring device, 910...external input device, 940...image display device

Claims

1. A failure cause analysis device that analyzes the cause of a failure occurring in a device based on an error code generated from measurement data of the device, and outputs information about the analysis result, a fault knowledge database in which knowledge relating to faults of the device is stored; a collection information determination unit that determines the content of information to be collected from the device based on the error code and the information stored in the failure knowledge database; a collected information evaluation unit that evaluates information collected from the device; a failure mode estimation unit that estimates a failure mode of the device from the evaluation result by the collected information evaluation unit and information stored in the failure knowledge database, The collected information determination unit lists inspection items required to determine a failure mode associated with the error code, and determines to collect measurement data and image information of the measurement values ​​with sufficient accuracy to determine the inspection items. Failure cause analysis device.

2. The failure knowledge database stores information on devices, failure modes, thresholds for failure determination, and accuracy. The failure factor analysis device according to claim 1 .

3. The collected information determination unit determines to collect measurement data and image information of measurement values ​​with high accuracy when the information on the accuracy of the failure determination included in the failure knowledge database indicates high accuracy, and determines to collect measurement data and image information of measurement values ​​with low accuracy when the information on the accuracy of the failure determination indicates low accuracy. The failure factor analysis device according to claim 2 .

4. If there is no information on the threshold and accuracy of fault determination stored in the fault knowledge database, the collected information determination unit collects the information additionally using an AI model. The failure factor analysis device according to claim 3 .

5. The collected information evaluation unit verifies the comprehensiveness and necessary sufficiency of the collected information. The failure factor analysis device according to claim 3 .

6. an information compression unit that compresses and outputs information on the evaluation result by the collected information evaluation unit and information on the estimation result by the failure mode estimation unit; The failure factor analysis device according to claim 1 .

7. A failure cause analysis method for a failure cause analysis device that analyzes a failure cause of a failure occurring in a device based on an error code generated from measurement data of the device and outputs information about the analysis result, a collection information determination step of determining the content of information to be collected from the device based on the error code and the information stored in a failure knowledge database that stores knowledge about device failures; a collected information evaluation step of evaluating information collected from the device; a failure mode estimation step of estimating a failure mode of the device from the evaluation result in the collected information evaluation step and information stored in the failure knowledge database, In the step of determining information to be collected, inspection items necessary for determining a failure mode associated with the error code are listed, and it is determined that measurement data and image information of measurement values ​​are to be collected with sufficient accuracy for determining the inspection items. Failure cause analysis method.

8. The failure knowledge database stores information on devices, failure modes, thresholds for failure determination, and accuracy. The failure cause analysis method according to claim 7.

9. In the collection information determination step, if the information on the accuracy of the fault determination included in the fault knowledge database indicates high accuracy, it is determined to collect measurement data and image information of measurement values ​​with high accuracy, and if the information on the accuracy of the fault determination indicates low accuracy, it is determined to collect measurement data and image information of measurement values ​​with low accuracy. The failure cause analysis method according to claim 8.

10. If there is no information on the threshold and accuracy of fault determination stored in the fault knowledge database, the collected information determination step uses an AI model to collect the information additionally. The failure cause analysis method according to claim 9.

11. In the collected information evaluation step, the comprehensiveness and necessary sufficiency of the collected information are verified. The failure cause analysis method according to claim 9.

12. an information compression step of compressing and outputting information on the evaluation result obtained by the collected information evaluation step and information on the estimation result obtained by the failure mode estimation step; The failure cause analysis method according to claim 7.

Images