Fault analysis method and device, electronic equipment and medium

Abstract

The embodiment of the invention provides a fault analysis method and device, electronic equipment and a medium, and the method comprises the steps: obtaining collected key data, and determining a fault related index according to the key data; constructing a dynamic rule tree according to the fault related indexes; and obtaining fault list information, and generating fault summary information for the fault list information through a pre-trained target model according to the fault list information and the dynamic rule tree. According to the embodiment of the invention, fault analysis is carried out by using the dynamic rule tree and the model, and the accuracy of fault analysis is improved.

Description

Technical Field

[0001] The present invention relates to the field of operation and maintenance technology, and in particular to a fault analysis method, device, electronic equipment and medium. Background Art

[0002] In power and other systems, intelligent operation and maintenance systems are usually installed to perform fault analysis to assist operation and maintenance personnel in troubleshooting. However, existing fault analysis is usually only performed through simple data comparison, and the accuracy of fault analysis is low. Summary of the Invention

[0003] In view of the above problems, a method, device, electronic device and medium for fault analysis are proposed to overcome the above problems or at least partially solve the above problems, including:

[0004] A fault analysis method, comprising:

[0005] Obtaining collected key data and determining fault-related indicators based on the key data;

[0006] Constructing a dynamic rule tree based on the fault-related indicators;

[0007] Fault ticket information is acquired, and fault summary information for the fault ticket information is generated according to the fault ticket information and the dynamic rule tree through a pre-trained target model.

[0008] Optionally, the generating of fault summary information for the fault ticket information according to the fault ticket information and the dynamic rule tree using the pre-trained target model includes:

[0009] Determine, by using a pre-trained target model, a node matching result of the fault ticket information in the dynamic rule tree; wherein the dynamic rule tree includes a plurality of nodes, and different nodes correspond to different fault causes;

[0010] The pre-trained target model is used to generate fault summary information for the fault ticket information based on the fault ticket information and the node matching result.

[0011] Optionally, determining the node matching result of the trouble ticket information in the dynamic rule tree by using a pre-trained target model includes:

[0012] Through the pre-trained target model, node matching is performed in the dynamic rule tree according to the fault description information in the fault ticket information to obtain a node matching result.

[0013] Optionally, constructing a dynamic rule tree according to the fault-related indicators includes:

[0014] Determine the cause of the target fault based on the fault-related indicators;

[0015] In a preset dynamic rule tree, nodes corresponding to the target fault cause are marked to construct the marked dynamic rule tree.

[0016] Optionally, it also includes:

[0017] The fault summary information is fed back using a preset format.

[0018] Optionally, the fault ticket information includes any one or more of the following: fault description information, fault occurrence time, and fault type.

[0019] Optionally, the fault summary information includes any one or more of the following: fault cause, fault impact scope, and fault handling solution.

[0020] Optionally, the target model is a large language model.

[0021] Optionally, the key data is key data in the power system, and the fault ticket information is fault ticket information for the power system.

[0022] A fault analysis method, comprising:

[0023] Pre-built decision tree models;

[0024] Acquire the collected key data, and determine a feature vector based on the key data;

[0025] Fault ticket information is acquired, and feature vectors corresponding to the fault ticket information are analyzed using the decision tree model to generate fault summary information for the fault ticket information.

[0026] A fault analysis method, comprising:

[0027] Establish fault judgment rules in advance;

[0028] Obtaining collected key data and determining fault-related indicators based on the key data;

[0029] Obtain fault ticket information, and generate fault summary information for the fault ticket information in combination with the fault judgment rule and the fault-related indicators.

[0030] A device for fault analysis, wherein the device is used to implement the fault analysis method described above.

[0031] An electronic device includes a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program implements the fault analysis method described above when executed by the processor.

[0032] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the fault analysis method described above is implemented.

[0033] The embodiments of the present invention have the following advantages:

[0034] In an embodiment of the present invention, by acquiring collected key data and determining fault-related indicators based on the key data, a dynamic rule tree is constructed based on the fault-related indicators to acquire fault ticket information, and through a pre-trained target model, fault summary information for the fault ticket information is generated based on the fault ticket information and the dynamic rule tree, thereby realizing the use of dynamic rule trees and models to perform fault analysis and improving the accuracy of fault analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the description of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0036] Figure 1 is a schematic diagram of a system architecture provided by some embodiments of the present invention;

[0037] Figure 2 is a flowchart of the steps of fault analysis provided by some embodiments of the present invention;

[0038] Figure 3 is a flowchart of the steps of another fault analysis method provided by some embodiments of the present invention;

[0039] Figure 4 is a flowchart of the steps of another fault analysis method provided by some embodiments of the present invention;

[0040] Figure 5 is a flowchart of the steps of another fault analysis method provided by some embodiments of the present invention;

[0041] Figure 6 is a flowchart of the steps of another fault analysis method provided by some embodiments of the present invention;

[0042] Figure 7This is a flowchart of the steps of another fault analysis method provided by some embodiments of the present invention. DETAILED DESCRIPTION

[0043] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments described are only a portion of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are also within the scope of protection of the present invention.

[0044] In an embodiment of the present invention, through the combined application of real-time data collection, calculation of relevant evaluation indicators, dynamic construction of large-model dynamic analysis trees, and intelligent summary of large-model analysis results, intelligent judgment and analysis of electric energy metering device faults are achieved. It can accurately identify power supply faults, communication faults, and terminal meter faults, and provide targeted alarm analysis suggestions. In addition, the solution has strong intelligent capabilities and flexibility, and can perform dynamic analysis and rule tree construction based on real-time data and evaluation indicators to adapt to different fault types and scenarios.

[0045] like Figure 1 The power scenario-based intelligent operation and maintenance system based on the pre-trained large language model includes capability modules such as system real-time data collection, calculation of relevant evaluation indicators, dynamic construction of large model analysis tree, and intelligent summary of large model analysis results. Through the intelligent operation and maintenance system based on the pre-trained large language model, it overcomes the shortcomings of the decision tree judgment method and rule-based judgment method in other solutions, and achieves higher fault judgment accuracy, intelligence level and flexibility.

[0046] Reference Figure 2 , shows a step flow chart of a fault analysis method provided by some embodiments of the present invention, which can be applied to an intelligent operation and maintenance system, which can be an intelligent operation and maintenance system for a power system.

[0047] The intelligent operation and maintenance system is an application system based on artificial intelligence technology. It is designed to intelligently analyze system alarms and provide alarm analysis recommendations to assist on-site personnel in troubleshooting. This system uses automated methods to monitor, analyze, and diagnose system alarms in real time, helping operation and maintenance personnel locate and resolve issues more quickly and accurately. The intelligent operation and maintenance system has the following key features and functions: Alarm Data Collection and Processing: The intelligent operation and maintenance system collects, processes, and categorizes alarm information generated by the system in real time. Using intelligent algorithms and models, the system converts alarm information into structured data for subsequent analysis and processing. Intelligent Analysis and Diagnosis: The system utilizes artificial intelligence technologies, including machine learning, data mining, and natural language processing, to analyze and diagnose alarm data. By learning from historical data and performing pattern recognition, the system automatically identifies common fault patterns and anomalies and generates corresponding alarm analysis reports. Alarm Analysis Recommendations: The system provides alarm analysis recommendations based on the analysis and diagnosis results of alarm data. These suggestions can address specific fault causes, solutions, and preventative measures, helping field personnel better understand the problem and take appropriate action. Assisted troubleshooting: The intelligent O&M system provides field personnel with assisted troubleshooting. The system's alarm analysis reports and suggestions allow field personnel to conduct more targeted troubleshooting and resolution, reducing the time required to locate and recover from faults. The application of the intelligent O&M system can significantly improve O&M efficiency and system stability, reducing the impact of faults on services. It automatically processes and analyzes large amounts of alarm data, providing field personnel with accurate analysis results and suggestions, helping them resolve issues more quickly and improving system reliability and availability.

[0048] Specifically, the following steps may be included:

[0049] Step 201: Acquire collected key data and determine fault-related indicators based on the key data.

[0050] In some embodiments of the present invention, the key data may be key data in the power system, such as a fault diagnosis sheet of an electric energy metering device, work order information, a power outage analysis result, and message information.

[0051] The energy metering device fault diagnosis sheet provides customer information, meter parameters, power outage information, and other data. By collecting this information, we can understand the basic situation and operating status of the energy metering device, including the manufacturer, wiring method, transformation ratio, and accuracy level.

[0052] Work order information may include work order number, work order type, application time, work order status, etc. In addition, the outage assessment results may include the outage asset number, outage start time, outage end time, outage type, etc.

[0053] Message information can include message time and signal strength, which can help the system monitor the communication status and stability of the device, as well as evaluate and analyze the communication quality.

[0054] In some examples, key data may also include indicators such as the meter reading success rate, line loss rate, and terminal online rate of the substation (line). These indicators reflect the stability and reliability of the electric energy metering device and have an important impact on the accuracy and effectiveness of data collection.

[0055] In practical applications, a real-time data acquisition module can be set up to collect various key information from the power grid in real time, providing data on energy metering devices and communication status, and providing reliable basic data for subsequent alarm analysis and fault diagnosis. By analyzing this data, equipment failures and abnormal conditions can be discovered in a timely manner, assisting on-site personnel in troubleshooting and repairing.

[0056] After obtaining the key data, fault-related indicators are calculated from the key data according to preset rules. The preset rules may indicate the method for calculating the fault-related indicators. For example, if the fault-related indicator is the current imbalance rate, the preset rule is to calculate the amplitude and phase difference of the three-phase current using the key data to obtain the current imbalance rate. For details, please refer to the relevant calculations below. After obtaining the fault-related indicators, the fault indicators are calculated and analyzed to promptly identify faults and abnormal conditions of the electric energy metering device, and provide corresponding alarm analysis and suggestions, which help on-site personnel quickly locate and troubleshoot faults, improving the efficiency and accuracy of fault handling.

[0057] In some embodiments of the present invention, fault-related indicators may be relevant indicators in the power system, such as current imbalance rate data, voltage imbalance rate data, and main and auxiliary meter deviation data. These indicators are important bases for evaluating the operating status and fault conditions of the electric energy metering device.

[0058] Among them, the current imbalance rate data is an indicator for evaluating the degree of uneven current load distribution. The calculation formula is (maximum phase current - minimum phase current) / maximum phase current. By collecting the current data contained in the key data, such as current data that can be included in the energy meter parameters or in other data, the amplitude and phase difference of the three-phase current are calculated to obtain the current imbalance rate. The higher the current imbalance rate, the more uneven the load distribution, and there may be problems such as power supply failure or poor cable contact.

[0059] Voltage imbalance rate data is an indicator used to assess the uneven distribution of supply voltage. The calculation formula is (maximum phase voltage - minimum phase voltage) / maximum phase voltage. The voltage imbalance rate is calculated by calculating the amplitude and phase difference of the three-phase voltages using voltage data included in key data collection, such as energy meter parameters or other data. A higher voltage imbalance rate indicates a more uneven distribution of supply voltage, potentially impacting the normal operation of equipment and the accuracy of energy metering.

[0060] The deviation between the primary and secondary meters is an indicator used to assess the error between them. By calculating the energy metering data for the primary and secondary meters, using the primary and secondary meter counts contained in key data collected (e.g., meter parameters or other data), the deviation between the primary and secondary meters can be determined. A larger deviation indicates a significant energy metering error between the primary and secondary meters, potentially requiring calibration or equipment replacement.

[0061] Step 202: construct a dynamic rule tree based on the fault-related indicators.

[0062] After obtaining fault-related indicators, a dynamic rule tree can be constructed. The dynamic rule tree can have different branches, and each branch can have different nodes. The branches and nodes of the tree represent different fault types and possible fault causes (each branch can correspond to a larger range of fault types, and the node can be the specific fault cause under the fault type). By matching and comparing with pre-set rules, the system can automatically determine the possible fault type and possible fault cause, and give corresponding analysis suggestions and processing methods. On-site personnel can conduct targeted fault troubleshooting and repair based on the intelligent analysis results provided by the system.

[0063] In an embodiment of the present invention, a rule tree is dynamically constructed based on the evaluation indicators collected from real-time data to achieve flexible judgment of different fault types. Compared with traditional static rule trees that are difficult to adapt to complex scenarios and changing fault conditions, the technical innovation of dynamically constructing rule trees enables the system to dynamically adjust the structure and judgment logic of the rule tree according to actual conditions and changes in evaluation indicators, thereby improving the flexibility and accuracy of fault judgment.

[0064] In some embodiments of the present invention, constructing a dynamic rule tree according to the fault-related indicators includes:

[0065] According to the fault-related indicators, a target fault cause is determined; in a preset dynamic rule tree, nodes corresponding to the target fault cause are marked to construct the marked dynamic rule tree.

[0066] In actual applications, multiple nodes can be set in the preset dynamic rule tree, and each node can correspond to a corresponding fault cause. Based on the calculated fault-related indicators, the possible fault causes can be dynamically analyzed. The fault cause can directly correspond to a certain fault type, or it can be a specific fault cause under a certain fault type, such as mainly covering power supply faults, communication faults, and terminal meter faults, to determine whether there is a corresponding type of fault. After determining the target fault cause, the corresponding node in the dynamically constructed rule tree can be marked according to the analysis results. By marking the corresponding node, the target fault cause that may currently exist can be marked, that is, the fault cause corresponding to the marked node is the target fault cause.

[0067] Power supply faults refer to faults related to the power supply to the energy metering device, such as power outages and power fluctuations. By analyzing indicators such as the current imbalance rate and voltage imbalance rate, the system can determine whether a power supply fault exists and mark the corresponding node in the dynamically constructed rule tree based on the analysis results.

[0068] Communication failures refer to failures related to the communication transmission of the energy metering device, such as communication interruptions or abnormalities. By analyzing indicators such as communication success rate and message information, the system can determine whether a communication failure exists and mark the corresponding node in the dynamically constructed rule tree.

[0069] Terminal meter faults refer to faults related to the functions or parameters of the energy metering device itself, such as an open meter cover or a serial port anomaly. By analyzing indicators such as primary and secondary meter deviation data and terminal online rates, the system can determine whether a terminal meter fault exists and mark the corresponding node in the dynamically constructed rule tree.

[0070] Step 203: Acquire fault ticket information, and generate fault summary information for the fault ticket information based on the fault ticket information and the dynamic rule tree through a pre-trained target model.

[0071] In some embodiments of the present invention, the fault ticket information may be fault ticket information for the power system, and the fault ticket information may include any one or more of the following: fault description information, fault occurrence time, and fault type. The fault summary information may include any one or more of the following: fault cause, fault impact scope, and fault recommended handling plan.

[0072] In some embodiments of the present invention, the target model may be a large language model. In some examples, the target model can be fine-tuned using a training dataset using an adaptive learning rate adjustment approach. In some examples, LORA (Layer-wise Optimized Rate Adaptation), a novel fine-tuning strategy, is employed for model training. By optimizing the learning rate layer by layer, LORA avoids vanishing or exploding gradients during fine-tuning, improving the fine-tuning performance of the model. LORA first adaptively adjusts the learning rate based on the parameters of each layer, ensuring smoother parameter updates. Then, through a layer-by-layer recursive approach, the updated results of the previous layer serve as input to the next layer, enabling information transfer and parameter adjustment between layers. This layer-by-layer information transfer fully utilizes the model's hierarchical structure and feature representation capabilities during fine-tuning, improving the model's expressiveness and performance. Furthermore, LORA technology is optimized for large model fine-tuning tasks. Through adaptive learning rate adjustment and layer-by-layer information transfer, it effectively improves the convergence speed and performance of large model fine-tuning. This approach can better cope with the challenges of large-scale model training and improve the generalization ability and performance of the model.

[0073] In this embodiment of the present invention, a pre-trained large language model is fine-tuned for specific tasks and goals in the intelligent O&M field, making it more adaptable to the needs of fault diagnosis and analysis. During the fine-tuning process, parameters are adjusted and training data is refined based on domain knowledge and actual application scenarios, thereby improving the model's accuracy and generalization capabilities in fault diagnosis. The technical innovation of fine-tuning the large language model enables this solution to better understand semantics and context, thereby achieving more accurate and intelligent fault diagnosis and analysis.

[0074] In practical applications, an interactive interface can be provided to users, through which they can submit trouble ticket information. After obtaining the trouble ticket information, the trouble ticket information and the dynamic rule tree can be input into a pre-trained target model. The model can then determine the possible causes of the fault and output a fault summary for the trouble ticket information.

[0075] In this embodiment of the present invention, a fine-tuned large language model is used to generate highly accurate and interpretable intelligent fault summaries based on input trouble ticket data and a dynamically constructed rule tree. This large model combines pre-trained language models with domain knowledge to generate detailed fault cause descriptions and recommended solutions through a deep understanding of fault causes and context. This enables the system to provide more comprehensive, accurate, and personalized fault diagnosis and analysis results, providing more valuable guidance and decision support to on-site personnel.

[0076] Compared to decision tree-based judgment methods, using a pre-trained large language model for fault cause diagnosis offers greater intelligence and semantic understanding capabilities. By learning from large amounts of data, pre-trained large language models can better understand context and semantics, improving the accuracy and intelligence of fault cause diagnosis. Compared to rule-based judgment methods, the intelligent summary module of large model analysis results utilizes a fine-tuned large model to generate intelligent fault summaries, providing detailed fault causes and recommended solutions. The large model combines pre-trained language models with domain knowledge for fault analysis, improving the accuracy and interpretability of fault summaries.

[0077] In some embodiments of the present invention, generating fault summary information for the trouble ticket information based on the trouble ticket information and the dynamic rule tree using a pre-trained target model includes:

[0078] The node matching result of the trouble ticket information in the dynamic rule tree is determined through a pre-trained target model; and the trouble summary information for the trouble ticket information is generated according to the trouble ticket information and the node matching result through the pre-trained target model.

[0079] The dynamic rule tree may include multiple nodes, and different nodes correspond to different fault causes.

[0080] In some embodiments of the present invention, determining the node matching result of the trouble ticket information in the dynamic rule tree by using a pre-trained target model includes:

[0081] Through the pre-trained target model, node matching is performed in the dynamic rule tree according to the fault description information in the fault ticket information to obtain a node matching result.

[0082] In actual applications, based on the dynamically constructed rule tree, the system can determine the node matching results by matching keywords in the fault description with the rule tree nodes, and then determine the possible cause of the fault.

[0083] In some examples, if multiple nodes are matched, the fault causes corresponding to these nodes can be directly summarized to obtain the final fault cause. Of course, the most likely fault cause can also be determined from multiple fault causes as the final fault cause.

[0084] For example, by performing node matching in a dynamic rule tree, the fault causes corresponding to the nodes include uneven current load distribution and uneven voltage distribution. The uneven current load distribution and uneven voltage distribution can be directly used as the final fault causes. Of course, the one with a larger degree of unevenness can also be selected as the final fault cause based on the degree of uneven current load distribution and uneven voltage distribution.

[0085] After obtaining node matching results, the system uses a fine-tuned large model to intelligently analyze the input trouble ticket data and the matching results of the rule tree, generating a fault summary for the trouble ticket information. For example, the large model uses a pre-trained language model, combined with a dynamically constructed rule tree, domain knowledge, and contextual information, to determine and infer the cause of the fault. Based on the key information in the trouble ticket, the large model generates an intelligent summary, including the cause of the fault, the scope of impact, and recommended solutions.

[0086] In some embodiments of the present invention, the present invention further includes:

[0087] The fault summary information is fed back using a preset format.

[0088] After obtaining the fault summary information, the system can output it in a fixed format for on-site personnel to quickly understand and apply. Based on the intelligent summary output by the system, on-site personnel can conduct targeted troubleshooting and repairs, improving the efficiency and accuracy of troubleshooting. For example, the format can include a text description of the fault cause, recommended treatment measures, and related alarm levels and priorities.

[0089] In an embodiment of the present invention, by acquiring collected key data and determining fault-related indicators based on the key data, a dynamic rule tree is constructed based on the fault-related indicators to acquire fault ticket information, and through a pre-trained target model, fault summary information for the fault ticket information is generated based on the fault ticket information and the dynamic rule tree, thereby realizing the use of dynamic rule trees and models to perform fault analysis and improving the accuracy of fault analysis.

[0090] Reference Figure 3 , shows a flowchart of another fault analysis method provided by some embodiments of the present invention, which may specifically include the following steps:

[0091] Step 301: Acquire collected key data and determine fault-related indicators based on the key data.

[0092] Step 302: construct a dynamic rule tree based on the fault-related indicators.

[0093] Step 303: Obtain trouble ticket information.

[0094] Step 304: Determine the node matching result of the fault ticket information in the dynamic rule tree through the pre-trained target model; wherein the dynamic rule tree includes multiple nodes, and different nodes correspond to different fault causes.

[0095] Step 305 : Generate fault summary information for the fault ticket information based on the fault ticket information and the node matching result through a pre-trained target model.

[0096] In an embodiment of the present invention, by combining the target model, dynamic rule tree, and fault ticket information, node matching is performed in the dynamic rule tree, and fault summary information is generated based on the node matching results and fault ticket information, thereby improving the efficiency of the model in predicting faults and improving the accuracy of fault analysis.

[0097] Reference Figure 4 , shows a flowchart of another fault analysis method provided by some embodiments of the present invention, which may specifically include the following steps:

[0098] Step 401: Acquire collected key data and determine fault-related indicators based on the key data.

[0099] Step 402: Determine the target fault cause based on the fault-related indicators.

[0100] Step 403: Mark the nodes corresponding to the target fault cause in the preset dynamic rule tree to construct the marked dynamic rule tree.

[0101] Step 404 , obtaining fault ticket information, and generating fault summary information for the fault ticket information based on the fault ticket information and the dynamic rule tree through a pre-trained target model.

[0102] In an embodiment of the present invention, a dynamic rule tree is preset, and the target fault cause is determined based on the collected key data. The target fault cause is marked in the dynamic rule tree according to the currently existing target fault cause, and the dynamic rule tree can be updated in real time and dynamically, thereby improving the real-time and accuracy of the data.

[0103] Reference Figure 5 , shows a flowchart of another fault analysis method provided by some embodiments of the present invention, which may specifically include the following steps:

[0104] Step 501: Acquire collected key data and determine fault-related indicators based on the key data.

[0105] Step 502: construct a dynamic rule tree based on the fault-related indicators.

[0106] Step 503 : Acquire fault ticket information, and generate fault summary information for the fault ticket information by using the pre-trained target model, the fault information and the dynamic rule tree.

[0107] Step 504: Feedback the fault summary information in a preset format.

[0108] In the embodiment of the present invention, by using a preset format to feed back the fault summary information, the user can view the fault summary information more intuitively, thereby improving the interactive experience.

[0109] refer to Figure 6 , shows a flowchart of another fault analysis method provided by some embodiments of the present invention, which may specifically include the following steps:

[0110] Step 601: pre-build a decision tree model.

[0111] Step 602: Acquire the collected key data and determine a feature vector based on the key data.

[0112] Step 603: Obtain fault ticket information, analyze the feature vector corresponding to the fault ticket information through the decision tree model, and generate fault summary information for the fault ticket information.

[0113] It should be noted that for relevant descriptions of key data, fault ticket information, fault summary information, etc., please refer to the relevant descriptions above and will not be repeated here.

[0114] In practical applications, the decision tree algorithm can be used to perform fault analysis. The decision tree is a classification model based on rules and attributes. It judges and branches the input features and finally obtains the decision result of the cause of the fault.

[0115] Specifically, a decision tree model can be constructed by pre-collecting sample data. After the decision tree model is constructed, key data can be characterized to obtain the feature vectors required for fault diagnosis. Upon receiving a fault ticket, the feature vectors associated with the fault ticket can be searched. The decision tree model can then be used to compare the values ​​of each feature vector with a preset threshold to determine the fault cause. This can then generate corresponding alarm analysis recommendations based on the cause, resulting in a final fault summary.

[0116] refer to Figure 7 , shows a flowchart of another fault analysis method provided by some embodiments of the present invention, which may specifically include the following steps:

[0117] Step 701: Establish fault judgment rules in advance.

[0118] Step 702: Acquire the collected key data and determine fault-related indicators based on the key data.

[0119] Step 703: Obtain fault ticket information, combine the fault judgment rules and the fault-related indicators, and generate fault summary information for the fault ticket information.

[0120] It should be noted that for relevant descriptions of key data, fault-related indicators, fault ticket information, fault summary information, etc., please refer to the relevant descriptions above and will not be repeated here.

[0121] In practical applications, rules can be used to analyze faults. By pre-establishing fault judgment rules, each rule corresponds to a fault cause. These rules can be used to determine faults based on expert knowledge and experience, using specific conditions and logic. Key data is used to identify fault-related indicators. Once fault ticket information is obtained, the associated fault-related indicators can be identified. Fault judgment rules are then used to determine the fault-related indicators and determine the cause of the fault. Based on the cause, corresponding alarm analysis recommendations can be generated, resulting in a final fault summary.

[0122] It should be noted that for the sake of simplicity, the method embodiments are described as a series of actions. However, those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0123] Some embodiments of the present invention further provide a fault analysis device, which is used to implement the fault analysis method described above, specifically comprising:

[0124] Obtaining collected key data and determining fault-related indicators based on the key data;

[0125] Constructing a dynamic rule tree based on the fault-related indicators;

[0126] Fault ticket information is acquired, and fault summary information for the fault ticket information is generated according to the fault ticket information and the dynamic rule tree through a pre-trained target model.

[0127] In some embodiments of the present invention, generating fault summary information for the trouble ticket information based on the trouble ticket information and the dynamic rule tree using a pre-trained target model includes:

[0128] Determine, by using a pre-trained target model, a node matching result of the fault ticket information in the dynamic rule tree; wherein the dynamic rule tree includes a plurality of nodes, and different nodes correspond to different fault causes;

[0129] The pre-trained target model is used to generate fault summary information for the fault ticket information based on the fault ticket information and the node matching result.

[0130] In some embodiments of the present invention, determining the node matching result of the trouble ticket information in the dynamic rule tree by using a pre-trained target model includes:

[0131] Through the pre-trained target model, node matching is performed in the dynamic rule tree according to the fault description information in the fault ticket information to obtain a node matching result.

[0132] In some embodiments of the present invention, constructing a dynamic rule tree according to the fault-related indicators includes:

[0133] Determine the cause of the target fault based on the fault-related indicators;

[0134] In a preset dynamic rule tree, nodes corresponding to the target fault cause are marked to construct the marked dynamic rule tree.

[0135] In some embodiments of the present invention, the present invention further includes:

[0136] The fault summary information is fed back using a preset format.

[0137] In some embodiments of the present invention, the fault ticket information includes any one or more of the following: fault description information, fault occurrence time, and fault type.

[0138] In some embodiments of the present invention, the fault summary information includes any one or more of the following: fault cause, fault impact scope, and fault handling solution.

[0139] In some embodiments of the present invention, the target model is a large language model.

[0140] In some embodiments of the present invention, the key data is key data in the power system, and the fault ticket information is fault ticket information for the power system.

[0141] Some embodiments of the present invention further provide a fault analysis device, which is used to implement the fault analysis method described above, specifically comprising:

[0142] Pre-built decision tree models;

[0143] Acquire the collected key data, and determine a feature vector based on the key data;

[0144] Fault ticket information is acquired, and feature vectors corresponding to the fault ticket information are analyzed using the decision tree model to generate fault summary information for the fault ticket information.

[0145] Some embodiments of the present invention further provide a fault analysis device, which is used to implement the fault analysis method described above, specifically comprising:

[0146] Establish fault judgment rules in advance;

[0147] Obtaining collected key data and determining fault-related indicators based on the key data;

[0148] Obtain fault ticket information, and generate fault summary information for the fault ticket information in combination with the fault judgment rule and the fault-related indicators.

[0149] Some embodiments of the present invention further provide an electronic device, which may include a processor, a memory, and a computer program stored in the memory and capable of running on the processor, and the computer program implements the above fault analysis method when executed by the processor.

[0150] Some embodiments of the present invention further provide a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above fault analysis method is implemented.

[0151] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0152] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0153] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0154] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, apparatus, or computer program products. Thus, embodiments of the present invention may take the form of a fully hardware embodiment, a fully software embodiment, or an embodiment combining software and hardware. Furthermore, embodiments of the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, compact disc read-only memory, optical storage, etc.) containing computer-usable program code.

[0155] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the process in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0156] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0157] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0158] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they become aware of the basic creative concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0159] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or terminal device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or terminal device that includes the element.

[0160] The above is a detailed introduction to the provided method, device, electronic device and medium for fault analysis. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core ideas. At the same time, for those skilled in the art, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the contents of this specification should not be understood as limiting the present invention.

Claims

1. A method for fault analysis, characterized in that: The method comprises: Obtaining collected key data and determining fault-related indicators based on the key data; Constructing a dynamic rule tree based on the fault-related indicators; Fault ticket information is acquired, and fault summary information for the fault ticket information is generated according to the fault ticket information and the dynamic rule tree through a pre-trained target model.

2. The method according to claim 1, characterized in that The pre-trained target model generates fault summary information for the fault ticket information according to the fault ticket information and the dynamic rule tree, including: Determine, by using a pre-trained target model, a node matching result of the fault ticket information in the dynamic rule tree; wherein the dynamic rule tree includes a plurality of nodes, and different nodes correspond to different fault causes; The pre-trained target model is used to generate fault summary information for the fault ticket information based on the fault ticket information and the node matching result.

3. The method according to claim 2, characterized in that Determining the node matching result of the fault ticket information in the dynamic rule tree by using the pre-trained target model includes: Through the pre-trained target model, node matching is performed in the dynamic rule tree according to the fault description information in the fault ticket information to obtain a node matching result.

4. The method according to any one of claims 1 to 3, characterized in that: The constructing of a dynamic rule tree according to the fault-related indicators includes: Determine the cause of the target fault based on the fault-related indicators; In a preset dynamic rule tree, nodes corresponding to the target fault cause are marked to construct the marked dynamic rule tree.

5. The method according to any one of claims 1 to 3, characterized in that Also includes: The fault summary information is fed back using a preset format.

6. The method according to any one of claims 1 to 3, characterized in that The fault ticket information includes any one or more of the following: fault description information, fault occurrence time, and fault type.

7. The method according to any one of claims 1 to 3, characterized in that The fault summary information includes any one or more of the following: fault cause, fault impact scope, and fault handling solution.

8. The method according to any one of claims 1 to 3, characterized in that The target model is a large language model.

9. The method according to any one of claims 1 to 3, characterized in that The key data is key data in the power system, and the fault ticket information is fault ticket information for the power system.

10. A method for fault analysis, characterized in that: The method comprises: Pre-built decision tree models; Acquire the collected key data, and determine a feature vector based on the key data; Fault ticket information is acquired, and feature vectors corresponding to the fault ticket information are analyzed using the decision tree model to generate fault summary information for the fault ticket information.

11. A method for fault analysis, characterized in that: The method comprises: Establish fault judgment rules in advance; Obtaining collected key data and determining fault-related indicators based on the key data; Obtain fault ticket information, and generate fault summary information for the fault ticket information in combination with the fault judgment rule and the fault-related indicators.

12. A device for fault analysis, characterized in that: The device is used to implement the fault analysis method according to any one of claims 1 to 11.

13. An electronic device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein when the computer program is executed by the processor, the method for fault analysis according to any one of claims 1 to 11 is implemented.

14. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the fault analysis method according to any one of claims 1 to 11 is implemented.

Images