Thermal power production equipment fault cause diagnosis method and system

Abstract

The invention relates to the technical field of artificial intelligence and thermal power production, in particular to a fault cause diagnosis method and system for thermal power production equipment. The method comprises the steps of obtaining a thermal power plant measuring point time series data abnormal result; inputting a thermal power plant measuring point time sequence data anomaly result into the trained fault diagnosis big language model, analyzing input measuring point data anomaly based on the trained fault diagnosis big language model, analyzing and reasoning possible fault types, reasons, processing suggestions and later maintenance suggestions according to measuring point meanings and types, and performing fault diagnosis. And generating a diagnosis report, wherein the diagnosis report at least comprises the fault type, the fault generation reason and the fault processing suggestion. According to the method, the user is assisted to provide decision suggestions, and the user friendliness and the interpretability of professional tasks are improved.

Description

Technical Field

[0001] The present invention relates to the technical fields of artificial intelligence and thermal power production, and specifically relates to a method and system for diagnosing the causes of faults in thermal power production equipment. Background Art

[0002] Currently, during the process of diagnosing faults in thermal power production, it is usually relied on manual reasoning and judgment of the causes of production equipment faults by combining the actual situation at the fault site, and then solutions and prevention and control measures suggestions are obtained. In recent years, the application of deep learning time series anomaly detection models has begun to bring benefits to the power generation industry, but the detection results of time series models are not natural language and are not easy for non-professionals to understand. If the manual professional experience is insufficient or experts cannot arrive at the scene for assistance in a timely manner, it is easy to have a deviation in understanding the anomaly detection results, resulting in missed detections and misjudgments of equipment, causing losses in time or economic benefits. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method and system for diagnosing the causes of faults in thermal power production equipment for solving the technical problems that the on-site personnel in the process of power generation production lack professionalism and cannot efficiently confirm the fault type and root cause through the multi-measurement point detection results of the time series model.

[0004] The object of the present invention is achieved by the following technical solutions: In the first aspect, the present invention provides a method for diagnosing the causes of faults in thermal power production equipment, including: Obtaining the abnormal results of the time series data of the measuring points in the thermal power plant; Inputting the abnormal results of the time series data of the measuring points in the thermal power plant into a trained large language model for fault diagnosis. Based on the trained large language model for fault diagnosis, analyzing the input data, and analyzing and reasoning out the possible fault types, causes, handling suggestions, and later maintenance suggestions according to the meanings and types of the measuring points, and generating a diagnostic report, where the diagnostic report at least includes the fault type, the cause of the fault, and the fault handling suggestions; The large language model for fault diagnosis is trained using a preprocessed expert knowledge base of thermal power plant equipment.

[0005] As a further improvement of the present invention, the abnormal results of the time series data of the measuring points in the thermal power plant are obtained according to a pre-trained time series anomaly detection model.

[0006] As a further improvement of the present invention, during the training process of the fault diagnosis model, it also includes fine-tuning the fault diagnosis model using a low-load fine-tuning method.

[0007] As a further improvement of the present invention, the low-load fine-tuning method adopts the LOMO or LoRA method.

[0008] As a further improvement of the present invention, during the training process of the fault diagnosis model, it further includes constructing a large model Q&A system using retrieval enhancement technology, and obtaining a diagnosis report using the large model Q&A system.

[0009] As a further improvement of the present invention, the expert knowledge base of thermal power plant equipment at least includes a historical fault document library of thermal power plant production equipment, a fault case record file in the thermal power plant production equipment database, a fault analysis report of thermal power production equipment written by experts, a maintenance manual for thermal power production equipment, and industry standards and specification documents of the thermal power plant industry.

[0010] As a further improvement of the present invention, the preprocessing steps of the expert knowledge base of thermal power plant equipment include: proofreading and organizing the data in the knowledge base, text cleaning, and editing and segmentation processing.

[0011] In a second aspect, the present invention provides a fault cause diagnosis system for thermal power production equipment, which is used to implement the above-mentioned fault cause diagnosis method for thermal power production equipment, and includes: A data acquisition module, which is used to acquire the abnormal results of the time series data of the measuring points in the thermal power plant; A fault cause acquisition module, which inputs the abnormal results of the time series data of the measuring points in the thermal power plant into the trained fault diagnosis model, analyzes the input data based on the trained fault diagnosis model, and analyzes and infers the possible fault types, causes, handling suggestions, and later maintenance suggestions according to the meaning and type of the measuring points, and generates a diagnosis report, and the diagnosis report at least includes the fault type, the cause of the fault, and the fault handling suggestions; The fault diagnosis model is trained using the preprocessed expert knowledge base of thermal power plant equipment.

[0012] In a third aspect, the present invention provides a computer-readable storage medium storing one or more programs, and the one or more programs include instructions, and when the instructions are executed by a computing device, the computing device is caused to execute the above-mentioned fault cause diagnosis method for thermal power production equipment.

[0013] In a fourth aspect, the present invention provides a computing device, including: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include steps for executing the above-mentioned fault cause diagnosis method for thermal power production equipment.

[0014] The beneficial effects of the present invention are as follows: The method for diagnosing the causes of faults in thermal power production equipment provided by the present invention can quickly identify possible fault types and causes through the analysis of measured point time-series data by the fault diagnosis model, which helps to reduce the time of traditional manual troubleshooting. Moreover, by combining the fault diagnosis model with the expert knowledge base for fault diagnosis, the accuracy and efficiency of diagnosis can be greatly improved. Trained with the preprocessed expert knowledge base of thermal power plant equipment, the fault diagnosis model can comprehensively consider various factors and provide more comprehensive and accurate fault diagnosis results. For operators or maintenance personnel without profound professional knowledge, this method can provide easily understandable explanatory texts to help maintenance personnel quickly master the fault situation and take corresponding measures. Through the analysis of the automated fault diagnosis model, the dependence on manual experience can be reduced, and the labor cost and error rate can be lowered.

[0015] Furthermore, the present invention applies the large language model for fault diagnosis across domains in the process of analyzing the causes of faults in thermal power production equipment. Although the fault diagnosis model does not directly participate in the algorithms related to unit fault diagnosis of the deep learning time-series model, it can assist users in providing decision-making suggestions, improving user-friendliness and the interpretability of professional tasks. This cross-domain application demonstrates the potential value of the fault diagnosis model in dealing with complex industrial problems.

[0016] Furthermore, through low-load fine-tuning, the model can more accurately identify fault types, analyze fault causes, and provide more precise processing and maintenance suggestions. Low-load fine-tuning allows the model to quickly adapt to new data and new changes without the need to retrain the entire model. Description of the Drawings

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0018] Figure 1 It is a schematic flow chart of the method for diagnosing the causes of faults in thermal power production equipment in the embodiments of the present invention; Figure 2 It is a schematic structural diagram of the system for diagnosing the causes of faults in thermal power production equipment in the embodiments of the present invention; Figure 3 It is a schematic diagram of an electronic device in the embodiments of the present invention. Detailed Embodiments

[0019] To make the objectives and technical solutions of the present invention clearer and easier to understand, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. The specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0020] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings and specific embodiments. Among them, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0021] Embodiment 1 As Figure 1 shown, this embodiment provides a method for diagnosing the causes of faults in thermal power production equipment. The following are the specific implementation manners.

[0022] First, obtain the abnormal results of the time-series data of the measuring points in the thermal power plant.

[0023] Among them, the abnormal results of the time-series data of the measuring points in the thermal power plant are obtained according to the trained time-series anomaly detection model. The time-series anomaly detection model is an existing deep learning model. After obtaining the abnormal results of the measuring point time-series data detected by the time-series anomaly detection model, such as image fluctuations. The fault diagnosis model can be used to generate explanatory text, describing possible abnormal causes, impacts, and recommended response measures.

[0024] Input the abnormal results of the time-series data of the measuring points in the thermal power plant into the trained fault diagnosis model. Based on the trained fault diagnosis model, analyze the input data, and analyze and infer the possible fault types, causes, handling suggestions, and later maintenance suggestions according to the meaning and type of the measuring points, and generate a diagnostic report. The diagnostic report includes at least the fault type, the cause of the fault, and the fault handling suggestions. The fault diagnosis model uses a large language model.

[0025] The fault diagnosis model can integrate knowledge from multiple sources (such as equipment manuals, historical fault records, professional papers, etc.), and use this information to provide in-depth analysis of the detected anomalies described by the user. The model can infer the correlation between the abnormal phenomena of the detection results of different measuring points in the user's description, and explain why certain abnormal changes may indicate specific types of equipment faults.

[0026] In this embodiment, an expert knowledge base of thermal power plant equipment is used to train the fault diagnosis model. The expert knowledge base of thermal power plant equipment at least includes a historical fault document library of thermal power plant production equipment, a fault case record file in the thermal power plant production equipment database, a fault analysis report of thermal power production equipment written by experts, a maintenance manual of thermal power production equipment, and thermal power plant industry standards and specification documents. Specifically, a large number of materials such as historical fault case records, expert analysis reports, maintenance manuals, equipment operation logs, power generation industry technical manuals and operation guides, safety regulations and standards, engineering and technical papers, industry standards and specifications in the thermal power production field are collected.

[0027] The preprocessing steps of the expert knowledge base of thermal power plant equipment include: proofreading and collating the data in the knowledge base, text cleaning, and editing and segmenting.

[0028] Specifically, proofreading and collating mainly include: verifying the sources of all documents to ensure they come from reliable channels such as official maintenance manuals, historical fault records, industry standards, etc. Comparing the same information in different documents to ensure there are no contradictions. For example, the descriptions of a certain fault phenomenon in different versions of the maintenance manual of the same equipment should be consistent. Identifying and correcting spelling mistakes, grammar mistakes, and obvious factual mistakes. This can be done through a combination of automated tools and manual review.

[0029] Text cleaning mainly includes: filtering out irrelevant information to make the text content in the knowledge base clear. For example, deleting the header, footer, comments, watermarks, etc. in the document and retaining the core information. Converting documents in different formats (such as PDF, Word, HTML) into a unified plain text format for subsequent processing.

[0030] Editing and segmenting is mainly used to divide large chunks of text into smaller and more meaningful parts for easy model understanding and processing. First, divide the document into chapters and paragraphs according to the logical structure. For example, the fault case record can be divided according to the time sequence, location, and equipment type involved in the event; mark the key information points in the text, such as fault phenomena, cause analysis, treatment measures, etc. This can be achieved using markup languages (such as Markdown, XML, or JSON). Then classify the documents according to factors such as fault type and equipment type for subsequent retrieval and matching. For example, classify all faults related to the electrical system into one category and those related to the mechanical system into another category. Finally, create an index for each document or paragraph for quick searching and reference. The index can be based on keywords, topic tags, etc.

[0031] In addition, during the training process of the fault diagnosis model in this embodiment, it also includes fine-tuning the fault diagnosis model using the low-load fine-tuning method. The low-load fine-tuning method uses the LOMO or LoRA method.

[0032] Alternatively, it is also possible to choose not to use the large model fine-tuning technology. Based on the collected expert knowledge materials, a large model question-answering system is constructed using the RAG method (Retrieval Augmented Generation). A diagnostic report is obtained using the large model question-answering system.

[0033] Based on the input information of on-site personnel, the fault diagnosis model quickly analyzes and infers the possible fault types, causes, handling suggestions, and subsequent maintenance suggestions according to the meaning and type of measurement points. By generating a diagnostic report, the diagnostic report can help on-site personnel quickly locate the root cause of the problem, reduce misjudgment, and improve work efficiency. The fault diagnosis large language model in this embodiment has the ability of continuous learning and can continuously update the model knowledge base according to actual feedback and newly emerging faults to ensure the continuous improvement of the large language model's ability and improve diagnostic accuracy. Collect the misjudgment and missed detection situations of the large language model to optimize the large model and improve the accuracy and efficiency of fault cause analysis.

[0034] Embodiment 2 Retrieve case records in advance from the historical fault document library or database of thermal power professional enterprises. These records should contain detailed information such as the time and location of the fault, equipment type, fault phenomenon, handling process and results, etc. At the same time, collect a large amount of corpora such as analysis reports written by experts, equipment maintenance manuals, industry standards and specifications, etc., and perform preprocessing work such as proofreading and collation, text cleaning, editing and segmentation to facilitate effective learning of the large model. Classification can be carried out according to factors such as fault type and equipment type to facilitate subsequent model learning and matching.

[0035] Select a fault diagnosis model as the basic model. For example, in order to achieve effective fine-tuning under limited computing resources, use low-load fine-tuning methods such as LOMO and LoRA to perform low-load fine-tuning on the fault diagnosis model. According to the corpus characteristics of the thermal power production field, set reasonable fine-tuning parameters to ensure that the model can achieve efficient learning under limited computing resources.

[0036] On-site personnel observe the detection results of the deep learning time-series data anomaly detection model, check which measurement points show abnormal results, and input them in natural language to the fault diagnosis model. For example: "I have obtained data of several measurement points related to the induced draft fan failure in a certain power plant. Among them, it is found that the flue gas temperature at the outlet of induced draft fan A, the flue gas pressure at the outlet of electrostatic precipitator A, the flue gas pressure at the outlet of induced draft fan A, the horizontal vibration of the bearing of induced draft fan A, the vertical vibration of the bearing of induced draft fan A, the temperature 1 of the non-drive end bearing of induced draft fan A, the temperature 2 of the non-drive end bearing of induced draft fan A, the temperature 3 of the non-drive end bearing of induced draft fan A, the temperature 4 of the middle bearing of induced draft fan A, the temperature 5 of the middle bearing of induced draft fan A, the temperature 6 of the middle bearing of induced draft fan A, the temperature 7 of the drive end bearing of induced draft fan A, the temperature 8 of the drive end bearing of induced draft fan A, the temperature 9 of the drive end bearing of induced draft fan A, the temperature 1 of the stator winding A of the motor of induced draft fan A, the temperature 2 of the stator winding A of the motor of induced draft fan A, the temperature 1 of the stator winding B of the motor of induced draft fan A, the temperature 2 of the stator winding B of the motor of induced draft fan A, the temperature 1 of the stator winding C of the motor of induced draft fan A, the temperature 2 of the stator winding C of the motor of induced draft fan A have significantly higher anomaly scores than the threshold value during a certain period, and the true values are significantly lower than the predicted values. Please analyze what possible faults may have occurred in this power plant?" Based on the input information from on-site personnel, the fault diagnosis model quickly analyzes and matches the possible fault types, causes, handling suggestions, and later maintenance suggestions. A diagnostic report is generated to facilitate providing reference for on-site personnel, quickly understanding the problem, and taking corresponding handling measures. Such as: "According to the information you provided, the anomalies in the data points of the induced draft fan and its related components (such as flue gas temperature, pressure, and the temperature of bearings and motors, etc.) may indicate problems in multiple aspects. The significant changes in these data indicators are usually closely related to the operating state of the equipment, and these abnormal manifestations may imply the following faults or problems: Mechanical faults: The anomalies in the flue gas temperature at the outlet of induced draft fan A and the flue gas pressure at the outlet of electrostatic precipitator A, as well as the increase in bearing vibration levels (horizontal, vertical), may be due to mechanical faults such as blade damage, wear, or rotor imbalance. These situations will change the hydrodynamic characteristics of the fluid passing through the equipment, thus affecting the performance of the above parameters.

[0037] Temperature anomalies: The temperature data of the bearings and stator windings of induced draft fan A being lower than expected and significantly deviating from the normal range may mean problems such as overly strong cooling effect or errors in the monitoring instruments themselves. But it is more likely to indicate that the actual working state of some components (such as insufficient lubrication) is not good, or the true value is lower than the historical data on which the prediction model is based because the equipment has stopped running.

[0038] Electrical system problems: The abnormal temperature of the stator windings of the motor may also imply faults in the electrical system, such as unstable power supply, too high or too low current, etc., which will also affect the working efficiency and stability of the induced draft fan.

[0039] Combined with the above analysis, the specific possible causes of failure include but are not limited to: Damage to the internal parts of the induced draft fan (such as blades) leading to reduced working efficiency. Bearing lubrication problems or severe wear causing increased vibration. Unstable power supply to the electrical system or problems with the motor itself. Abnormal operation of the cooling system or errors in the monitoring equipment.

[0040] Given this information, an on-site inspection is recommended to confirm the specific circumstances.

[0041] For example: Check whether there are signs of damage to the blades and internal structure of the induced draft fan. Carry out a detailed inspection of the bearings, including lubrication and wear. The stability of the electrical system and the operating status of the motor require further diagnosis by professionals. At the same time, consider recalibrating the monitoring equipment or adjusting the parameters of the prediction model to more accurately reflect the current actual situation. " In addition, this embodiment regularly collects the misjudgments and missed detections of the large model, optimizes and upgrades the fault diagnosis model, and improves the judgment ability and adaptability of the model.

[0042] Example 3 like Figure 2 As shown, this embodiment provides a thermal power production equipment fault cause diagnosis system, which is used to implement the thermal power production equipment fault cause diagnosis method of the above-mentioned embodiment 1 and embodiment 2, including: Data acquisition module, used to obtain abnormal results of time series data of thermal power plant measurement points; The fault cause acquisition module inputs the abnormal results of the time series data of the thermal power plant measurement points into the trained fault diagnosis model, analyzes the input data based on the trained fault diagnosis model, analyzes and infers the possible fault type, cause, handling suggestions and subsequent maintenance suggestions according to the meaning and type of the measurement points, and generates a diagnosis report, which at least includes the fault type, cause of the fault, and fault handling suggestions.

[0043] Example 4 In another embodiment of the present invention, the present invention further provides a storage medium, specifically a computer-readable storage medium (Memory). The computer-readable storage medium is a memory device in a terminal device, used to store programs and data. It can be understood that the computer-readable storage medium here can include both the built-in storage medium in the terminal device, and of course, the extended storage medium supported by the terminal device. It can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, device, or component. The computer-readable storage medium provides storage space, and this storage space stores the operating system of the terminal. And, in this storage space, there is also stored one or more instructions suitable for being loaded and executed by a processor, and these instructions can be one or more computer programs (including program codes). It should be noted that more specific examples (non-exhaustive list) of the computer-readable storage medium here include: electrical connections with one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above.

[0044] The computer-readable storage medium also includes data signals propagated in a baseband or as part of a carrier wave, which carry readable program codes. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable storage medium can also be any readable medium other than the readable storage medium, and this readable medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, device, or component. The program codes contained on the readable storage medium can be transmitted by any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above.

[0045] The program code for performing the operations of the present invention can be written in any combination of one or more programming languages, including object-oriented programming languages such as Python, Java, C++, etc., and also including conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, executed as a stand-alone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user's computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (e.g., by connecting through the Internet using an Internet service provider).

[0046] One or more instructions stored in a computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the method for diagnosing the causes of faults in thermal power production equipment in the above embodiments; one or more instructions in the computer-readable storage medium are loaded and executed by the processor to perform the following steps: Obtain the abnormal results of the time-series data of the measuring points in the thermal power plant; Input the abnormal results of the time-series data of the measuring points in the thermal power plant into the trained fault diagnosis model. Based on the trained fault diagnosis model, analyze the input data, and analyze and infer the possible fault types, causes, handling suggestions, and later maintenance suggestions according to the meaning and type of the measuring points, and generate a diagnostic report. The diagnostic report at least includes the fault type, the cause of the fault, and the fault handling suggestions; The fault diagnosis model is trained using a preprocessed expert knowledge base of thermal power plant equipment.

[0047] Embodiment 5 Please refer to Figure 3 , the terminal device is a computer device. The computer device 60 in this embodiment includes: a processor 61, a memory 62, and a computer program 63 stored in the memory 62 and executable on the processor 61. When the computer program 63 is executed by the processor 61, it implements the method for diagnosing the causes of faults in thermal power production equipment in the embodiment. To avoid repetition, it will not be elaborated here one by one. Alternatively, when the computer program 63 is executed by the processor 61, it implements the functions of each model / unit in the computing system of the fault diagnosis model in Embodiment 1. To avoid repetition, it will not be elaborated here one by one.

[0048] The computer device 60 can be a computing device such as a desktop computer, a notebook, a handheld computer, and a cloud server. The computer device 60 may include, but is not limited to, a processor 61 and a memory 62. Those skilled in the art can understand, Figure 3This is only an example of the computer device 60, which does not constitute a limitation on the computer device 60. It may include more or fewer components than those shown in the figure, or combine certain components, or different components. For example, the computer device may also include input / output devices, network access devices, buses, etc.

[0049] The so-called processor 61 may be a central processing unit (CPU), or it may also be other general-purpose processors, central processors, graphics processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, data processing logic units based on quantum computing, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0050] The memory 62 may be an internal storage unit of the computer device 60, such as the hard disk or memory of the computer device 60. The memory 62 may also be an external storage device of the computer device 60, such as a plug-in hard disk equipped on the computer device 60, a smart media card (SMC), a secure digital (SD) card, a flash card, etc.

[0051] Furthermore, the memory 62 may also include both the internal storage unit and the external storage device of the computer device 60. The memory 62 is used to store computer programs and other programs and data required by the computer device. The memory 62 may also be used to temporarily store data that has been output or will be output.

[0052] In each of the embodiments provided by the present application, any reference to a memory, database, or other medium may include at least one of non-volatile and volatile memories. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0053] In each of the embodiments provided by the present application, the database involved may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., without limitation. In each of the embodiments provided by the present application, the processor involved may be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without limitation.

Claims

1. A method for diagnosing the cause of a thermal power production equipment failure, characterized in that: include: Obtain abnormal results of time series data of thermal power plant measurement points; The abnormal results of the time series data of the thermal power plant measurement points are input into the trained fault diagnosis model, and based on the trained fault diagnosis model, the input data is analyzed and inferred according to the significance and type of the measurement points, the possible fault types, causes, handling suggestions and subsequent maintenance suggestions, and a diagnosis report is generated, wherein the diagnosis report at least includes the fault type, the cause of the fault, and the fault handling suggestions; The fault diagnosis model adopts a large language model; the fault diagnosis model is trained by using a preprocessed thermal power plant equipment expert knowledge base.

2. The method for diagnosing the cause of a thermal power production equipment failure according to claim 1, characterized in that: The abnormal results of the time series data of the thermal power plant measurement points are obtained according to a pre-trained time series anomaly detection model.

3. The method for diagnosing the cause of a failure of thermal power production equipment according to claim 1, characterized in that: During the training process of the fault diagnosis model, the fault diagnosis model is further fine-tuned by using a low-load fine-tuning method.

4. The method for diagnosing the cause of a thermal power production equipment failure according to claim 3, characterized in that: The low load fine tuning method adopts the LOMO or LoRA method.

5. The method for diagnosing the cause of a failure of thermal power production equipment according to claim 1, characterized in that: During the training process of the fault diagnosis model, the method further includes constructing a large model question-answering system using retrieval enhancement technology, and obtaining a diagnosis report using the large model question-answering system.

6. The method for diagnosing the cause of a failure of thermal power production equipment according to claim 1, characterized in that: The thermal power plant equipment expert knowledge base at least includes a thermal power plant production equipment historical fault document library, a fault case record file in a thermal power plant production equipment database, a thermal power production equipment fault analysis report written by experts, a thermal power production equipment maintenance manual, and thermal power plant industry standards and industry specification documents.

7. The method for diagnosing the cause of a failure of thermal power production equipment according to claim 6, characterized in that: The preprocessing steps of the thermal power plant equipment expert knowledge base include: proofreading and arranging the data in the knowledge base, text cleaning, and segmentation processing.

8. A thermal power production equipment fault cause diagnosis system, used to implement the thermal power production equipment fault cause diagnosis method according to any one of claims 1 to 7, characterized in that: include: Data acquisition module, used to obtain abnormal results of time series data of thermal power plant measurement points; The fault diagnosis module inputs the abnormal results of the time series data of the thermal power plant measurement points into the trained fault diagnosis model, analyzes the input data based on the trained fault diagnosis model, analyzes and infers the possible fault types, causes, handling suggestions and subsequent maintenance suggestions according to the meaning and type of the measurement points, and generates a diagnosis report, which at least includes the fault type, the cause of the fault, and the fault handling suggestions; The fault diagnosis model is trained using a preprocessed thermal power plant equipment expert knowledge base.

9. A computer-readable storage medium storing one or more programs, characterized in that: The one or more programs include instructions, which, when executed by a computing device, enable the computing device to execute the method for diagnosing causes of failures of thermal power production equipment as described in any one of claims 1 to 7.

10. A computing device, characterized in that include: One or more processors, a memory and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include steps for executing the method for diagnosing the cause of failure of thermal power production equipment as described in any one of claims 1 to 7.

Images