Power grid information system fault positioning system and method based on adaptive learning

Abstract

The invention relates to the technical field of power grid information, in particular to a power grid information system fault positioning system and method based on adaptive learning, and the system comprises a knowledge base which is formed based on historical fault data and expert knowledge experience and is used for assisting fault diagnosis and decision making; the data processing module is used for preprocessing and extracting data; the model training module is used for constructing a fault prediction and positioning model and training the model by using historical fault data; the data processing module comprises a data cleaning unit and a deep learning unit; the deep learning unit automatically extracts spatial features and time sequence features of the power signals from the data by using a convolutional neural network and a long-short term memory network; the method combines two deep learning technologies of the convolutional neural network and the long and short term memory network, automatically extracts the features in the power signals, achieves the functions of rapid diagnosis and positioning of faults, improves the operation and maintenance efficiency of a power grid information system, and guarantees the safe and stable operation of the information system.

Description

Technical Field

[0001] The present invention relates to the field of power grid information technology, and in particular to a power grid information system fault location system and method based on adaptive learning. Background Art

[0002] The power grid information system is characterized by high complexity, strong real-time performance, high dynamics, and high safety requirements. At the same time, the aging of system machines and the gradual decline in performance have led to an increase in the failure rate of the information system year by year, late fault detection time, and long fault recovery cycle. The reliability problem of the system is becoming increasingly prominent. Therefore, how to use new technologies to create intelligent diagnosis of power grid information system faults is particularly important, and fault location is the basis of fault diagnosis.

[0003] A Chinese patent with publication number CN109726200A discloses a power grid information system fault location system and method based on a bidirectional deep neural network, including a fault monitoring module, an inference engine, a database, and a deep learning module; the deep neural network is trained forward and backward on the fault tree analysis results to obtain the corresponding relationship between the fault characteristics and the fault location and store it in the expert knowledge base, and the power grid information system fault monitoring is performed, and the fault characteristics obtained according to the monitoring data are written into the characteristic database, and the fault location result is inferred based on the expert knowledge base and the fault characteristics; the invention can effectively solve the problems of difficult, slow and inaccurate fault location in the power grid information system, so that the system can issue an alarm as soon as the fault occurs, and combined with the fault self-processing system, it can realize the intelligent fault location and intelligent processing functions of the information system, which can further improve the operation and maintenance efficiency of the power grid information system and ensure the safe and stable operation of the information system.

[0004] In the current existing technology, the training process of the bidirectional deep neural network is relatively complicated, requiring a large amount of historical fault data for forward and reverse training. The bidirectional deep neural network also needs to consider future information, and the training cost is high. Due to the complexity of the model, a large amount of computing resources and time are required for training, especially when processing large-scale data sets, the feature extraction speed is affected.

[0005] To this end, the present invention provides a power grid information system fault location system and method based on adaptive learning. Summary of the invention

[0006] To this end, the technical problem to be solved by the present invention is to overcome the problem that the training process of the bidirectional deep neural network in the prior art is relatively complicated, requiring a large amount of historical fault data for forward and reverse training, and the bidirectional deep neural network needs to consider future information, and the training cost is high. Due to the complexity of the model, a large amount of computing resources and time are required for training, especially when processing large-scale data sets, the feature extraction speed is affected.

[0007] In order to solve the above technical problems, the present invention provides a power grid information system fault location system based on adaptive learning, comprising:

[0008] Knowledge base, which is formed based on historical fault data and expert knowledge and experience, is used to assist fault diagnosis and decision-making;

[0009] Data processing module, used for data preprocessing and extraction;

[0010] Model training module, used to build fault prediction and location models, and train the models using historical fault data;

[0011] The data processing module includes a data cleaning unit and a deep learning unit;

[0012] The deep learning unit automatically extracts the spatial features and time series features of the power signal from the data using a convolutional neural network (CNN) and a long short-term memory (LSTM) network;

[0013] The model training module includes:

[0014] The model building unit selects a suitable machine learning algorithm to build a fault prediction and location model based on the extracted features and sets the model parameters;

[0015] Supervised learning method unit, a neural network-based algorithm uses historical fault data to train the model so that the model can recognize and classify different fault types;

[0016] The adaptive learning algorithm unit adopts an adaptive learning algorithm to optimize the performance of the model by adjusting model parameters and algorithms based on the changing dynamics of the grid structure and operating status.

[0017] In one embodiment of the present invention, the working process of the model training module includes the following steps:

[0018] A1: Dataset division: random division is used to randomly divide the entire dataset into training set and test set;

[0019] A2: Use the training set data to train the model. During the training process, the parameters are continuously adjusted through the back propagation algorithm.

[0020] A3: After the model training is completed, use the test set data to evaluate the performance of the model; if the performance is poor, you need to further adjust the model parameters or improve the data preprocessing.

[0021] In one embodiment of the present invention, it also includes a data acquisition module for collecting real-time operation data of the power grid; the data acquisition module collects real-time operation data of the power grid through sensors and intelligent devices; the real-time operation data includes voltage, current, power, temperature and humidity, and also includes the operation status and fault record information of the power grid equipment.

[0022] In one embodiment of the present invention, it also includes a fault diagnosis and positioning module; the fault diagnosis and positioning module includes a fault diagnosis unit and a fault positioning unit; the fault diagnosis unit is used to diagnose the type of fault. When the data of the data acquisition module is compared with the data of the knowledge base, the diagnosis is performed based on the trained model parameters. When the data of the data acquisition module is similar to the data of the knowledge base and is less than the set threshold, it is recorded as abnormal data. The abnormal data is diagnosed by the fault diagnosis unit to determine the type of fault; the fault positioning unit determines the specific location of the fault based on the fault diagnosis, by numbering the power grid and editing the power grid number on the electronic map, and setting a GPS sensor for each number, locating the power grid fault through the GPS sensor, and sending the corresponding number on the electronic map to the operation and maintenance personnel, providing the operation and maintenance personnel with positioning information.

[0023] In one embodiment of the present invention, it also includes a policy generation unit, which is used to infer and generate a fault operation policy based on the fault diagnosis and location results; the fault operation policy is sent to the user interface for operation and maintenance personnel to view the fault location results and perform related operations.

[0024] In one embodiment of the present invention, an optimization and update module is also included, and the optimization and update module includes an effect evaluation unit and a model update unit; the effect evaluation unit evaluates the performance of the model through feedback from operation and maintenance personnel on the effect of the fault diagnosis and positioning module; the model update unit adjusts the model parameters based on abnormal data for updating and optimizing the model.

[0025] In one embodiment of the present invention, a method for locating a fault in a power grid information system based on adaptive learning is provided, the method comprising the following steps:

[0026] S1: Data collection and preprocessing, by collecting real-time operation data and historical fault data, extracting features useful for fault location;

[0027] S2: Build an adaptive learning model. Design an adaptive learning model based on the requirements of fault location and data characteristics.

[0028] S3: Model training, training the model through the data set and evaluating the model performance;

[0029] S4: Adaptive learning and optimization, inputting new data into the model in real time for online learning, and continuously monitoring the performance of the model;

[0030] S5: Fault location: the model diagnoses the fault location and type based on the input data and sends fault information to the operation and maintenance personnel;

[0031] S6: Feedback and iteration: collect feedback from operation and maintenance personnel on the fault location results, and further train and optimize the model.

[0032] In one embodiment of the present invention, the adaptive learning and optimization in S4 comprises the following steps:

[0033] S401: Online learning, inputting new data into the model in real time for online learning and updating model parameters;

[0034] S402: Performance monitoring: continuously monitor the performance of the model, and trigger model optimization when performance degradation is found;

[0035] S403: Model optimization: tuning the model based on the performance monitoring results.

[0036] In one embodiment of the present invention, the fault location in S5 includes the following steps:

[0037] S501: Input data, input the real-time data in the power grid information system into the trained model;

[0038] S502: Fault prediction: the model predicts the fault location and type based on the input data;

[0039] S503: Output the results and display the fault prediction results in a visual manner for operation and maintenance personnel to troubleshoot and repair.

[0040] In one embodiment of the present invention, the feedback and iteration in S6 includes the following steps:

[0041] S601: Collect feedback, collect feedback from operation and maintenance personnel on the fault location result;

[0042] S602: Data labeling: labeling the newly added data based on the feedback results for further training and optimization of the model;

[0043] S603: Iterative update: based on feedback results and new data, the model is continuously updated iteratively.

[0044] The above technical solution of the present invention has the following advantages compared with the prior art:

[0045] The invention discloses a fault location system and method for a power grid information system based on adaptive learning. The system and method cooperate with a data acquisition module, a data processing module and a model training module to collect real-time operation data of the power grid and then perform data cleaning. The convolutional neural network (CNN) and the long short-term memory (LSTM) network can be used to automatically extract spatial features and time series features of power signals from the data. According to the extracted features, a suitable machine learning algorithm is selected to construct a fault prediction and location model, and model parameters are set. The neural network-based algorithm uses historical fault data to train the model so that the model can identify and classify different fault types. The adaptive learning algorithm is used to adjust the model parameters and the algorithm according to the dynamic changes in the power grid structure and operation status. The core of the system lies in its adaptive learning algorithm, which continuously learns and adjusts the model parameters to adapt to the changes in the power grid structure and operation status. The system combines two deep learning technologies, the convolutional neural network and the long short-term memory network, to automatically extract features in the power signal, realize the function of rapid diagnosis and location of faults, improve the operation and maintenance efficiency of the power grid information system, and ensure the safe and stable operation of the information system.

[0046] The present invention discloses a fault location system and method for a power grid information system based on adaptive learning. The system uses a fault diagnosis and location module to number the power grid and edit the power grid numbers on an electronic map. A GPS sensor is set for each number. The data of the data acquisition module is compared with the data of the knowledge base, and a diagnosis is performed based on the trained model parameters. When the data of the data acquisition module is similar or close to the data of the knowledge base and is less than a set threshold, it is judged as abnormal data and needs to be diagnosed, thereby determining the type and location of the fault. The number on the electronic map is identified by the GPS sensor and the corresponding number is sent to the operation and maintenance personnel, providing the operation and maintenance personnel with location information. According to the fault diagnosis and location results, a fault handling operation strategy is generated by adaptive reasoning and sent to the operation and maintenance personnel along with the location results. The operation and maintenance personnel can directly view the fault location results and perform related operations, which facilitates fast and accurate maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to make the contents of the present invention more clearly understood, the present invention is further described in detail below according to specific embodiments of the present invention in conjunction with the accompanying drawings.

[0048] Figure 1 It is a schematic diagram of the framework of the power grid information system fault location system in the present invention;

[0049] Figure 2 It is a flow chart of the method for model training in the present invention;

[0050] Figure 3 It is a method flow chart of the method for locating faults in the power grid information system of the present invention;

[0051] Figure 4 is a flow chart of the method for adaptive learning and optimization in the present invention;

[0052] Figure 5 is a flow chart of the method for fault location in the present invention;

[0053] Figure 6 is a flow chart of the feedback and iteration method in the present invention; DETAILED DESCRIPTION

[0054] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0055] See also Figure 1 - Figure 6 The present invention provides a fault location system for a power grid information system based on adaptive learning, comprising: a knowledge base, which is formed based on historical fault data and expert knowledge and experience and is used to assist fault diagnosis and decision-making; a data processing module, which is used to preprocess and extract data; a model training module, which is used to build a fault prediction and location model and train the model using historical fault data; the data processing module comprises a data cleaning unit and a deep learning unit; the deep learning unit automatically extracts spatial features and time series features of power signals from data using a convolutional neural network (CNN) and a long short-term memory (LSTM) network; the model training module comprises: a model building unit, which selects a suitable machine learning algorithm to build a fault prediction and location model and sets model parameters based on the extracted features; a supervised learning method unit, which uses a neural network-based algorithm to train the model using historical fault data so that the model can identify and classify different fault types; an adaptive learning algorithm unit, which uses an adaptive learning algorithm to optimize the performance of the model by adjusting model parameters and algorithms based on the changing dynamics of the power grid structure and operating status.

[0056] When used, the power grid information system fault location system provided by the present invention collects real-time operation data of the power grid and then performs data cleaning, and can use convolutional neural network CNN and long short-term memory network LSTM to automatically extract spatial features and time series features of power signals from the data; according to the extracted features, a suitable machine learning algorithm is selected to build a fault prediction and location model, and model parameters are set. The neural network-based algorithm uses historical fault data to train the model so that the model can identify and classify different fault types; an adaptive learning algorithm is used to adjust the model parameters and algorithm according to the changing dynamics of the power grid structure and operating status; the core of this system lies in its adaptive learning algorithm, which continuously learns and adjusts the model parameters to adapt to changes in the power grid structure and operating status; combining the two deep learning technologies of convolutional neural network and long short-term memory network, the features in the power signal are automatically extracted to achieve the function of rapid diagnosis and location of faults.

[0057] Furthermore, if Figure 2 As shown, the working process of the model training module includes the following steps:

[0058] A1: Dataset division: random division is used to randomly divide the entire dataset into training set and test set;

[0059] A2: Use the training set data to train the model. During the training process, the parameters are continuously adjusted through the back propagation algorithm.

[0060] A3: After the model training is completed, use the test set data to evaluate the performance of the model; if the performance is poor, you need to further adjust the model parameters or improve the data preprocessing.

[0061] When the model training provided by the present invention is used, the entire data set is randomly divided into a training set and a test set by a random partitioning method, the training set data is used to train the model, and the test set data is used to evaluate the performance of the model; during the training process, the model continuously adjusts parameters through the back propagation algorithm to minimize the loss function; if the model performs well on the test set, it means that it has strong generalization ability; if the performance is poor, it is necessary to further adjust the model parameters or improve the data preprocessing.

[0062] Furthermore, if Figure 1 As shown, it also includes a data acquisition module for collecting real-time operation data of the power grid; the data acquisition module collects real-time operation data of the power grid through sensors and intelligent devices; the real-time operation data includes voltage, current, power, temperature and humidity, and also includes the operation status and fault record information of the power grid equipment.

[0063] When in use, the data acquisition module provided by the present invention collects the voltage, current, power, temperature and humidity of the power grid, as well as the operating status and fault record information of the power grid equipment through sensors and intelligent devices. The data acquisition module is the main feature of fault detection.

[0064] Furthermore, if Figure 1 As shown, it also includes a fault diagnosis and positioning module; the fault diagnosis and positioning module includes a fault diagnosis unit and a fault positioning unit; the fault diagnosis unit is used to diagnose the type of fault. When the data of the data acquisition module is compared with the data of the knowledge base, the diagnosis is performed based on the trained model parameters. When the data of the data acquisition module is similar to the data of the knowledge base and is less than the set threshold, it is recorded as abnormal data. The abnormal data is diagnosed by the fault diagnosis unit to determine the type of fault; the fault positioning unit determines the specific location of the fault based on the fault diagnosis, by numbering the power grid and editing the power grid number on the electronic map, and each number is set with a GPS sensor, the power grid fault is located by the GPS sensor, and the corresponding number on the electronic map is sent to the operation and maintenance personnel to provide positioning information for the operation and maintenance personnel.

[0065] When the fault diagnosis and positioning module provided by the present invention is in use, the power grid is numbered and the power grid numbers are edited on an electronic map, and a GPS sensor is set for each number. The data of the data acquisition module is compared with the data of the knowledge base, and diagnosis is performed based on the trained model parameters. When the data of the data acquisition module is similar or close to the data of the knowledge base, and is less than a set threshold, it is judged to be abnormal data and needs to be diagnosed, so as to determine the type and location of the fault, identify the number on the electronic map through the GPS sensor, and send the corresponding number to the operation and maintenance personnel, so as to provide the operation and maintenance personnel with positioning information.

[0066] Furthermore, if Figure 1 As shown, it also includes a strategy generation unit, which is used to infer and generate a fault operation strategy based on the fault diagnosis and location results; the fault operation strategy is sent to the user interface for operation and maintenance personnel to view the fault location results and perform related operations.

[0067] When in use, the decision generation module provided by the present invention adaptively infers and generates a fault handling operation strategy based on the fault diagnosis and location results, and sends it to the operation and maintenance personnel along with the location results. The operation and maintenance personnel can directly view the fault location results and perform related operations, which facilitates fast and accurate maintenance.

[0068] Furthermore, if Figure 1As shown, it also includes an optimization and update module, which includes an effect evaluation unit and a model update unit; the effect evaluation unit evaluates the performance of the model through feedback from the operation and maintenance personnel on the effect of the fault diagnosis and positioning module; the model update unit adjusts the model parameters based on abnormal data for model updating and optimization.

[0069] When the feedback and optimization module provided by the present invention is used, firstly, the effectiveness of the system is verified by establishing fault handling information understanding model evaluation indicators and test data sets; secondly, an online learning algorithm is introduced to enable the model to continuously learn and update the knowledge base in practical applications; thirdly, the database is regularly updated and maintained to ensure that the data therein is updated in real time and the fault characteristics and location information are accurately stored.

[0070] Furthermore, if Figure 2 As shown, a fault location method for a power grid information system based on adaptive learning, the fault location method comprises the following steps:

[0071] S1: Data collection and preprocessing, by collecting real-time operation data and historical fault data, extracting features useful for fault location;

[0072] S2: Build an adaptive learning model. Design an adaptive learning model based on the requirements of fault location and data characteristics.

[0073] S3: Model training, training the model through the data set and evaluating the model performance;

[0074] S4: Adaptive learning and optimization, inputting new data into the model in real time for online learning, and continuously monitoring the performance of the model;

[0075] S5: Fault location: the model diagnoses the fault location and type based on the input data and sends fault information to the operation and maintenance personnel;

[0076] S6: Feedback and iteration: collect feedback from operation and maintenance personnel on the fault location results, and further train and optimize the model.

[0077] When the fault location method provided by the present invention is used, firstly, real-time operation data and historical fault data are collected to extract features useful for fault location; then, an adaptive learning model is designed according to the requirements of fault location and data characteristics; the model is trained and the performance of the model is evaluated through the data set, new data is input into the model in real time for online learning, and the performance of the model is continuously monitored. The model diagnoses the fault location and type according to the input data and sends fault information to the operation and maintenance personnel; feedback from the operation and maintenance personnel on the fault location results is collected, and the model is further trained and optimized; the system is enabled to issue an alarm at the first time a fault occurs, and combined with feedback and iteration, the operation and maintenance efficiency of the power grid information system can be further improved to ensure the safe and stable operation of the information system.

[0078] Furthermore, if Figure 4 As shown, the adaptive learning and optimization in S4 includes the following steps:

[0079] S401: Online learning, inputting new data into the model in real time for online learning and updating model parameters;

[0080] S402: Performance monitoring: continuously monitor the performance of the model, and trigger model optimization when performance degradation is found;

[0081] S403: Model optimization: tuning the model based on the performance monitoring results.

[0082] When the adaptive learning and optimization provided by the present invention is used, new data is input into the model in real time for online learning, and the performance of the model is continuously monitored. When the performance is found to have declined, model optimization is triggered. Based on the performance monitoring results, the model is optimized, thereby realizing the function of coordination between optimization and monitoring.

[0083] Furthermore, if Figure 5 As shown, the fault location in S5 includes the following steps:

[0084] S501: Input data, input the real-time data in the power grid information system into the trained model;

[0085] S502: Fault prediction: the model predicts the fault location and type based on the input data;

[0086] S503: Output the results and display the fault prediction results in a visual manner for operation and maintenance personnel to troubleshoot and repair.

[0087] When the fault location provided by the present invention is used, the fault location and type are predicted by inputting data, and the fault prediction results are displayed in a visual manner, which is convenient for operation and maintenance personnel to intuitively view and maintain.

[0088] Furthermore, if Figure 6As shown, the feedback and iteration in S6 includes the following steps:

[0089] S601: Collect feedback, collect feedback from operation and maintenance personnel on the fault location result;

[0090] S602: Data labeling: labeling the newly added data based on the feedback results for further training and optimization of the model;

[0091] S603: Iterative update: based on feedback results and new data, the model is continuously updated iteratively.

[0092] The feedback and iteration provided by the present invention, when used, continuously iterates and updates the model according to feedback and new data, thereby improving the accuracy and practicality of fault location.

[0093] Working principle: By collecting real-time operation data of the power grid and then cleaning the data, the convolutional neural network (CNN) and the long short-term memory network (LSTM) can be used to automatically extract the spatial features and time series features of the power signal from the data; according to the extracted features, the appropriate machine learning algorithm is selected to build a fault prediction and location model, and the model parameters are set. The neural network-based algorithm uses historical fault data to train the model so that the model can identify and classify different fault types; the adaptive learning algorithm is used to adjust the model parameters and algorithm according to the dynamic changes in the power grid structure and operating status; the core of this system lies in its adaptive learning algorithm, which continuously learns and adjusts the model parameters to adapt to changes in the power grid structure and operating status; combining the two deep learning technologies of convolutional neural network and long short-term memory network, the features in the power signal are automatically extracted to achieve the function of rapid diagnosis and location of faults;

[0094] By numbering the power grids and editing the power grid numbers on the electronic map, and setting a GPS sensor for each number, the data from the data acquisition module is compared with the data from the knowledge base, and diagnosis is performed based on the trained model parameters. When the data from the data acquisition module is similar or close to the data from the knowledge base, and is less than the set threshold, it is judged as abnormal data and needs to be diagnosed, so as to determine the type and location of the fault. The number on the electronic map is identified by the GPS sensor and the corresponding number is sent to the operation and maintenance personnel to provide positioning information for the operation and maintenance personnel; according to the fault diagnosis and positioning results, the fault handling operation strategy is generated by adaptive reasoning and sent to the operation and maintenance personnel along with the positioning results. The operation and maintenance personnel can directly view the fault positioning results and perform related operations to facilitate fast and accurate maintenance.

[0095] Obviously, the above embodiments are merely examples for clear explanation and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from these are still within the protection scope of the invention.

Claims

1. A power grid information system fault location system based on adaptive learning, comprising: Knowledge base, which is formed based on historical fault data and expert knowledge and experience, is used to assist fault diagnosis and decision-making; Data processing module, used for data preprocessing and extraction; Model training module, used to build fault prediction and location models, and train the models using historical fault data; Features: The data processing module includes a data cleaning unit and a deep learning unit; The deep learning unit automatically extracts the spatial features and time series features of the power signal from the data using a convolutional neural network (CNN) and a long short-term memory (LSTM) network; The model training module includes: The model building unit selects a suitable machine learning algorithm to build a fault prediction and location model based on the extracted features and sets the model parameters; Supervised learning method unit, a neural network-based algorithm uses historical fault data to train the model so that the model can recognize and classify different fault types; The adaptive learning algorithm unit adopts an adaptive learning algorithm to optimize the performance of the model by adjusting model parameters and algorithms based on the changing dynamics of the power grid structure and operating status.

2. The power grid information system fault location system based on adaptive learning according to claim 1, characterized in that: The working process of the model training module includes the following steps: A1: Dataset division: random division is used to randomly divide the entire dataset into training set and test set; A2: Use the training set data to train the model. During the training process, the parameters are continuously adjusted through the back propagation algorithm. A3: After the model training is completed, use the test set data to evaluate the performance of the model; if the performance is poor, you need to further adjust the model parameters or improve the data preprocessing.

3. The power grid information system fault location system based on adaptive learning according to claim 1, characterized in that: It also includes a data acquisition module for collecting real-time operation data of the power grid; the data acquisition module collects real-time operation data of the power grid through sensors and intelligent devices; the real-time operation data includes voltage, current, power, temperature and humidity, and also includes the operation status and fault record information of the power grid equipment.

4. The power grid information system fault location system based on adaptive learning according to claim 1, characterized in that: It also includes a fault diagnosis and positioning module; the fault diagnosis and positioning module includes a fault diagnosis unit and a fault positioning unit; the fault diagnosis unit is used to diagnose the type of fault. When the data of the data acquisition module is compared with the data of the knowledge base, the diagnosis is performed based on the trained model parameters. When the data of the data acquisition module is similar to the data of the knowledge base and is less than the set threshold, it is recorded as abnormal data. The abnormal data is diagnosed by the fault diagnosis unit to determine the type of fault; the fault positioning unit determines the specific location of the fault based on the fault diagnosis, by numbering the power grid and editing the power grid number on the electronic map, and setting a GPS sensor for each number, locating the power grid fault through the GPS sensor, and sending the corresponding number on the electronic map to the operation and maintenance personnel, providing the operation and maintenance personnel with positioning information.

5. The power grid information system fault location system based on adaptive learning according to claim 4, characterized in that: It also includes a strategy generation unit, which is used to generate a fault operation strategy based on the fault diagnosis and location results; The fault operation strategy is sent to the user interface for operation and maintenance personnel to view the fault location results and perform related operations.

6. A power grid information system fault location system based on adaptive learning according to claim 1, characterized in that: It also includes an optimization and update module, which includes an effect evaluation unit and a model update unit; the effect evaluation unit evaluates the performance of the model through feedback from operation and maintenance personnel on the effect of the fault diagnosis and positioning module; the model update unit adjusts the model parameters based on abnormal data for model updating and optimization.

7. A method for locating faults in a power grid information system based on adaptive learning, applicable to a system for locating faults in a power grid information system based on adaptive learning as claimed in any one of claims 1 to 6, characterized in that: The fault location method comprises the following steps: S1: Data collection and preprocessing, by collecting real-time operation data and historical fault data, extracting features useful for fault location; S2: Build an adaptive learning model. Design an adaptive learning model based on the requirements of fault location and data characteristics. S3: Model training, training the model through the data set and evaluating the model performance; S4: Adaptive learning and optimization, inputting new data into the model in real time for online learning, and continuously monitoring the performance of the model; S5: Fault location: the model diagnoses the fault location and type based on the input data and sends fault information to the operation and maintenance personnel; S6: Feedback and iteration: collect feedback from operation and maintenance personnel on the fault location results, and further train and optimize the model.

8. The method for locating faults in a power grid information system based on adaptive learning according to claim 7, characterized in that: The adaptive learning and optimization in S4 includes the following steps: S401: Online learning, inputting new data into the model in real time for online learning and updating model parameters; S402: Performance monitoring: continuously monitor the performance of the model, and trigger model optimization when performance degradation is found; S403: Model optimization: tuning the model based on the performance monitoring results.

9. The method for locating faults in a power grid information system based on adaptive learning according to claim 7, characterized in that: The fault location in S5 includes the following steps: S501: Input data, input the real-time data in the power grid information system into the trained model; S502: Fault prediction: the model predicts the fault location and type based on the input data; S503: Output the results and display the fault prediction results in a visual manner for operation and maintenance personnel to troubleshoot and repair.

10. A method for locating faults in a power grid information system based on adaptive learning according to claim 7, characterized in that: The feedback and iteration in S6 includes the following steps: S601: Collect feedback, collect feedback from operation and maintenance personnel on the fault location result; S602: Data labeling: labeling the newly added data based on the feedback results for further training and optimization of the model; S603: Iterative update: based on feedback results and new data, the model is continuously updated iteratively.

Images