JP7199986B2 - Equipment maintenance system and method - Google Patents

Description

The present invention relates to equipment maintenance technology, and more particularly to equipment maintenance technology suitable for effectively utilizing IoT data for measuring equipment such as production facilities for maintenance.

2. Description of the Related Art In recent years, information processing apparatuses have come into wide use for systems that maintain equipment based on sensor data for measuring equipment such as production equipment in factories.

In particular, with the explosive spread of the Internet, not only information processing equipment but also things that exist in the world have communication functions, and by connecting to the Internet and communicating with each other, automatic recognition, automatic control, remote control, etc. The so-called IoT (Internet of Things), which performs measurement, has attracted attention. IoT makes it possible to instantly grasp the situation of things, including many people, and to realize new systems and services by utilizing and combining such information.

Under such circumstances, there is an urgent need to utilize IoT data such as a wide variety of enormous sensor data in equipment maintenance systems. A system that uses sensor data for equipment monitoring and preventive maintenance is disclosed, for example, in Japanese Unexamined Patent Application Publication No. 2002-200010.

JP-A-2000-259222

In the prior art patent document 1, the device state is monitored and its state change is predicted by means of a calculation means for diagnosing the deterioration of the device and estimating the speed, which is provided in the management device, based on the information from the sensor, and the result is is displayed on the display means of the management device.

However, the technology described in the prior art diagnoses the deterioration of equipment using sensor data that is easy to understand as appropriate, and does not take into consideration the diagnosis that takes hidden causes into consideration.

An object of the present invention is to effectively utilize IoT data such as various sensor data, analyze the IoT data, and provide equipment maintenance technology that enables maintenance of equipment in consideration of background information of the equipment.

The configuration of the equipment maintenance system of the present invention preferably includes a processor and a storage device storing a maintenance information database, and is an equipment maintenance system that outputs maintenance information for failure events, and uses sensor data as background information. Factors that are rules for conversion, factors that classify failure events, background information acquisition policies that are associated with important factors of sensor data, and rankings of important factors are held in a storage device, and a processor stores failure event information. Accepts input, selects important factors based on defect event information, background information acquisition policy, and ranking of important factors, and acquires background information that stores sensor data in the background information format corresponding to the selected important factors Then, based on the defect event information and background information, the maintenance information database is searched to output the maintenance information, the input of the maintenance evaluation information for the output maintenance information is received, and the input maintenance evaluation information is processed. Based on this, the ranking of the important factors of the background information acquisition policy is updated .

According to the present invention, it is possible to provide an equipment maintenance technique that enables equipment maintenance by effectively utilizing IoT data such as various sensor data, analyzing the IoT data, and considering the background information of the equipment.

1 is an overall configuration diagram of an equipment maintenance system; FIG. 1 is a hardware/software configuration diagram of an equipment maintenance device; FIG. It is a figure which shows an example of sensor data. FIG. 10 is a diagram showing an example of a background information acquisition policy; It is a figure which shows an example of ranking. It is a figure which shows an example of background information. It is a figure which shows an example of a query. It is a figure which shows an example of a maintenance instruction sheet. 4 is a flow chart showing a series of flows from an event input by a field worker to output of maintenance information corresponding to the event input by the equipment maintenance system. It is a figure which shows an example of a malfunction event input screen. It is a figure which shows an example of a malfunction event evaluation screen.

An embodiment according to the present invention will be described below with reference to FIGS. 1 to 11. FIG.

First, the configuration of the overall equipment maintenance system will be described with reference to FIGS. 1 and 2. FIG.

The overall equipment maintenance system is, as shown in FIG.

The equipment maintenance system 100 is a device for inputting information about trouble events, outputting information about maintenance of equipment, and displaying it to the field worker 1 and the maintenance manager 2. A sensor data input/output unit 110, a maintenance information I It consists of functional units such as an /F (InterFace) unit 120 , a search engine front unit 140 , a search engine 150 , a similar case extraction unit 160 and an operation log utilization unit 170 .

The edge 180 is a device that controls and manages maintenance target equipment at the site, and has a sensor data background information conversion unit 130 .
The sensor data input/output unit 110 is a functional unit that takes the sensor data 30 into the edge 180 . The sensor data 30 is data obtained by measuring the state of the maintenance target device 10 installed at the site by the sensor 20, the maintenance target device 10 itself, and its control device.

A maintenance information I/F (InterFace) unit 120 is an interface that imports information on trouble events and information on maintenance from the on-site maintenance terminal 40 and the maintenance management terminal 50 into the equipment maintenance system 100, and outputs records and advice on maintenance. Department.

The sensor data background information conversion section 130 is a functional section that converts the sensor data 30 into background information and returns it in response to a request from the search engine front section 140 . The sensor data background information conversion section 130 is composed of subcomponents of a conversion interpretation rule setting section 131 , a conversion processing section 132 , a request reception section 133 and a response return section 134 .

The conversion interpretation rule setting unit 131 refers to the background information acquisition policy 230 and sets rules for converting the sensor data 30 into background information. The conversion processing unit 132 is the body of processing for converting the sensor data 30 into background information. The request receiving unit 133 and the response returning unit 134 receive requests from the search engine front unit 140 and return corresponding background information.

The search engine front unit 140 is a functional unit that performs front processing for processing and passing input information before processing with the search engine 150 . The search engine front section 140 is composed of subcomponents of a device background information acquisition control section 141 and a merge processing section 142 .

Before issuing a query to the search engine 150, the device background information acquisition control unit 141 requests necessary background information from the sensor data background information conversion unit 130. FIG. The merge processing unit 142 merges the background information returned from the sensor data background information conversion unit 130 in response to a request with other parameters required for the query.

The search engine 150 is a part that inputs a query in response to a request from the on-site maintenance terminal 40 or the maintenance management terminal 50 and performs search processing from the maintenance record data lake 200 .

The maintenance record data lake 200 is a database that stores maintenance records relating to past maintenance target equipment 10 .

The similar case retrieving unit 160 is a functional unit that retrieves past similar maintenance cases and sends them to the field maintenance terminal 40 or the maintenance management terminal 50 in response to a request from the field maintenance terminal 40 or the maintenance management terminal 50 .

The operation log utilization unit 170 is a functional unit provided to utilize the operation log 220 related to on-site maintenance equipment. The operation log utilization unit 170 is composed of subcomponents of an evaluation collection unit 171 , a knowledge edit accumulation unit 172 and an operation log I/F unit 173 .

The evaluation collection unit 171 collects evaluations of responses from the equipment maintenance system input by the maintenance manager 2 from the maintenance management terminal 50 . The knowledge editing and accumulating unit 172 allows field worker 1 or maintenance manager 2 to edit and accumulate knowledge related to equipment maintenance in operation log 220 . The operation log I/F unit 173 performs input/output between the operation log utilization unit 170 and the operation log 220 .

The operation log 220 is information that is stored as a record of the operation of the equipment maintenance system in addition to evaluations and knowledge about equipment maintenance.

From the on-site maintenance terminal 40, the on-site worker 1 inputs a problem event in the form of selecting a keyword or free description of a sentence when the equipment 10 to be maintained breaks down. A specific user interface will be described in detail later.

To the maintenance management terminal 50, the equipment maintenance system outputs past similar cases and recommended countermeasures that serve as responses to input of trouble events, and the maintenance manager 2 inputs the evaluation thereof. In addition, the maintenance manager 2 displays a maintenance instruction sheet on the maintenance management terminal 50 on the field maintenance terminal 40 and instructs the field worker 1 . Maintenance instructions are registered in the operation log 220 .
Furthermore, from the operation log 220, a background information acquisition policy 230 that is used when converting the sensor data 30 into background information is generated. The field maintenance terminal 40 and the maintenance management terminal 50 may be the same terminal.

Next, the hardware/software configuration of the equipment maintenance system will be described with reference to FIG.

As a hardware configuration of the equipment maintenance system 100, for example, as shown in FIG. 2, it is realized by an information processing system such as a server.

The equipment maintenance system 100 has a form in which a CPU (Central Processing Unit) 302, a main storage system 304, a network I/F 306, and an auxiliary storage I/F 312 are connected by a bus.

The CPU 302 controls each part of the equipment maintenance system 100, loads necessary programs into the main storage device 304, and executes them.

The main memory device 304 is normally composed of a volatile memory such as a RAM, and stores programs executed by the CPU 302 and data to be referred to.

A network I/F 306 is an interface for connecting with the network 5 .

Auxiliary storage I/F 312 is an interface for connecting auxiliary storage devices such as HDD (Hard Disk Drive) 350 and SSD (Solid Sate Drive).

The HDD 350 has a large storage capacity and stores programs for executing this embodiment. The equipment maintenance system 100 includes a sensor data input program 401, a maintenance information I/F program 402, a sensor data background information conversion program 403, a search engine front program 404, a search engine program 405, a similar case extraction program 406, and an operation log utilization program. 407 is installed.

A sensor data input program 401, a maintenance information I/F program 402, a sensor data background information conversion program 403, a search engine front program 404, a search engine program 405, a similar case extraction program 406, and an operation log utilization program 407 run on the CPU 302. By doing so, sensor data input/output unit 110, maintenance information I / F (InterFace) unit 120, sensor data background information conversion unit 130, search engine front unit 140, search engine 150, similar case extraction unit 160, operation log utilization It performs the functions of section 170 .

The HDD 350 also holds a maintenance record data lake 200 , an operation log 220 , and a background information acquisition policy 230 .

Next, details of data used in the equipment maintenance system will be described with reference to FIGS. 3 to 8, 10, and 11. FIG.
The sensor data 30 is data measured by the sensor 20 with respect to the maintenance target device 10, and is recorded paired with the date and time of measurement. The example of FIG. 3 shows an example of vibration data in which the vibration frequency, the amplitude, and the date and time of measurement that satisfy a predetermined condition (large amplitude, etc.) output from a vibration-related sensor are shown.

The background information acquisition policy 230 shown in FIG. 4 is data that becomes a rule for acquiring the corresponding sensor data 30 and converting it into background information when a field worker inputs information about an event. The example of FIG. 4 shows an example of a data structure in which defect/defective phenomena are classified into major, medium, and small categories, arranged on a tree, and important factors indicated in the terminal leaves. ing. Multiple important factors are set for terminal leaves. The important factor data structure also includes data measurement locations and background information formats (maximum amplitude, its frequency, frequency of amplitudes above a predetermined level, etc.). The background information format is data that is entered at the same time that field workers enter information about an event and issue a query for information about maintenance from the system.
The ranking 260 shown in FIG. 5 stores the degree of usefulness and ranking of the important factors set for the terminal leaves. This varies according to the evaluation from the maintenance worker .

The background information 240 shown in FIG. 6 is data obtained by inputting a value converted based on the sensor data 30 into the background information format. The example shown in FIG. 6 shows data values relating to the abnormal frequency and maximum amplitude, and information specifying the site of occurrence.

A query 250 shown in FIG. 7 is a command to be input to the search engine 150 in order to inquire information about maintenance from the system when a field worker inputs information about a trouble event. In the example shown in FIG. 7, for the background information, the maximum vibration frequency is 3 times/day, the maximum amplitude is 0.094 mm, the occurrence site is the processing pedestal, and the input keywords are "processing failure", "screw Queries for similar cases are shown for "tip" and "chipping crack".

The maintenance instruction sheet shown in FIG. 8 indicates maintenance instructions to the field worker 1 by the maintenance manager 2 with reference to past cases and operation logs, and is transmitted from the maintenance management terminal 50 to the field maintenance terminal 40. be. In this example, maintenance instructions are shown for the working pedestal.

A trouble event input screen 500 shown in FIG. 10 is a screen for the field worker 1 to input a trouble event regarding the maintenance target equipment 10 at the work site. It consists of an occurrence date entry field 501 , a time period entry field 502 , a processed item number entry field 503 , a malfunction event entry field 510 and a free description entry field 520 .

Occurrence date input field 501, time period input field 502, and processed product number input field 503 are fields for inputting the date on which the problem event occurred, its time period, and the processed product number related to the problem event, respectively.

The trouble event entry field 510 consists of a site entry field 511 and a symptom entry field 512, and is a field for entering the site and symptom related to the trouble event. In FIG. 10, both the site input field 511 and the symptom input field 512 are designed to select items from lists. In the example shown in FIG. 10, it is assumed that "poor processing, screwing tip" is selected as the site of the defect event in FIG. 9, and "chipping and cracking" is selected as the symptom.

The free description input field 520 is a field in which the field worker 1 can freely describe the trouble event.

In this embodiment, the keyword is obtained from the region input field 511 and the symptom input field 512, but it may be obtained from the text input in the free description input field 520. FIG. Further, if the information on the malfunction event can be automatically obtained from the target device or the like, the information may be obtained automatically.

The failure event evaluation screen 600 shown in FIG. 11 is a screen for outputting maintenance information from the equipment maintenance system 100 to the maintenance manager 2 as a response to the input failure event. It consists of an input information column 610 and a maintenance information column 620.

The input information column 610 is a column for displaying information input by the field worker 1 from the trouble event input screen 500 of FIG.

The maintenance information column 620 is a column for displaying maintenance information answered from the equipment maintenance system 100 , and consists of a background information column 621 , a past case column 622 , a recommended countermeasure column 623 and an evaluation button 624 .

The background information column 621 is a column for displaying the background information obtained by converting the sensor data by the sensor data background information conversion unit 130 in an easy-to-understand format. The past case column 622 is a column for displaying similar cases regarding the failure event. The recommended countermeasure column 623 is a column for displaying countermeasures recommended by the equipment maintenance system 100 for the failure event. The evaluation button 624 is a button for the maintenance manager 2 to input an evaluation of the maintenance information displayed by the equipment maintenance system 100 . In the example of FIG. 11, an evaluation of "Like!" is input, but a negative evaluation may be input, or a 5-level evaluation may be input. Evaluations of maintenance information made by the maintenance manager 2 are accumulated in the operation log 220 . By reflecting cases with high evaluations in the background information acquisition policy, it is expected that more accurate searches will become possible.

Next, processing related to the equipment maintenance system will be described with reference to FIG.
First, the equipment maintenance system 100 waits (S101, S102) until the site worker 1 inputs a problem event from the problem event input screen 500, and proceeds to S103 when the problem event information is input.

The equipment background information acquisition control unit of the search engine front unit 140 of the equipment maintenance system 100 receives the trouble event information, refers to the background information acquisition policy 230 corresponding to the trouble event information (S103), and determines that valid background information exists. It is determined whether or not to do so (S104). Here, the effective background information is taken out when the defect event information matches the criteria of the system, such as matching the nodes of the tree indicated by the background information acquisition policy 230 shown in FIG. 4 at a predetermined ratio or more. background information (format) In the background information format for nodes such as branches and leaves, a plurality of important factors are set for the data column. choose for. If there is no effective background information, it corresponds to the case where the problem can be solved without using sensor data or the case where there is no effective solution that the system can present. In this embodiment, it is assumed that the keywords "processing defect", "screw tip", and "chipping crack" are input from the defect event input screen 500, and that there is effective background information.

When the device background information acquisition control unit 141 of the search engine front unit 140 determines that valid background information exists (S104: YES), the background information acquisition policy 230 that has been input is included in the sensor data background information conversion unit 130. A background information request is sent to request background information (S105).

A request reception unit 133 of the sensor data background information conversion unit 130 receives a background information request including the background information acquisition policy 230 . The conversion/interpretation rule setting unit 131 sets a conversion rule for converting the relevant sensor data 30 into background information based on the background information acquisition policy 230 that has been input. The conversion processing unit 132 converts the sensor data 30 into background information based on the conversion rule, and returns the background information to the merge processing unit 142 of the search engine front unit 140 (S106). Here, the sensor data 30 may be either data acquired from the sensor after the background information request is accepted or data already acquired from the sensor and stored in a storage device (not shown).

For example, in the examples shown in FIGS. 3 to 6, as keywords "processing failure", "screw tip", and "chipping crack" are input, the background information acquisition policy 230 identifies important factors in the processing environment. , the "vibration of the pedestal" is taken out. Then, the conversion/interpretation rule setting unit 131 acquires the sensor data 30 related to the pedestal, furthermore, gives rules such as “acquisition of the abnormal frequency of the day” and “acquisition of the maximum amplitude”, and sends the data to the conversion processing unit 132. Instruct the conversion. As a result, background information is generated from the viewpoints of "abnormal frequency", "maximum amplitude", and "location of occurrence" as shown in FIG.

Next, in the search engine front section 140, the merge processing section 142 merges the input background information, creates a query (S108), and inputs it to the search engine 150. FIG. The merging section 142 can be omitted if the background information includes the defect event information in advance.

Here, as shown in FIG. 7, the input keywords are "processing defect", "screw tip", and "chipping crack", and the background information is "maximum vibration frequency is 3 times/ It is assumed that a query is generated by merging "day, maximum amplitude of 0.094 mm, occurrence site of machined pedestal" to inquire about similar cases of the input failure event.

Next, the search engine 150 performs search processing according to the input query (S109).

Next, the similar case retrieving unit 160 of the equipment maintenance system 100 displays the searched maintenance instructions (S110), and the maintenance manager 2 performs maintenance according to the maintenance instructions.
When the maintenance is completed, the trouble event evaluation screen 600 is output to the maintenance management terminal 50 according to the instruction of the maintenance manager 2 (S111). At that time, the similar case extractor 160 extracts a similar case of the problem event and displays it on the problem event evaluation screen 600 .

Then, the maintenance manager 2 inputs an evaluation of the response regarding the failure event from the equipment maintenance system 100 (S112).
The evaluation collection unit of the operation log utilization unit 170 stores the evaluation input via the operation log I/F unit 173 in the operation log 220 in association with the maintenance instruction sheet.
The knowledge editing storage unit edits the ranking 260 of the background information acquisition policy 230 based on the evaluations stored in the operation log 220 . For example, an item (important factor) that causes a search for maintenance instructions with a high evaluation is given a higher ranking 260, an item that leads to a maintenance instruction with a lower evaluation is given a lower ranking 260, and other items are ranked. 260 is edited to be relatively high. By doing so, the important factor of the background information format for the same failure event information can be replaced based on the evaluation by maintenance personnel, so that appropriate background information can be obtained by utilizing the knowledge of maintenance personnel. can.

As described above, according to the equipment maintenance system of the present embodiment, sensor data can be validated to convert sensor data into background information based on the entered defect event keyword and the background information acquisition policy. It is possible to maintain the equipment considering the background information of the equipment.

1 Field worker 2 Maintenance manager 5 Network 10 Equipment to be maintained 20 Sensor 30 Sensor data 100 Equipment maintenance system 110 Sensor data input/output unit 120 Maintenance information I /F unit 130 sensor data background information conversion unit 140 search engine front unit 150 search engine 160 similar case extraction unit 170 operation log utilization unit 200 maintenance record data lake 220 operation log , 230 Background information acquisition policy

Claims (6)

a processor;
a storage device storing a maintenance information database;
An equipment maintenance system that outputs maintenance information for failure events, a rule for converting sensor data into background information, and a background information acquisition policy that associates factors that classify failure events and important factors of sensor data. and the ranking of the important factors in the storage device,
The processor
Receiving input of defect event information,
selecting the important factors based on the failure event information, the background information acquisition policy, and the ranking of the important factors, and obtaining background information in which sensor data is stored in a background information format corresponding to the selected important factors; Acquired,
searching the maintenance information database and outputting the maintenance information based on the malfunction event information and the background information;
An equipment maintenance system, wherein input of maintenance evaluation information for said output maintenance information is received, and ranking of important factors of said background information acquisition policy is updated based on said input maintenance evaluation information. In claim 1,
transmitting the keyword of the malfunction event and the background information acquisition policy to an external device;
An equipment maintenance system, wherein background information containing sensor data is obtained from the external device. In claim 2,
The external device is
The background information is created by converting and storing the sensor data in a background information format associated with the important factor based on the keyword of the malfunction event and the background information acquisition policy. equipment maintenance system. In claim 1,
An equipment maintenance system, wherein the important factor includes a measurement position of a sensor. In claim 4,
The equipment maintenance system, wherein the important factor further includes a conversion method of sensor data. a processor;
An equipment maintenance method for outputting maintenance information for a malfunction event by an equipment maintenance system comprising a storage device storing a maintenance information database,
a step in which the storage device holds a background information acquisition policy, which is a rule for converting sensor data into background information, and in which factors that classify failure events and important factors of sensor data are associated with each other;
the storage device holding a ranking of the important factors;
the processor accepting input of fault event information;
The processor selects the important factors based on the failure event information, the background information acquisition policy, and the ranking of the important factors, and stores sensor data in a background information format corresponding to the selected important factors. obtaining background information;
a step in which the processor searches the maintenance information database based on the trouble event information and the background information and instructs to output the maintenance information;
a step of accepting input of maintenance evaluation information for the output maintenance information;
and updating a ranking of important factors of the background information acquisition policy based on the inputted maintenance evaluation information.

Images