Electromagnetic transient fault analysis methods, devices, and equipment based on large models

Abstract

This application discloses a method, apparatus, and equipment for electromagnetic transient fault analysis based on a large-scale model. The method includes: preprocessing real-time operational information obtained from the power grid to obtain time-series operational information; performing fault reasoning analysis based on a pre-set fault analysis large-scale model and a thought chain technique according to the time-series operational information to obtain fault analysis results, including fault location, fault type, and fault root cause; and analyzing the cascading effects of the fault based on the spatiotemporal feedback reasoning chain in the pre-set fault analysis large-scale model, combined with power grid topology information and the fault analysis results, to obtain fault impact analysis results. This application can solve the technical problems of insufficient analysis accuracy, large computational load, and difficulty in conducting impact analysis and evaluation for various types of faults in existing technologies.

Description

Technical Field

[0001] This application relates to the field of power grid fault analysis technology, and in particular to electromagnetic transient fault analysis methods, devices and equipment based on large models. Background Technology

[0002] Dynamic fault analysis of electromagnetic transients in power grids studies the electromagnetic transient phenomena of power systems during faults, including instantaneous changes in voltage and current and their impact on system stability. Electromagnetic transient fault processes typically occur between milliseconds and seconds, exhibiting transient, high-frequency, and nonlinear characteristics, and can potentially impact and damage power equipment and system operation.

[0003] Existing power grid electromagnetic transient fault analysis mainly relies on traditional numerical calculations and predefined models. These methods are prone to problems such as insufficient accuracy, large computational load, and inability to achieve real-time response during the analysis process. Moreover, power grid faults are diverse in type and have a wide range of impacts, making it difficult for existing technologies to accurately analyze the impact of faults. As a result, existing power grid electromagnetic transient solid-state analysis schemes are difficult to meet the application requirements of real-world scenarios. Summary of the Invention

[0004] This application provides a method, apparatus, and equipment for electromagnetic transient fault analysis based on a large model, which addresses the technical problems of insufficient analysis accuracy, large computational load, and difficulty in conducting impact analysis and evaluation for various types of faults in existing technologies.

[0005] In view of this, the first aspect of this application provides a method for electromagnetic transient fault analysis based on a large model, including:

[0006] Preprocessing operations are performed on the real-time operation information obtained from the power grid to obtain time-series operation information;

[0007] Based on a pre-set fault analysis model and thinking chain technology, fault reasoning analysis is performed according to the time-series operation information to obtain fault analysis results, including fault location, fault type and fault root cause.

[0008] Based on the spatiotemporal feedback inference chain in the preset fault analysis model, and combined with the power grid topology information and the fault analysis results, the chain effects of the fault are analyzed to obtain the fault impact analysis results.

[0009] Preferably, the preprocessing operation of the real-time operation information obtained from the power grid to obtain time-series operation information includes:

[0010] Obtain real-time operational information from the power grid's sensors and measurement systems;

[0011] The real-time operation information is cleaned, denoised, outlier detected, and standardized to obtain time-series operation information.

[0012] Preferably, the fault reasoning analysis based on the preset fault analysis model and the thinking chain technology, according to the time-series operation information, to obtain the fault analysis results, includes:

[0013] The fault detection mechanism based on the thinking chain technology in the pre-set fault analysis model performs fault detection and analysis based on the time-series operation information, and uses a multi-layer analysis mechanism to determine the fault location and fault type.

[0014] The root cause of the failure is analyzed using the fault root cause analysis mechanism based on the aforementioned thinking chain technology, and the root cause of the failure is obtained.

[0015] Preferably, the step of analyzing the cascading effects of a fault based on the spatiotemporal feedback inference chain in the preset fault analysis model, combined with power grid topology information and the fault analysis results, to obtain fault impact analysis results, further includes:

[0016] Based on the fault analysis results and the fault impact analysis results, electromagnetic transient faults at future times are predicted to obtain fault prediction results;

[0017] Early warning information is generated based on the fault prediction results, and fault warnings are issued.

[0018] Preferably, the step of analyzing the cascading effects of a fault based on the spatiotemporal feedback inference chain in the preset fault analysis model, combined with power grid topology information and the fault analysis results, to obtain fault impact analysis results, further includes:

[0019] The power grid dispatch strategy is optimized by combining the real-time power grid operation status and the results of the fault impact analysis, resulting in an optimized power grid strategy.

[0020] The second aspect of this application provides an electromagnetic transient fault analysis device based on a large model, comprising:

[0021] The information processing unit is used to preprocess the real-time operation information obtained from the power grid to obtain time-series operation information.

[0022] The fault analysis unit is used to perform fault reasoning analysis based on the time-series operation information using a preset fault analysis model and thinking chain technology to obtain fault analysis results, including fault location, fault type and fault root cause.

[0023] The impact analysis unit is used to analyze the chain effects of the fault based on the spatiotemporal feedback inference chain in the preset fault analysis model, combined with the power grid topology information and the fault analysis results, and to obtain the fault impact analysis results.

[0024] Preferably, the fault analysis unit is specifically used for:

[0025] The fault detection mechanism based on the thinking chain technology in the pre-set fault analysis model performs fault detection and analysis based on the time-series operation information, and uses a multi-layer analysis mechanism to determine the fault location and fault type.

[0026] The root cause of the failure is analyzed using the fault root cause analysis mechanism based on the aforementioned thinking chain technology, and the root cause of the failure is obtained.

[0027] Preferably, it further includes:

[0028] The fault prediction unit is used to predict electromagnetic transient faults at future times based on the fault analysis results and the fault impact analysis results, and obtain fault prediction results.

[0029] The early warning generation unit is used to generate early warning information based on the fault prediction results and to issue fault warnings.

[0030] Preferably, it further includes:

[0031] The strategy optimization unit is used to optimize the power grid dispatch strategy by combining the real-time power grid operating status and the fault impact analysis results, so as to obtain an optimized power grid strategy.

[0032] A third aspect of this application provides an electromagnetic transient fault analysis device based on a large model, the device including a processor and a memory;

[0033] The memory is used to store program code and transmit the program code to the processor;

[0034] The processor is used to execute the electromagnetic transient fault analysis method based on a large model as described in the first aspect, according to the instructions in the program code.

[0035] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:

[0036] This application provides a method for electromagnetic transient fault analysis based on a large model, including: preprocessing real-time operating information obtained from the power grid to obtain time-series operating information; performing fault reasoning analysis based on a pre-set fault analysis large model and thinking chain technology according to the time-series operating information to obtain fault analysis results, including fault location, fault type, and fault root cause; and analyzing the chain effects of the fault based on the spatiotemporal feedback reasoning chain in the pre-set fault analysis large model, combined with power grid topology information and fault analysis results, to obtain fault impact analysis results.

[0037] The electromagnetic transient fault analysis method based on a large model provided in this application uses a pre-constructed large-scale fault analysis model to perform a comprehensive analysis of the fault. This process incorporates chain-of-thought techniques and a feedback mechanism, ensuring that the fault analysis at each stage is derived step-by-step, making the analysis process interpretable and the results more accurate and reliable. Moreover, this process relies on a trained large model, so the actual computational load is relatively small. Furthermore, this process can perform cascading effect analysis on various types of faults, and since it is based on the feedback inference chain in the large model, the accuracy of the analysis results can be ensured. Therefore, this application can solve the technical problems of insufficient analysis accuracy, large computational load, and difficulty in performing impact analysis and evaluation on various types of faults in existing technologies. Attached Figure Description

[0038] Figure 1 A flowchart illustrating the electromagnetic transient fault analysis method based on a large model provided in this application embodiment;

[0039] Figure 2 A schematic diagram of the structure of the electromagnetic transient fault analysis device based on a large model provided in the embodiments of this application. Detailed Implementation

[0040] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0041] Terminology Explanation:

[0042] Thinking Chain: A logical reasoning-based analytical method that breaks down complex problems into multiple logically related steps, gradually deriving a solution.

[0043] Electromagnetic transients refer to the phenomenon where, under normal operating conditions, the state of a power system changes significantly within a short period of time due to external interference or internal faults.

[0044] Dynamic faults refer to instantaneous or short-term faults that occur in a power system, such as lightning strikes and short circuits.

[0045] For easier understanding, please refer to Figure 1 The embodiments of the electromagnetic transient fault analysis method based on a large model provided in this application include:

[0046] Step 101: Perform preprocessing operations on the real-time operation information obtained from the power grid to obtain time-series operation information.

[0047] Further, step 101 includes:

[0048] Obtain real-time operational information from the power grid's sensors and measurement systems;

[0049] The real-time operation information is cleaned, denoised, outlier detected, and standardized to obtain time-series operation information.

[0050] It should be noted that the process of acquiring data from the power grid is not limited to power grid equipment; measurement systems include PMUs or SCADA systems, as long as the relevant data can be obtained. Furthermore, the data acquisition stage can also acquire all the data that needs to be extracted from the power grid in this embodiment. Although only real-time operational information is mentioned here, it can also include subsequent power grid topology information, real-time power grid operating status, and operational or historical data required for model training, etc., which will not be elaborated further. It is understood that real-time operational information includes, but is not limited to, key electromagnetic parameters such as voltage, current, and frequency.

[0051] To improve the accuracy and reliability of fault analysis, this embodiment performs unified preprocessing operations on all acquired data, specifically including, but not limited to, data cleaning, noise reduction, outlier detection, and standardization. Data cleaning removes noise and invalid data, which can be achieved using filtering algorithms such as Kalman filtering and low-pass filtering. Noise reduction reduces electromagnetic interference, ensuring data accuracy. Outlier detection uses statistical analysis or machine learning algorithms to identify abnormal data values; machine learning algorithms include Isolation Forest and DBSCAN. Data format standardization unifies the data format, ensuring compatibility with data from different sources.

[0052] Understandably, the acquired and processed data is presented in time series form and can be directly input into subsequent large models for fault analysis. Because the data has undergone various preprocessing operations, it is high-quality data, which can ensure the accuracy and reliability of fault analysis based on this data.

[0053] Step 102: Based on the preset fault analysis model and thinking chain technology, perform fault reasoning analysis according to the time sequence operation information to obtain the fault analysis results, which include fault location, fault type and fault root cause.

[0054] Further, step 102 includes:

[0055] The fault detection mechanism based on thinking chain technology in the pre-set fault analysis model performs fault detection and analysis based on time-series operation information, and uses a multi-layer analysis mechanism to determine the fault location and fault type.

[0056] The root cause of a failure is analyzed using a fault root cause analysis mechanism based on thinking chain technology, and the root cause of the failure is obtained.

[0057] The pre-defined fault analysis model in this embodiment employs a feedback mechanism and introduces a thought chain technique. This enables temporal and spatial feedback and reasoning of power grid data, ensuring that the correlation characteristics between related data can be analyzed step by step, thereby guaranteeing the reliability of fault analysis. The pre-defined fault analysis model in this embodiment can be built based on multi-layer neural networks or graph neural networks, such as convolutional neural networks and recurrent neural networks. Models designed on these network architectures can accurately capture the spatiotemporal characteristics of data information.

[0058] The entire pre-designed fault analysis model involves multiple stages, such as fault detection, fault location, fault classification, root cause analysis, and subsequent cascading effect analysis. The chain-of-thinking technique can break down the fault analysis process into these stages, which can form a chain-like reasoning structure. This step-by-step deductive analysis method makes the model more interpretable and the analysis process more reliable.

[0059] The pre-defined fault analysis model in this embodiment can be pre-learned using historical fault data through supervised learning or reinforcement learning. Through reinforcement learning, the inference chain can be continuously optimized based on fault feedback, improving the model's adaptability and analytical accuracy. Cross-validation is used to optimize model parameters, further enhancing the model's analytical accuracy. Specific model evaluation can be achieved based on metrics such as accuracy, recall, and F1 score, followed by hyperparameter tuning.

[0060] In this embodiment, the time-series operational information input to the large model is real-time information. After input, a fault detection analysis is first performed to determine whether a fault exists in the power grid. If a fault exists, a multi-layered analysis mechanism in the thought process chain analyzes the location and type of the fault, thus obtaining the specific fault location and type. The subsequent root cause analysis mechanism further analyzes the root cause of the fault based on the detected fault location and type, thereby obtaining the root cause. It is understood that the large model can also, as needed, combine electromagnetic transient characteristics with the above fault analysis results to perform a brief analysis of the fault severity. The electromagnetic transient characteristics can be calculated using existing technology, generally within the time range of 20ms-100ms. This is helpful for subsequent fault impact analysis; the specific process is not limited here and can be set as needed.

[0061] Step 103: Based on the spatiotemporal feedback inference chain in the preset fault analysis model, and combined with the power grid topology information and fault analysis results, analyze the chain effects of the fault to obtain the fault impact analysis results.

[0062] The spatiotemporal feedback inference chain in the pre-defined fault analysis model has already been mentioned in the fault analysis phase. It combines concepts such as thought chain, feedback mechanism and neural network to perform time and space feedback inference analysis on the data, ensuring that all related information in the fault process can be taken into account, thereby ensuring the reliability of the chain effect analysis.

[0063] The cascading effect analysis in this embodiment mainly includes two parts: basic impact analysis and cascading reaction prediction. Basic impact analysis primarily analyzes the dynamic impact of a fault on load distribution, voltage stability, frequency, and other aspects based on fault type and grid topology information. Cascading reaction prediction, on the other hand, predicts the potential subsequent impacts of a fault, providing theoretical guidance for subsequent grid operation and maintenance.

[0064] Furthermore, step 103, followed by:

[0065] Based on the fault analysis results and the fault impact analysis results, the electromagnetic transient faults at future times are predicted, and the fault prediction results are obtained.

[0066] Early warning information is generated based on the fault prediction results, and fault warnings are issued.

[0067] The fault prediction process in this embodiment involves comprehensively analyzing existing fault information and predicting potential future faults. This part can be implemented based on a graph neural network model, ensuring prediction accuracy. Predicted faults include, but are not limited to, overload, voltage fluctuations, voltage instability, and load transfer. Different early warning messages can be generated based on different prediction results, providing early warnings to power grid dispatchers to take preventative measures and improve system stability.

[0068] Furthermore, step 103, followed by:

[0069] The power grid dispatch strategy is optimized by combining the real-time power grid operation status and fault impact analysis results, resulting in an optimized power grid strategy.

[0070] The fault impact analysis results of this embodiment can be used for subsequent power grid dispatch strategy optimization. The strategy optimization proposed in this embodiment is only one application scenario of the fault analysis results. The fault analysis results can also be applied to other scenarios according to the actual situation, without any specific limitations.

[0071] In addition to relying on the results of fault impact analysis, the optimization process also needs to incorporate analysis of the real-time power grid operating status. This process can generate optimized power grid strategies, such as specifying switching devices and adjusting generator output; furthermore, it can develop detailed fault handling plans to ensure stable power grid operation.

[0072] The optimized power grid strategy generated in this embodiment can also be transformed into specific automated control commands and sent to power grid equipment for dynamic adjustment. A real-time multi-verification mechanism ensures the security and accuracy of the control commands. By directly adjusting power grid equipment through automated control commands, real-time dynamic response is achieved. The multi-verification mechanism ensures the security and accuracy of command execution, preventing misoperation. This closed-loop control from analysis to automated execution enhances the efficiency and security of fault handling. Therefore, the fault analysis method provided in this embodiment can provide optimized decision support for the stable operation of the power grid, improve the overall reliability of the power grid through optimized scheduling strategies, and reduce outage time due to faults.

[0073] Understandably, deploying the electromagnetic transient fault analysis method of this embodiment in high-voltage transmission lines allows for real-time monitoring of line operation. Through the large-scale model thinking chain construction module, the system can identify transient faults caused by lightning strikes and predict their potential impact on downstream power grids. The system automatically generates switching strategies to disconnect affected lines, preventing the fault from spreading to the entire power grid, while simultaneously adjusting the load distribution of other lines to ensure the continuity and stability of power supply.

[0074] The electromagnetic transient fault analysis method based on a large model provided in this application uses a pre-constructed large-scale fault analysis model to perform a comprehensive analysis of the fault. This process incorporates chain-of-thought techniques and a feedback mechanism, ensuring that the fault analysis at each stage is derived step-by-step, making the analysis process interpretable and the results more accurate and reliable. Furthermore, this process relies on a trained large model, so the actual computational load is relatively small. In addition, this process can perform cascading effect analysis on various types of faults, and since it is based on the feedback inference chain within the large model, the accuracy of the analysis results is ensured. Therefore, this application's embodiments can solve the technical problems of insufficient analysis accuracy, high computational load, and difficulty in performing impact analysis and evaluation on various types of faults in existing technologies.

[0075] For easier understanding, please refer to Figure 2 This application provides an embodiment of an electromagnetic transient fault analysis device based on a large model, including:

[0076] Information processing unit 201 is used to preprocess real-time operation information obtained from the power grid to obtain time-series operation information;

[0077] The fault analysis unit 202 is used to perform fault reasoning analysis based on a preset fault analysis model and thinking chain technology according to the time-series operation information to obtain fault analysis results, including fault location, fault type and fault root cause.

[0078] The impact analysis unit 203 is used to analyze the chain effects of faults based on the spatiotemporal feedback inference chain in the preset fault analysis model, combined with the power grid topology information and fault analysis results, and obtain the fault impact analysis results.

[0079] Furthermore, the fault analysis unit 202 is specifically used for:

[0080] The fault detection mechanism based on thinking chain technology in the pre-set fault analysis model performs fault detection and analysis based on time-series operation information, and uses a multi-layer analysis mechanism to determine the fault location and fault type.

[0081] The root cause of a failure is analyzed using a fault root cause analysis mechanism based on thinking chain technology, and the root cause of the failure is obtained.

[0082] Furthermore, it also includes:

[0083] The fault prediction unit 204 is used to predict electromagnetic transient faults at future times based on fault analysis results and fault impact analysis results, and obtain fault prediction results.

[0084] The early warning generation unit 205 is used to generate early warning information based on the fault prediction results and to issue fault warnings.

[0085] Furthermore, it also includes:

[0086] The strategy optimization unit 206 is used to optimize the power grid dispatch strategy by combining the real-time power grid operation status and fault impact analysis results, so as to obtain an optimized power grid strategy.

[0087] This application also provides an electromagnetic transient fault analysis device based on a large model, the device including a processor and a memory;

[0088] The memory is used to store program code and transfer the program code to the processor;

[0089] The processor is used to execute the electromagnetic transient fault analysis method based on a large model in the above method embodiments according to the instructions in the program code.

[0090] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0091] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0092] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0093] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for executing all or part of the steps of the methods described in the various embodiments of this application through a computer device (which may be a personal computer, server, or network device, etc.). The aforementioned storage medium includes: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program code.

[0094] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for electromagnetic transient fault analysis based on a large model, characterized in that, include: Preprocessing operations are performed on the real-time operation information obtained from the power grid to obtain time-series operation information; Based on a pre-defined fault analysis model and thinking chain technology, fault reasoning analysis is performed according to the time-series operational information to obtain fault analysis results. The fault analysis results include fault location, fault type, and fault root cause. The specific process includes: The fault detection mechanism based on the thinking chain technology in the pre-set fault analysis model performs fault detection and analysis based on the time-series operation information, and uses a multi-layer analysis mechanism to determine the fault location and fault type. The root cause of the failure is analyzed using the fault root cause analysis mechanism based on the aforementioned thinking chain technology to obtain the root cause of the failure. Based on the spatiotemporal feedback inference chain in the preset fault analysis model, and combined with the power grid topology information and the fault analysis results, the chain effect of the fault is analyzed to obtain the fault impact analysis results; Chain reaction analysis includes two parts: basic impact analysis and chain reaction prediction. Basic impact analysis is based on fault type and power grid topology information to analyze the dynamic impact of faults on load distribution, voltage stability and frequency. Chain reaction prediction is to predict the subsequent impacts that faults may cause. Based on the graph neural network model, the electromagnetic transient faults at future times are predicted according to the fault analysis results and the fault impact analysis results, and the fault prediction results are obtained. Early warning information is generated based on the fault prediction results, and fault warnings are issued.

2. The electromagnetic transient fault analysis method based on a large model according to claim 1, characterized in that, The preprocessing operation of the real-time operation information obtained from the power grid to obtain time-series operation information includes: Obtain real-time operational information from the power grid's sensors and measurement systems; The real-time operation information is cleaned, denoised, outlier detected, and standardized to obtain time-series operation information.

3. The electromagnetic transient fault analysis method based on a large model according to claim 1, characterized in that, The step involves analyzing the cascading effects of a fault based on the spatiotemporal feedback inference chain in the preset fault analysis model, combined with power grid topology information and the fault analysis results, to obtain fault impact analysis results. This process further includes: The power grid dispatch strategy is optimized by combining the real-time power grid operation status and the results of the fault impact analysis, resulting in an optimized power grid strategy.

4. An electromagnetic transient fault analysis device based on a large model, characterized in that, include: The information processing unit is used to preprocess the real-time operation information obtained from the power grid to obtain time-series operation information. The fault analysis unit is used to perform fault reasoning analysis based on a preset fault analysis model and thinking chain technology according to the time-series operation information, and obtain fault analysis results. The fault analysis results include fault location, fault type, and fault root cause. Specifically, the fault analysis unit is used for: The fault detection mechanism based on the thinking chain technology in the pre-set fault analysis model performs fault detection and analysis based on the time-series operation information, and uses a multi-layer analysis mechanism to determine the fault location and fault type. The root cause of the failure is analyzed using the fault root cause analysis mechanism based on the aforementioned thinking chain technology to obtain the root cause of the failure. The impact analysis unit is used to analyze the chain effects of the fault based on the spatiotemporal feedback inference chain in the preset fault analysis model, combined with the power grid topology information and the fault analysis results, and to obtain the fault impact analysis results. Chain reaction analysis includes two parts: basic impact analysis and chain reaction prediction. Basic impact analysis is based on fault type and power grid topology information to analyze the dynamic impact of faults on load distribution, voltage stability and frequency. Chain reaction prediction is to predict the subsequent impacts that faults may cause. The fault prediction unit is used to predict electromagnetic transient faults at future times based on the fault analysis results and the fault impact analysis results using a graph neural network model, and to obtain fault prediction results. The early warning generation unit is used to generate early warning information based on the fault prediction results and to issue fault warnings.

5. The electromagnetic transient fault analysis device based on a large model according to claim 4, characterized in that, Also includes: The strategy optimization unit is used to optimize the power grid dispatch strategy by combining the real-time power grid operating status and the fault impact analysis results, so as to obtain an optimized power grid strategy.

6. An electromagnetic transient fault analysis device based on a large model, characterized in that, The device includes a processor and a memory; The memory is used to store program code and transmit the program code to the processor; The processor is used to execute the electromagnetic transient fault analysis method based on a large model as described in any one of claims 1-3 according to the instructions in the program code.

Images