Information processing method, information processing device, and computer program

Abstract

This invention provides an information processing method, an information processing device, and a computer program for performing processing to obtain countermeasures against problems in a plant. [Solution] The information processing method obtains queries related to troubles in a plant that incinerates biomass, waste, or sludge, or burns gas generated from biomass, waste, or sludge, obtains countermeasures for the troubles corresponding to the obtained queries from a language model that uses a database that records knowledge about the plant, and outputs the obtained countermeasures.

Description

Technical Field

[0001] The present invention relates to an information processing method, an information processing apparatus, and a computer program.

Background Art

[0002] In various fields of business, when some trouble occurs, workers investigate the trouble, find a solution to the trouble, and perform work to solve the trouble. In order to stabilize the business, it is desirable that any worker can handle the trouble in the same way. Patent Document 1 discloses a system that outputs an instruction for dealing with trouble according to information from a resident when trouble occurs in a residence.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In plants such as waste incineration plants, various troubles such as component failures can occur. In order to operate the plant stably, it is desirable that any worker can handle the trouble in the same way. If a countermeasure against the occurred trouble can be obtained, the worker can handle the trouble.

[0005] An object of the present invention is to provide an information processing method, an information processing apparatus, and a computer program that perform processing for obtaining a countermeasure against trouble in a plant.

Means for Solving the Problems

[0006] An information processing method according to one embodiment of the present invention is characterized by acquiring queries related to troubles in a plant that incinerates biomass, waste, or sludge, or burns gas generated from biomass, waste, or sludge, acquiring countermeasures for the troubles corresponding to the acquired queries from a language model that uses a database recording knowledge about the plant, and outputting the acquired countermeasures.

[0007] An information processing device according to one embodiment of the present invention comprises a calculation unit, the calculation unit acquires queries relating to troubles in a plant that incinerates biomass, waste, or sludge, or burns gas generated from biomass, waste, or sludge, and acquires countermeasures for the troubles corresponding to the acquired queries from a language model using a database that records knowledge about the plant, and outputs the acquired countermeasures.

[0008] A computer program according to one embodiment of the present invention is characterized by causing a computer to execute a process that acquires queries relating to troubles in a plant that incinerates biomass, waste, or sludge, or burns gas generated from biomass, waste, or sludge, acquires countermeasures for the troubles corresponding to the acquired queries from a language model that uses a database that records knowledge about the plant, and outputs the acquired countermeasures.

[0009] In one embodiment of the present invention, in response to a query regarding a plant trouble, a language model using a database that records plant-related knowledge is used to obtain and output countermeasures for the trouble. By using the language model and database, appropriate countermeasures for troubles can be output. Any worker, including inexperienced workers, can obtain appropriate countermeasures in the event of a plant trouble. [Effects of the Invention]

[0010] The present invention offers excellent benefits, such as providing countermeasures against problems in plants. [Brief explanation of the drawing]

[0011] [Figure 1] This is a schematic diagram showing an example of the configuration of an information processing system. [Figure 2] This is a schematic diagram showing an example of a plant configuration. [Figure 3] This is a block diagram showing an example of the internal configuration of an information processing device. [Figure 4] This is a block diagram showing an example of the internal configuration of a terminal device. [Figure 5] This flowchart shows an example of the steps an information processing device takes to provide solutions to a problem. [Figure 6] This flowchart shows an example of the steps an information processing system performs to control a plant. [Figure 7] This flowchart shows an example of the steps an information processing device performs to output recommendations. [Modes for carrying out the invention]

[0012] The present invention will be described in detail below with reference to the drawings illustrating its embodiments. Figure 1 is a schematic diagram showing an example configuration of the information processing system 100. The information processing system 100 is a system that processes and outputs countermeasures for dealing with problems related to plant 3. The information processing system 100 includes an information processing device 1. The information processing device 1 is connected to plant 3. For example, the information processing device 1 is connected to plant 3 via a communication network using wireless or wired communication. The information processing device 1 may be connected to multiple plants 3. The information processing system 100 also includes a terminal device 2. The terminal device 2 is capable of communicating with the information processing device 1. For example, the terminal device 2 communicates with the information processing device 1 via a communication network using wireless or wired communication. The terminal device 2 is used by workers 4. The information processing system 100 may include multiple terminal devices 2.

[0013] Figure 2 is a schematic diagram showing an example configuration of Plant 3. Plant 3 is a power plant that generates electricity by incinerating biomass, waste, or sludge, or by burning gas generated from biomass, waste, or sludge, and utilizing the heat generated. Plant 3 consists of multiple pieces of equipment. Plant 3 includes a boiler 31. The boiler 31 includes a combustion furnace 311 and a heat exchanger 312. Plant 3 includes a dryer 32 for drying sludge, a conveyor 33 for transporting waste or biomass, or a fermentation bed 34 for generating methane gas by igniting waste, biomass, or sludge. Sludge dried in the dryer 32, or waste or biomass transported by the conveyor 33, is supplied as fuel to the combustion furnace 311, and the combustion furnace 311 burns the fuel. In this way, waste, biomass, or sludge is incinerated. Alternatively, methane gas generated in the fermentation bed 34 is supplied as fuel to the combustion furnace 311, and the combustion furnace 311 burns the fuel.

[0014] Combustion in the combustion furnace 311 generates heat, which in turn generates steam in the heat exchanger 312. Plant 3 includes a turbine 351, a generator 352 connected to the turbine 351, a condenser 353, and a feedwater pump 36. The heat exchanger 312 and the turbine 351 are connected by steam piping, and the steam generated in the heat exchanger 312 rotates the turbine 351. Electricity is generated by the generator 352 in accordance with the rotation of the turbine 351. The steam that rotated the turbine 351 is condensed in the condenser 353. Water is supplied to the heat exchanger 312 by the feedwater pump 36. Plant 3 includes a bag filter 37. Exhaust from the combustion furnace 311 is discharged after passing through the bag filter 37. The bag filter 37 collects dust contained in the exhaust. Plant 3 includes a urea decomposition reactor 38. The urea decomposition reactor 38 supplies the denitrifying agent produced by decomposing urea to the combustion furnace 311.

[0015] The combustion furnace 311 is equipped with a camera 51 and a sensor 52. The camera 51 photographs the inside of the combustion furnace 311. The sensor 52 measures some physical quantity generated in the combustion furnace 311. For example, the sensor 52 measures temperature, pressure, vibration, or sound. The camera 51 or sensor 52 is provided in each piece of equipment included in the plant 3. For example, the dryer 32 is equipped with a sensor 52 to measure the moisture content of the dried sludge, the conveyor belt 33 is equipped with a camera 51 to photograph the waste or biomass being conveyed, and the fermentation bed 34 is equipped with a sensor 52 to measure the concentration of the generated methane gas. For example, the feedwater pump 36 is equipped with a camera 51 to observe water leakage or a sensor 52 to measure temperature. For example, the bag filter 37 is equipped with a sensor 52 to measure pressure loss, and the urea decomposition reactor 38 is equipped with a sensor 52 to measure pressure.

[0016] The cameras 51 and sensors 52 installed on each piece of equipment are connected to the information processing device 1. For example, each camera 51 and sensor 52 is connected to the information processing device 1 via a communication network. For example, the information processing system 100 includes a control device (not shown) that controls the plant 3, and each camera 51 and sensor 52 is connected to the information processing device 1 via the control device. The configuration of the plant 3 shown in Figure 2 is just one example, and the plant 3 may include equipment not shown in Figure 2, or may not include any of the equipment shown in Figure 2. For example, the plant 3 may be in a configuration that does not generate electricity.

[0017] FIG. 3 is a block diagram showing a configuration example inside the information processing apparatus 1. The information processing apparatus 1 is configured using a computer such as a server apparatus. The information processing apparatus 1 executes an information processing method. The information processing apparatus 1 includes an arithmetic unit 11, a memory 12, a storage unit 13, a reading unit 14, an input / output unit 15, and a communication unit 16. The arithmetic unit 11 is a processor and is configured using, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or a multi-core CPU. The arithmetic unit 11 may be configured using a quantum computer. The memory 12 stores temporary data generated during arithmetic operations. The memory 12 is, for example, a RAM (Random Access Memory). The storage unit 13 is non-volatile and is, for example, a hard disk or a non-volatile semiconductor memory. The reading unit 14 reads information from a recording medium 10 such as an optical disk or a portable memory.

[0018] The input / output unit 15 is connected to the plant 3 and performs input / output of data to / from the plant 3. For example, the input / output unit 15 is an interface. The information processing apparatus 1 receives data from cameras 51 and sensors 52 provided in each device included in the plant 3 at the input / output unit 15. The communication unit 16 communicates with the outside of the information processing apparatus 1 by wired communication or wireless communication. Specifically, the communication unit 16 communicates with the terminal device 2. The communication unit 16 may also serve as the input / output unit 15.

[0019] The arithmetic unit 11 causes the reading unit 14 to read a computer program (program product) 131 recorded on the recording medium 10 and stores the read computer program 131 in the storage unit 13. The arithmetic unit 11 executes processing for realizing the functions of the information processing apparatus 1 according to the computer program 131. The computer program 131 may be stored in the storage unit 13 in advance or may be downloaded from outside the information processing apparatus 1. In this case, the information processing apparatus 1 may not include the reading unit 14.

[0020] The computer program 131 can be deployed on a single computer, located at a single site, or distributed across multiple sites and run on multiple computers interconnected by a communication network. That is, the information processing device 1 may consist of multiple computers, and the computer program 131 may run on multiple computers connected via a communication network. The information processing device 1 may also be configured using a cloud server.

[0021] The processing steps described below for executing the information processing method can be performed on multiple computers. The processing steps can also be performed on different computers. The data used during processing may be stored on multiple computers. The processing steps can also be performed using a virtual machine. The processing steps may be performed by multiple arithmetic units. The processing steps may also be performed by different arithmetic units. For example, part of the processing may be performed on one computer, and other parts on other computers.

[0022] The information processing device 1 is equipped with a language model 132. The language model 132 is a pre-trained model that, when given a prompt containing an instruction sentence requesting some kind of response and information attached to the prompt, outputs a sentence corresponding to the input prompt and information. For example, the language model 132 is a large-scale language model. More specifically, the language model 132 is BERT, GPT-4, Bard, or LLaMA, etc. The language model 132 is pre-trained.

[0023] The language model 132 is realized by the arithmetic unit 11 executing information processing according to the computer program 131. The storage unit 13 stores data for realizing the language model 132. The language model 132 may be configured using dedicated hardware other than the arithmetic unit 11 and the storage unit 13. The language model 132 may be realized using a quantum computer. Alternatively, the language model 132 may be provided outside the information processing device 1, and the information processing device 1 may execute processing using the external language model 132. The language model 132 may be realized using multiple computers connected via a communication network, or it may be realized using a cloud.

[0024] The memory unit 13 stores a knowledge database 133 that records information referenced by the language model 132. The knowledge database 133 records predetermined plant data for plant 3, operating history which is a history of data generated in connection with the operation of plant 3, and case data including examples of troubles in plant 3.

[0025] For example, plant data includes design drawings, specifications, and instruction manuals for Plant 3, or drawings, specifications, and instruction manuals for each piece of equipment included in Plant 3, etc., relating to the design or specifications of Plant 3. For example, plant data includes operation manuals, operating procedures, procedures for starting up and shutting down Plant 3, or maintenance plans and procedures, etc., relating to the operation or maintenance of Plant 3. For example, plant data includes troubleshooting guides, or methods or procedures for dealing with problems when they occur, relating to troubleshooting Plant 3. For example, plant data includes safety and regulatory data, such as safety procedures, regulatory compliance guides, emergency response plans, environmental protection regulations, or information on emission standards.

[0026] For example, the operation history includes logs of the operation or management of Plant 3, such as a history of operation data showing the history of Plant 3's operation, a history of operations performed by worker 4, Plant 3's operation log, logs of data from cameras 51 and sensors 52, trouble reports, or a history of abnormal events. For example, the operation history includes a list of consumables or spare parts such as lubricants. For example, the operation history includes maintenance records or history. Knowledge DB 133 may record operation histories for multiple Plants 3. The data recorded in Knowledge DB 133 is vectorized. Knowledge DB 133 may also record plant data, operation history, and case data for Plant 3 other than the Plant 3 that worker 4 is responsible for.

[0027] For example, the case data may include a list of troubles in Plant 3 and countermeasures for those troubles, or a list of troubles and countermeasures for each piece of equipment included in Plant 3. For example, the case data may include examples of troubles that have occurred in the past and the countermeasures taken for those troubles. For example, the case data may include a list of customer inquiries and answers to those inquiries. Knowledge DB133 may record case data for multiple Plant 3s.

[0028] Figure 4 is a block diagram showing an example of the internal configuration of terminal device 2. Terminal device 2 is used by worker 4 and is a computer such as a personal computer, tablet computer, or smartphone. Terminal device 2 comprises an arithmetic unit 21, memory 22, storage unit 23, operation unit 24, display unit 25, and communication unit 26. The arithmetic unit 21 is a processor and is configured using, for example, a CPU, GPU, or multi-core CPU. The arithmetic unit 21 may also be configured using a quantum computer. Memory 22 stores temporary data generated in connection with calculations. Memory 22 is, for example, RAM. Storage unit 23 is non-volatile and is, for example, a hard disk or non-volatile semiconductor memory.

[0029] The operation unit 24 accepts input of information such as text by receiving operations from the worker 4. The operation unit 24 is, for example, a touch panel, keyboard, or pointing device. The display unit 25 displays images. The display unit 25 can display images including text. The display unit 25 is, for example, a liquid crystal display or an EL display (Electroluminescent Display). The operation unit 24 and the display unit 25 may be integrated. The communication unit 26 communicates with the outside of the terminal device 2 by wired or wireless communication. Specifically, the communication unit 26 communicates with the information processing device 1.

[0030] The storage unit 23 stores the computer program 231. For example, the computer program 231 is downloaded using the communication unit 26 and stored in the storage unit 23. For example, the computer program 231 may be stored in the storage unit 23 in advance. For example, the computer program 231 may be read to the terminal device 2 via an interface unit (not shown) and stored in the storage unit 23. The arithmetic unit 21 executes processing according to the computer program 231.

[0031] Worker 4 is a person who operates Plant 3, a person who monitors the operation of Plant 3, or a person who performs maintenance on Plant 3. The information processing system 100 receives queries from worker 4 regarding troubles in Plant 3 and processes to provide worker 4 with countermeasures for the troubles. Figure 5 is a flowchart of an example of the procedure of processing that the information processing device 1 performs to provide countermeasures for troubles. Hereinafter, steps will be abbreviated as S. The information processing device 1 performs the following processing by having the arithmetic unit 11 perform information processing according to the computer program 131.

[0032] If any trouble occurs in Plant 3, worker 4 inputs a query regarding the trouble in Plant 3 by operating the control unit 24. In terminal device 2, the calculation unit 21 transmits the input query to information processing device 1 via the communication unit 26. Information processing device 1 obtains the query regarding the trouble in Plant 3 (S11). In S11, the calculation unit 11 obtains the query regarding the trouble in Plant 3 by receiving the query transmitted from terminal device 2 via the communication unit 16. Worker 4 may input a query regarding a trouble in a Plant 3 other than the Plant 3 he is in charge of, and in S11, information processing device 1 may obtain a query regarding a trouble in a different Plant 3.

[0033] The information processing device 1 acquires sensor data obtained from the camera 51 or sensor 52, operation data representing the history of the operation of plant 3, or input data entered by the worker 4 regarding the status of plant 3 (S12). In S12, the information processing device 1 acquires sensor data by receiving images measured by the camera 51 or data measured by the sensor 52 such as temperature, pressure, vibration, or sound at the input / output unit 15.

[0034] The operation data is data that represents the history of the operation of Plant 3, such as data that associates the time with the content of the processing being performed at Plant 3 at that time, and the information processing device 1 acquires the operation data from Plant 3. For example, operation data is transmitted from the control device that controls Plant 3 to the information processing device 1, and the information processing device 1 receives the operation data with the input / output unit 15. The input data is data that is input from the worker 4 to the terminal device 2 regarding the status of Plant 3. For example, the input data is numerical values ​​shown on instruments installed in Plant 3, which the worker 4 inputs to the terminal device 2 by operating the operation unit 24. For example, the terminal device 2 has an imaging unit (not shown), and the input data is an image taken by the imaging unit of the numerical values ​​shown on instruments installed in Plant 3. The input data is transmitted from the terminal device 2 to the information processing device 1, and the information processing device 1 acquires the input data by receiving the input data with the communication unit 16. The information processing device 1 may also acquire sensor data, operation data, or input data related to another plant.

[0035] The information processing device 1 then creates a prompt requesting countermeasures for the trouble (S13). In S13, the calculation unit 11 creates a prompt that includes the acquired query, the acquired sensor data, operation data or input data, and the acquisition and use of necessary data from the knowledge DB 133. The calculation unit 11 searches the knowledge DB 133 for data containing similar content to the acquired query, or data containing keywords similar to the keywords included in the query, and includes in the prompt the acquisition and use of the retrieved data from the knowledge DB 133. The calculation unit 11 searches the knowledge DB 133 for data using a vector search. The calculation unit 11 may also include in the prompt the referencing of other data specified by a link associated with the retrieved data. The calculation unit 11 may also search for data relating to another plant. The calculation unit 11 may also include in the prompt information specifying the data to be used from the data recorded in the knowledge DB 133. The calculation unit 11 may also include in the prompt the retrieved data.

[0036] The information processing device 1 then uses the language model 132 to obtain countermeasures for the problem (S14). In S14, the calculation unit 11 adds data from the knowledge DB 133 to the created prompt and inputs the prompt to the language model 132. The calculation unit 11 causes the language model 132 to execute processing according to the prompt. The language model 132 performs calculations according to the input prompt and outputs countermeasures for the problem. At this time, the language model 132 may perform calculations using specified data from the data recorded in the knowledge DB 133, or it may perform calculations using all the data recorded in the knowledge DB 133. The calculation unit 11 obtains the countermeasures output by the language model 132. The obtained countermeasures for the problem consist of, for example, text. The countermeasures may include data extracted from the knowledge DB 133, such as diagrams or documents. The calculation unit 11 stores the obtained countermeasures in the memory 12 or storage unit 13.

[0037] The information processing device 1 then outputs the acquired countermeasures (S15). In S15, the calculation unit 11 transmits the acquired countermeasures to the terminal device 2 via the communication unit 16. The terminal device 2 receives the countermeasures transmitted from the information processing device 1 via the communication unit 26. The calculation unit 21 generates an image including the received countermeasures and displays the generated image on the display unit 25. This outputs the countermeasures for the trouble in plant 3. After S15 is completed, the information processing device 1 terminates processing.

[0038] Worker 4 refers to the outputted countermeasures and addresses the trouble in Plant 3. If necessary, Worker 4 may re-enter the query into Terminal Device 2, and processes S11 to S15 may be executed again. Processes S11 to S15 are executed as needed. The information processing device 1 may omit process S12.

[0039] This section presents several examples of countermeasures provided for problems in Plant 3. Examples of problems with the conveyor belt 33 In plant 3, which incinerates biomass or waste, if a problem occurs with the transport conveyor 33, the processes S11 to S15 are executed. In S11, the following query is obtained. Query: The motor on the conveyor belt stopped due to overload. Please tell me how to fix this. In S12, the current value measured by the sensor 52 installed on the conveyor belt 33 is acquired as sensor data. In S13, the following prompt is created. Prompt: You are a veteran operator of a biomass (or waste) incineration facility. Based on Knowledge Database 133, provide solutions for the following: {The conveyor motor has stopped due to overload. Please advise on how to resolve this issue.} and {Current Value}.

[0040] The data used by the language model 132 may be specified from the data recorded in the knowledge DB 133. For example, the data used by the language model 132 may be the history of the current value of the conveyor belt 33, the procedures for handling abnormalities in the conveyor belt 33, the details of past troubles and examples of countermeasures, or records of past maintenance. In the procedures for handling abnormalities in the conveyor belt 33, the details of the abnormality in the conveyor belt 33, the estimated cause of the abnormality, and the actions to be taken are recorded in association with each other.

[0041] In S14, the following countermeasures are taken and output in S15. Countermeasures: The following are possible causes: (1) Foreign object caught in the conveying mechanism. (2) Foreign object caught in the drive mechanism. (3) Chain tension is too high. Chain tension has been checked from the previous maintenance records. Upon checking the conveyor data, the motor current value has been stable since the last maintenance. However, the current value has shown a tendency to rise instantaneously over the past week or so. Therefore, it is suspected that the amount of foreign object in the fuel is increasing. Please check the conveying mechanism for any foreign objects.

[0042] • Examples of problems with water supply pumps (36 types) In plant 3, a problem occurred with the feedwater pump 36. In S11, the following query is obtained: Query: The bearing temperature of the water supply pump is abnormally high. Please tell me how to deal with this. In S12, the temperature value measured by the sensor 52 installed on the water supply pump 36 is acquired as sensor data. In S13, a prompt is created using the query and sensor data. For example, the data used by the language model 132 may be the installation record of the water supply pump 36 during construction, the procedure for when the water supply pump 36 malfunctions, the details of past troubles and examples of countermeasures, the records of past maintenance, or a list of lubricants. In the procedure for when the water supply pump 36 malfunctions, the details of the abnormality of the conveyor belt 33, the estimated cause of the abnormality, and the actions to be taken are recorded in association with each other.

[0043] In S14, the following countermeasures are taken and output in S15. Countermeasures: Similar problems have occurred at other plants, and the top two causes were (1) insufficient lubrication oil and (2) poor installation precision of the feedwater pump. We checked the installation records for feedwater pump 36, and there were no abnormalities in the installation dimensions, etc. Therefore, we believe that insufficient lubrication oil is the cause. Please check the lubrication oil level and increase it if necessary. The last time the lubrication oil was changed was one year ago. The lubrication oil is "Daphne Turbine Oil".

[0044] Examples of steam piping problems A problem occurred in the steam piping at Plant 3. In S11, the following query is obtained: Query: There is a steam leak from the flange of a steam pipe. How should I address this? In S12, the sensor data may be an image of the steam piping taken by the camera 51, and the input data may be data indicating the location of a steam leak. In S13, a prompt is created using a query. For example, the data used by the language model 132 may be a drawing of the plant 3 containing information on the position of the steam piping flanges and the thermal expansion of the steam piping, the calculation result of the thermal expansion of the steam piping, or examples of past troubles and countermeasures.

[0045] In S14, the following countermeasures are taken and output in S15. Countermeasures: Steam leaks can occur due to (1) loose flange bolts or (2) gasket deterioration. Steam piping undergoes thermal expansion, so the U-bolts on the pipe support should be loosened, but there have been cases where they were fixed in place. Therefore, please check that the position of the pipe support where the steam is being generated is not fixed. Then, tighten the bolts. The countermeasures may include drawings such as diagrams showing the location of pipe supports. Alternatively, in response to a prompt, the language model 132 may provide instructions to input the location of the steam leak, and a prompt including the location of the steam leak may be entered.

[0046] Examples of problems with dryers (32 models) A problem occurred with the dryer 32 that dries the sludge. In S11, the following query is obtained. Query: The drying process in the dryer is working very well, and the dried sludge is very fine and free-flowing. It seems okay, but what kind of problems can be expected if this continues? Also, please tell me how to return the drying process to normal.

[0047] In S14, the following countermeasures are taken and output in S15. Countermeasures: Problems around the dryer and combustion furnace are possible. Around the dryer, there have been cases of dust explosions due to over-drying of sludge. Also, since over-drying means that excess energy is being input, it is not desirable to continue operation from an energy conservation perspective. In the combustion furnace, if the sludge is burned in a loose state, combustion will occur immediately inside the furnace, resulting in explosive combustion. In addition, suspended combustion tends to occur, making combustion unstable (generating NOx or CO, etc.). Furthermore, a large amount of dust flows to the subsequent equipment, soot is collected by the bag filter, and pressure loss increases. Methods to return to the dry state include (1) reducing the rotation speed of the dryer and (2) reducing the temperature of the supplied steam.

[0048] • Example of investigating the degree of deterioration of combustion furnace 311 In S11, a query is obtained to inquire about the degree of deterioration of the internal walls of the combustion furnace 311. In S12, an image of the inside of the combustion furnace 311 taken by the camera 51 is obtained as sensor data. In S13, a prompt is created that includes the obtained query and image. The language model 132 outputs countermeasures including the degree of deterioration of the walls of the combustion furnace 311 in response to the prompt, and in S14, the countermeasures including the degree of deterioration of the walls of the combustion furnace 311 are obtained. The worker 4 can then know the degree of deterioration of the internal walls of the combustion furnace 311 and take appropriate measures such as repairing or replacing the internal walls of the combustion furnace 311 according to the degree of deterioration.

[0049] • Example of investigating clinker inside combustion furnace 311 Clinker is solidified ash produced by combustion. Clinker adheres to the bottom of the combustion furnace 311. In S11, a query is obtained to inquire about the amount of clinker adhering to the combustion furnace 311. The query may also include an inquiry about the processing to be performed if the amount of clinker adhering is large. In S12, an image of the clinker inside the combustion furnace 311, taken by camera 51, is obtained as sensor data. In S13, a prompt is created that includes the obtained query and image. In response to the prompt, the language model 132 outputs countermeasures including the amount of clinker adhering in the combustion furnace 311, and in S14, the countermeasures including the amount of clinker adhering is obtained. Worker 4 can know the amount of clinker adhering in the combustion furnace 311 and take appropriate measures according to the amount of clinker adhering, such as removing the clinker from the combustion furnace 311 if the amount is large.

[0050] The degree of deterioration or the amount of clinker buildup on the internal walls of the combustion furnace 311 may be determined using a pre-trained model separate from the language model 132. This pre-trained model is pre-trained to output the degree of deterioration or the amount of clinker buildup on the walls when an image of the inside of the combustion furnace 311 is input. The degree of deterioration or the amount of clinker buildup on the internal walls of the combustion furnace 311 can be expressed as a numerical value, for example. In S13, the information processing device 1 inputs an image of the inside of the combustion furnace 311 to the pre-trained model, obtains the degree of deterioration or the amount of clinker buildup on the walls output by the pre-trained model, and creates a prompt that includes the obtained degree of deterioration or the amount of clinker buildup on the walls. In S14, the information processing device 1 obtains countermeasures from the language model 132 that correspond to the prompt that includes the degree of deterioration or the amount of clinker buildup on the walls.

[0051] Furthermore, the information processing device 1 can also investigate the degree of deterioration of the walls or the state of the combustion furnace 311 other than the clinker. For example, the information processing device 1 may be configured to investigate the burn-out point, which is the position where the fuel burns completely inside the combustion furnace 311. In this configuration, a query to inquire about the burn-out point is obtained in S11. In S12, an image of the inside of the combustion furnace 311 taken by the camera 51 is obtained as sensor data. In S13, a prompt including the obtained query and image is created. The language model 132 outputs countermeasures including the burn-out point in response to the prompt, and in S14, the countermeasures including the burn-out point are obtained. The burn-out point may also be determined using a pre-trained model other than the language model 132. The pre-trained model is pre-trained to output the burn-out point when an image of the inside of the combustion furnace 311 is input. In S13, the information processing device 1 inputs the image to the pre-trained model, obtains the burn-out point output by the pre-trained model, and creates a prompt including the obtained burn-out point. In S14, the information processing device 1 obtains countermeasures corresponding to the prompt, including the burnout point, from the language model 132.

[0052] As described above, the information processing device 1, in response to queries regarding plant 3 troubles, retrieves countermeasures for the troubles from the language model 132 using the knowledge DB 133 and outputs the countermeasures. By using the language model 132 and the knowledge DB 133, appropriate countermeasures for troubles can be output. Any worker 4, including inexperienced workers, can obtain appropriate countermeasures from the information processing device 1 when a plant 3 trouble occurs, and by implementing the obtained countermeasures, can deal with the trouble appropriately. In other words, any worker 4 can deal with troubles in the same way. For this reason, plant 5 can continue to operate stably.

[0053] The information processing system 100 may perform processing to control plant 3 in response to a query. Figure 6 is a flowchart showing an example of the procedure for processing performed by the information processing device 1 to control plant 3. Worker 4 inputs a query to terminal device 2 regarding a problem in plant 3, including a request to perform control on plant 3. The calculation unit 21 of terminal device 2 transmits the input query to the information processing device 1 via the communication unit 26, and the information processing device 1 receives the query regarding the problem in plant 3 (S21). The information processing device 1 acquires sensor data, operation data, or input data, similar to S12 (S22). The information processing device 1 creates a prompt, similar to S13 (S23).

[0054] The information processing device 1 then uses the language model 132 to obtain countermeasures for the trouble, including the control content of plant 3 (S24). In response to a prompt containing a query that includes a request to perform control on plant 3, the language model 132 outputs countermeasures including the control content of plant 3, and the arithmetic unit 11 obtains the outputted countermeasures. The arithmetic unit 11 stores the obtained countermeasures in the memory 12 or storage unit 13.

[0055] Next, the information processing device 1 outputs the acquired countermeasures (S25). In S25, the information processing device 1 sends the acquired countermeasures to the terminal device 2, and the terminal device 2 displays the countermeasures, including the control content, on the display unit 25. The worker 4 confirms the countermeasures, including the control content, and inputs approval of the control content to the terminal device 2 by operating the operation unit 24. The terminal device 2 sends the approval of the control content to the information processing device 1, and the information processing device 1 obtains approval of the control content by receiving the approval. Note that the language model 132 may output multiple control content options, and the information processing device 1 may obtain approval of the control content by accepting the selection of any of the multiple control content options. For example, multiple control content options with different target temperatures may be output, and one of them may be selected. The information processing device 1 may also receive instructions to modify the control content included in the output countermeasures using the terminal device 2, and modify the control content according to the received instructions. The information processing device 1 can also obtain disapproval of the control content. If approval is not obtained, the information processing device 1 will not execute any processing from S26 onwards.

[0056] The information processing device 1 outputs a control command to plant 3 corresponding to the control content upon obtaining approval for the control content (S26). In S26, the arithmetic unit 11 generates a control command to control plant 3 according to the control content and transmits the generated control command to plant 3 from the input / output unit 15. Plant 3 performs control according to the control command. For example, a control command is input to a control device included in plant 3, the control device performs control according to the control command, and plant 3 executes the processing defined in the control content. After S26 is completed, the information processing device 1 terminates its processing.

[0057] The processes S21 to S26 are executed as needed. The processes S21 to S26 control plant 3 in response to queries from worker 4. Necessary control is performed on plant 3 depending on the trouble that occurs. Note that the information processing device 1 may omit process S22. The information processing device 1 may refrain from executing process S26 if it does not obtain approval for the control content. Alternatively, the information processing device 1 may perform process S26 without requiring approval for the control content.

[0058] The information processing system 100 automatically acquires data related to the plant 3 and performs processing to output recommendations for the worker 4. Figure 7 is a flowchart showing an example of the procedure for processing performed by the information processing device 1 to output recommendations. The information processing device 1 acquires sensor data, operation data, or input data (S31). For example, sensor data or operation data is periodically transmitted from the plant 3 to the information processing device 1, and the information processing device 1 receives the sensor data or operation data in the input / output unit 15. For example, input data is periodically input from the worker 4 to the terminal device 2, and the information processing device 1 receives the input data transmitted from the terminal device 2. The calculation unit 11 stores the acquired sensor data, operation data, or input data in the memory 12 or storage unit 13. The calculation unit 11 may also record the acquired sensor data, operation data, or input data in the knowledge DB 133.

[0059] The information processing device 1 creates a prompt requesting a recommendation based on the acquired sensor data, operation data, or input data (S32). In S32, the calculation unit 11 creates a prompt that includes a recommendation request and instructions to acquire and use the acquired sensor data, operation data, or input data, and necessary data from the knowledge DB 133. For example, the calculation unit 11 creates a prompt that includes a recommendation request that includes suggestions for worker 4 or tasks that worker 4 should perform, based on the acquired sensor data, operation data, or input data. The information processing device 1 may acquire sensor data, operation data, or input data for a plant 3 other than the plant 3 that worker 4 is in charge of in S31, and in S32 create a prompt requesting a recommendation based on the sensor data, operation data, or input data for the other plant 3.

[0060] The information processing device 1 then obtains a recommendation using the language model 132 (S33). In S33, the calculation unit 11 adds data from the knowledge DB 133 to the created prompt and inputs the prompt to the language model 132. The calculation unit 11 causes the language model 132 to execute processing according to the prompt. The language model 132 performs calculations according to the input of the prompt and outputs a recommendation. The calculation unit 11 obtains the recommendation output by the language model 132. The recommendation consists of, for example, text. The recommendation may also include data extracted from the knowledge DB 133, such as a diagram or document. The calculation unit 11 stores the obtained recommendation in the memory 12 or storage unit 13.

[0061] The information processing device 1 outputs the acquired recommendation (S34). In S34, the calculation unit 11 transmits the acquired recommendation to the terminal device 2 via the communication unit 16. The terminal device 2 receives the recommendation transmitted from the information processing device 1 via the communication unit 26. The calculation unit 21 generates an image containing the received recommendation and displays the generated recommendation on the display unit 25. This outputs a recommendation for worker 4. After S34 is completed, the information processing device 1 terminates processing.

[0062] Worker 4 checks the outputted recommendations and performs the work according to the recommendations. For example, worker 4 may input a query regarding the problem to terminal device 2 according to the recommendations, and processes S11-S15 or S21-S26 may be executed. Processes S31-S34 are executed as needed.

[0063] Several examples are shown where recommendations regarding Plant 3 are provided, and furthermore, countermeasures for problems are provided through processing S11 to S15. Examples of recommendations for water supply pump 36 Sensor data, operation data, or input data are acquired regarding a problem with the feedwater pump 36 at another plant 3, and processing S31 to S34 is executed. In S33, the following recommendation is acquired and output in S34. Recommendation: At another facility, a water leak occurred from the discharge side joint of the boiler feedwater pump. The cause is still under investigation, but pump vibration is a strong possibility. Therefore, please measure and record the pump vibration values ​​at this plant as well. Please also record video and audio information.

[0064] Worker 4, after reviewing the recommendation, inspects the water supply pump 36 according to the recommendation. He also measures the vibration value of the water supply pump 36 and inputs it into terminal device 2, and further inputs video and audio information into terminal device 2. Additionally, worker 4 inputs a query into terminal device 2, and processes S11 to S15 are executed. In S11, the following query is obtained. Query: Please collect and organize the following information from the knowledge database and display it. Information from sources other than the knowledge database is not needed, so general statements and external data are prohibited. (1) What is the progress of the investigation into the cause at other facilities? (2) Please find information related to "water leakage on the pump discharge side" in past trouble cases. (3) Please list the causes of pipe water leakage trouble other than vibration in bullet points. (4) If there are any investigation results from other facilities nationwide, please display them.

[0065] In S14, the following countermeasures are taken and output in S15. countermeasure: (1) No new information regarding the cause of the problem at other facilities has been registered in the database. Shall we contact the maintenance department? (2) Similar cases in past disputes are as follows: [Document 1] [Document 2] [Document 3] (3) Other than vibration, the following are possible causes of piping problems: Construction error: Since five years have passed since completion, this is considered unlikely. Material selection error: Please check the piping material on the discharge side of the boiler feedwater pump. (4) The results of the survey at other business locations nationwide are as follows: [Document 4]

[0066] Examples of recommendations for the urea decomposition reactor 38 Processing S31 to S34 is executed in response to the acquisition of input data related to Plant 3. Regarding the urea decomposition reactor 38, the following recommendation is acquired in S33 and output in S34. Recommendation: Based on the registered operation log information, it has been found that the catalyst regeneration interval in the urea decomposition reactor is becoming shorter. Furthermore, an increase in reactor pressure has also been observed. Please review the data for the past three years below. [Data for the past three years] Related problems have occurred in other cases, so please also check the information on those cases. [Trouble Example 1] [Trouble Example 2]

[0067] Worker 4, after reviewing the recommendations, checks the materials according to the recommendations. Worker 4 also enters a query into terminal device 2, and processes S11 to S15 are executed. In S11, the following query is obtained. Query: I have reviewed the documents and data. I do believe that this plant may indeed have a higher risk of similar problems occurring as in other projects. Please provide the investigation records from other projects. Please also provide what data should be recorded around the reactor and what countermeasures should be taken.

[0068] In S14, the following countermeasures are taken and output in S15. Countermeasures: (1) The investigation records for other cases are as follows: [Investigation Record 1] [Investigation Record 2] (2) The following should be recorded as part of the initial investigation of the reactor trouble: <Pressure in and around the reactor. Water quality analysis of the urea solution. Check for blockage of the urea spray nozzle> After checking these items as soon as possible, please contact the maintenance department as needed. (3) The next scheduled maintenance is planned for two months from now, at which time please perform an internal check. The following arrangements are necessary: ​​<Cleaning of the catalyst, gravel, nozzle (if blocked), and inside of the reactor>

[0069] • Example of a recommendation related to Bug Filter 37 Processing S31 to S34 is executed in response to the acquisition of sensor data or operational data related to Plant 3. Regarding the bug filter 37, the following recommendations are obtained in S33 and output in S34. Recommendation: The pressure loss of the bag filter, which was normally around 0.8 kPa, has risen to 1.0 kPa since yesterday evening. The amount of gas must be within the specified value based on the operating data. The current value of the discharge conveyor is also fine. The amount of BF recovered ash has not increased. That has been confirmed. Therefore, please confirm the following at the site. Pressure gauge (PG) reading for backwash compressed air (pulse air) Bag filter differential pressure gauge (PG) reading

[0070] After reviewing the recommendation, worker 4 checks the pressure gauge reading according to the recommendation, inputs a query including the pressure gauge reading into terminal device 2, and processes S11 to S15 are executed. A query including an image of the pressure gauge may also be input. In S11, the following query is obtained. Query: The pressure gauge reading for backwash compressed air is 0.2 kPa, and the bag filter differential pressure gauge reading is 0.95 kPa.

[0071] In S14, the following countermeasures are taken and output in S15. Countermeasures: The differential pressure of the bag filter matches the readings on the field pressure gauge and the sensor, so there is no problem with the sensor. Also, the pressure of the compressed air used for backwashing is the same as during the last field inspection. If the differential pressure increases further, increase the pressure of the compressed air used for backwashing to 0.25 MPa.

[0072] Examples of recommendations regarding fermentation layer 34 Processing S31 to S34 is executed in response to the acquisition of sensor data or operational data related to Plant 3. Regarding the fermentation layer 34, the following recommendation is acquired in S33 and output in S34. Recommendation: Last week, the methane gas concentration in biogas was 58%, but it has now dropped to 53%. Biogas per unit of waste input is also 150 Nm³. 3 The value was / ton, but it changed to 130Nm 3 The level has dropped to [number] hours. This suggests that methane fermentation is being inhibited or the load is too high. Please take the following actions. Please collect sludge from the inlet and outlet of the fermentation tank and measure its pH. Please measure the ammonia concentration in the sludge at the outlet of the fermentation tank.

[0073] After reviewing the recommendations, worker 4 measures the pH of the sludge and the ammonia concentration according to the recommendations. Furthermore, worker 4 inputs a query containing the measurement results into terminal device 2, and processes S11 to S15 are executed. In S11, the following query is obtained. Query: Fermenter pH is 7.1 at the inlet, 7.5 in the middle, and 7.9 at the outlet. Ammonia concentration at the outlet is 2200 mg / L.

[0074] In S14, the following countermeasures are taken and output in S15. Countermeasures: The ammonia concentration was 2300 mg / L in the previous analysis and has remained below 2500 mg / L, which inhibits fermentation, so there is no problem. The pH at the inlet was 7.3 last time and is trending downwards. It is suspected that the load is slightly higher than usual. Reduce the amount of waste input by approximately 80% and check the situation for 2-3 days to see if the methane gas concentration recovers.

[0075] As described above, the information processing device 1 obtains recommendations for worker 4 from the language model 132 using the knowledge DB 133 according to the status of plant 3, and outputs the recommendations. By using the language model 132 and the knowledge DB 133, appropriate recommendations for the status of plant 3 can be output. Worker 4 can operate plant 3 appropriately by performing tasks related to plant 3 according to the recommendations. Since any worker 4 can operate plant 3 appropriately, plant 5 can continue to operate stably.

[0076] The present invention is not limited to the embodiments described above, and various modifications are possible within the scope of the claims. That is, embodiments obtained by combining technical means that have been appropriately modified within the scope of the claims are also included in the technical scope of the present invention.

[0077] The matters described in each embodiment can be combined with each other. Furthermore, the independent and dependent claims described in the claims can be combined with each other in any combination, regardless of the form of reference. Moreover, although the claims do not use a form in which claims referencing two or more other claims (multi-claim form), they are not limited to this. They may be described using a multi-claim form, or a form in which multi-claims referencing at least one multi-claim (multi-multi-claim). [Explanation of Symbols]

[0078] 100 Information Processing Systems 1. Information Processing Device 10 Recording media 11 Arithmetic section 13 Storage section 131 Computer Programs 132 Language Models 133 Knowledge DB 2 Terminal devices 3 Plants 4 Workers 51 Camera 52 sensors

Claims

1. We retrieve queries regarding problems at plants that incinerate biomass, waste, or sludge, or burn gases generated from biomass, waste, or sludge. Using a language model that records knowledge about the aforementioned plant, countermeasures for the aforementioned trouble are obtained in response to the acquired query. Output the acquired countermeasures. An information processing method characterized by the following:

2. The system acquires sensor data obtained from sensors installed in the plant, operational data representing the history of the plant's operation, or input data entered by people regarding the plant's status. Using the language model, the countermeasures corresponding to the acquired sensor data, driving data, or input data and the query are obtained. The information processing method according to feature 1.

3. The aforementioned sensor data includes temperature, pressure, vibration, or sound from equipment included in the plant. The information processing method according to feature 2.

4. As the sensor data or input data, an image representing the status of the plant is acquired. Obtain the acquired image and the countermeasure corresponding to the query. The information processing method according to feature 2.

5. To represent the status of the aforementioned plant, an image of the inside of the combustion furnace is acquired. The information processing method according to feature 4.

6. Using the language model, the countermeasures are obtained, including images of the inside of the combustion furnace and the degree of deterioration of the furnace wall, corresponding to the query. The information processing method according to feature 5.

7. As an image taken inside the combustion furnace, an image of the clinker inside the combustion furnace is obtained. Using the language model, images of the clinker inside the combustion furnace and the countermeasures, including the amount of clinker attached, corresponding to the query are obtained. The information processing method according to feature 5.

8. Using the language model, the countermeasures, including control content for the equipment included in the plant, are obtained. A control command corresponding to the control content is output to the device. The information processing method according to feature 1.

9. The database records data defined for the plant, the history of data generated during the operation of the plant, or examples of troubles at the plant. The information processing method according to feature 1.

10. The system acquires sensor data obtained from sensors installed in the plant, operational data representing the history of the plant's operation, or input data entered by people regarding the plant's status. Using the language model, recommendations for the plant are obtained based on the acquired sensor data, operating data, or input data. Output the acquired recommendation. The information processing method according to feature 1.

11. The aforementioned sensor data, operating data, or input data may include data relating to a plant other than the plant to which the recommendation is made. The information processing method according to feature 10.

12. Equipped with a calculation unit, The aforementioned arithmetic unit, We retrieve queries regarding problems at plants that incinerate biomass, waste, or sludge, or burn gases generated from biomass, waste, or sludge. Using a language model that records knowledge about the aforementioned plant, countermeasures for the aforementioned trouble are obtained in response to the acquired query. Output the acquired countermeasures. An information processing device characterized by the following:

13. We retrieve queries regarding problems at plants that incinerate biomass, waste, or sludge, or burn gases generated from biomass, waste, or sludge. Using a language model that records knowledge about the aforementioned plant, countermeasures for the aforementioned trouble are obtained in response to the acquired query. Output the acquired countermeasures. A computer program characterized by causing a computer to perform a process.

Images