Power grid fault intelligent diagnosis and suggestion generation system and method

Abstract

The invention discloses a power grid fault intelligent diagnosis and suggestion generation system and method, and belongs to the technical field of fault diagnosis. The output end of a data acquisition module is connected to the input end of a fault diagnosis module; the output end of the fault diagnosis module is connected to the input end of the fault positioning module and the input end of the suggestion generation module; the output end of the fault positioning module is connected to the input end of the suggestion generation module and the input end of the visualization module; the output end of the suggestion generation module is connected to the input end of the visualization module; and the visualization module is configured to receive and integrate the fault type transmitted by the fault diagnosis module, the fault position transmitted by the fault positioning module and the maintenance suggestion transmitted by the suggestion generation module, and visually display the fault type, the fault position and the maintenance suggestion on a preset power grid topological graph. According to the method, the fault type, the corresponding maintenance suggestion and the fault position are displayed on the graph, so that operation and maintenance personnel can quickly know the fault condition of the power grid, and a more accurate decision can be made.

Description

Technical Field

[0001] This invention relates to the field of fault diagnosis technology, specifically to a system and method for intelligent diagnosis and suggestion generation of power grid faults. Background Technology

[0002] In smart grids, diagnosing the location of early faults is crucial for maintaining power supply reliability and conducting fault analysis. Once a fault occurs in the power grid, it propagates rapidly within the distributed system and can potentially cause secondary faults. With the development of modern society, higher demands are being placed on power systems to improve power supply reliability.

[0003] Currently, fault detection and diagnosis are widely used in smart grids. One approach is to use phasor measurement units (PMUs) to detect fault locations in the smart grid. However, PMU measurement costs are relatively high, and in practice, only a small number of PMUs are typically used. Then, Ohm's law is applied to calculate the voltage of other grid components. Another approach is intelligent detection based on support vector machines (SVMs), decision trees, etc., which train artificial intelligence models using data generated by node operation as training samples. However, the data generated by node operation consists of a large amount of unlabeled data, leading to less than ideal accuracy in fault diagnosis for the trained models. Regarding the visualization of grid faults, existing systems often only provide relatively simple grid topology diagrams, failing to intuitively display key information such as fault type, fault location, and maintenance recommendations, posing challenges for maintenance personnel in quickly and accurately grasping the fault situation. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by this invention is: how to address the fact that existing systems often only provide relatively simple power grid topology diagrams, and cannot intuitively display key information such as fault type, fault location, and maintenance suggestions on the diagram, which brings certain difficulties to operation and maintenance personnel in quickly and accurately grasping the fault situation.

[0006] To address the aforementioned technical problems, this invention provides the following technical solution: a power grid fault intelligent diagnosis and suggestion generation system, comprising a data acquisition module, a fault diagnosis module, a fault location module, a suggestion generation module, and a visualization module; the output of the data acquisition module is connected to the input of the fault diagnosis module, for transmitting the acquired power grid equipment operation data to the fault diagnosis module in real time; the output of the fault diagnosis module is connected to the inputs of the fault location module and the suggestion generation module, for transmitting the identified fault types in parallel to the fault location module and the suggestion generation module; the output of the fault location module is connected to the inputs of the suggestion generation module and the visualization module, for transmitting the located fault locations in parallel to the suggestion generation module and the visualization module; the output of the suggestion generation module is connected to the input of the visualization module, for transmitting the generated maintenance suggestions to the visualization module; the visualization module is configured to receive and integrate the fault types transmitted by the fault diagnosis module, the fault locations transmitted by the fault location module, and the maintenance suggestions transmitted by the suggestion generation module, and to visualize them on a preset power grid topology map.

[0007] As a preferred embodiment of the intelligent fault diagnosis and suggestion generation system for power grids described in this invention, the system comprises: a data acquisition module for collecting operating data of power grid equipment, including current, voltage, active power, reactive power, temperature, protection device action signals, and equipment status alarm signals; a fault diagnosis module for identifying fault types based on the operating data using a pre-trained fault identification model; a fault location module for acquiring a power grid topology map and locating the fault location using a particle swarm optimization algorithm based on the power grid topology map and the operating data; a suggestion generation module for generating maintenance suggestions based on the fault type, the fault location, and a preset fault maintenance strategy knowledge base; and a visualization module for loading a preset power grid topology map and displaying the fault type, fault location, and maintenance suggestions on the power grid topology map.

[0008] As a preferred embodiment of the intelligent fault diagnosis and suggestion generation system for power grids described in this invention, the fault diagnosis module includes a data preprocessing submodule, a feature extraction submodule, a fault identification module training submodule, and a fault classification submodule. The data preprocessing submodule filters and normalizes the collected operating data to remove noise and outliers. The feature extraction submodule extracts fault features from the processed operating data, including current surge features, voltage over-limit features, power anomaly features, temperature anomaly features, protection device action signal features, and equipment status alarm signal features. The fault identification module training submodule acquires historical operating data and extracts historical fault features from it. These historical fault features are used as a training sample set. An RNN is employed, and the training sample set is input into an untrained fault identification model. Based on the supervision labels representing faults, the model is trained to convergence, enabling it to identify fault types. The fault classification submodule inputs the fault features from the feature extraction submodule into the trained fault identification model to identify the fault type.

[0009] In a preferred embodiment of the intelligent power grid fault diagnosis and suggestion generation system of the present invention: the input end of the data preprocessing submodule is connected to the output end of the data acquisition module, for receiving the operating data, filtering and normalizing the operating data, and transmitting the processed operating data to the feature extraction submodule; the input end of the feature extraction submodule is connected to the output end of the data preprocessing submodule, for extracting fault features from the processed operating data, and transmitting the fault features to the fault classification submodule; the input end of the fault classification submodule is connected to the output end of the feature extraction submodule, for inputting the fault features into a pre-trained fault identification model, identifying the fault type, and outputting the fault type to the output end of the fault diagnosis module; the fault identification module training submodule operates offline, for training the fault identification model, and transmitting the trained model parameters to the fault classification submodule for the fault classification submodule to use.

[0010] As a preferred embodiment of the intelligent power grid fault diagnosis and suggestion generation system of the present invention, the fault location module includes a data acquisition submodule, a data extraction submodule, a chromosome encoding submodule, a fitness function construction submodule, a genetic evolution submodule, a convergence determination submodule, and a fault decoding submodule; the data acquisition submodule is used to acquire the power grid topology map and retrieve the line impedance parameters from a preset power grid topology parameter library; the data extraction submodule extracts the current mutation amplitude and voltage distortion rate based on the operating data; the chromosome encoding submodule is used to map the line nodes in the power grid topology map to chromosomes composed of binary gene sequences, each chromosome representing... The system identifies a potential fault location; the fitness function construction submodule constructs a fitness function based on the current fluctuation amplitude, voltage distortion rate, and line impedance parameters, with the goal of minimizing the electrical quantity deviation at the fault point; the genetic evolution submodule initializes the chromosome population and iteratively updates the population through roulette wheel selection, single-point crossover, and probabilistic mutation operations, retaining the chromosome with the highest fitness; the convergence determination submodule determines convergence when the fitness value of the chromosome with the highest fitness remains stable within a threshold range for N consecutive generations; and the fault decoding submodule decodes the gene sequence of the chromosome with the highest fitness into a power grid topology node number and outputs the fault location identifier in the topology map.

[0011] As a preferred embodiment of the intelligent power grid fault diagnosis and suggestion generation system of the present invention, wherein: the input end of the data acquisition submodule is connected to the output end of the data acquisition module, for acquiring the power grid topology map and retrieving line impedance parameters from a preset power grid topology parameter library, and transmitting the power grid topology map and line impedance parameters to the fitness function construction submodule and the chromosome encoding submodule; the input end of the data extraction submodule is connected to the output end of the data acquisition module, for extracting current fluctuation amplitude and voltage distortion rate from the operating data, and transmitting the current fluctuation amplitude and voltage distortion rate to the fitness function construction submodule; the input end of the chromosome encoding submodule is connected to the output end of the data acquisition submodule, for mapping line nodes in the power grid topology map to chromosomes, and transmitting chromosome data to the genetic evolution submodule; the input end of the fitness function construction submodule is connected to... The data acquisition submodule and the data extraction submodule outputs are used to construct a fitness function based on the current fluctuation amplitude, voltage distortion rate, and line impedance parameters, and transmit the fitness function to the genetic evolution submodule. The input of the genetic evolution submodule is connected to the output of the chromosome encoding submodule and the fitness function construction submodule, used to initialize the chromosome population, iteratively update the population through genetic operations, and transmit the chromosome with the highest fitness during the iteration process to the convergence determination submodule. The input of the convergence determination submodule is connected to the output of the genetic evolution submodule, used to determine whether the population has converged, and transmit the optimal chromosome after convergence to the fault decoding submodule. The input of the fault decoding submodule is connected to the output of the convergence determination submodule, used to decode the optimal chromosome into a fault location, and output the fault location to the output of the fault location module.

[0012] As a preferred embodiment of the intelligent power grid fault diagnosis and suggestion generation system of the present invention, the suggestion generation module includes a fault maintenance strategy knowledge base construction submodule, a matching submodule, an information packaging submodule, and a notification routing submodule. The fault maintenance strategy knowledge base construction submodule is used to construct a fault maintenance strategy knowledge base, including fault types and a three-dimensional response set corresponding to the fault types. The three-dimensional response set includes personnel operation procedures, equipment replacement lists, and safety isolation steps. The matching submodule is used to match the corresponding maintenance strategy in the fault maintenance strategy knowledge base according to the fault type output by the fault diagnosis module, as a maintenance suggestion. The information packaging submodule is used to encapsulate the matched maintenance suggestion and the fault location coordinates output by the fault location module into a structured alarm information package. The notification routing submodule is used to automatically associate the maintenance team communication terminal address according to the power grid management responsibility area corresponding to the fault location coordinates, and push the alarm information package to the target terminal.

[0013] As a preferred embodiment of the intelligent power grid fault diagnosis and suggestion generation system of the present invention, the following is provided: the fault maintenance strategy knowledge base construction submodule operates offline, used to construct and update the fault maintenance strategy knowledge base, and store the knowledge base data locally for the matching submodule to call; the input end of the matching submodule is connected to the output end of the fault diagnosis module and the output end of the fault maintenance strategy knowledge base construction submodule, used to match the corresponding maintenance strategy from the knowledge base according to the fault type, and transmit the matched maintenance strategy to the information packaging submodule; the input end of the information packaging submodule is connected to the output end of the matching submodule and the output end of the fault location module, used to encapsulate the maintenance strategy and the fault location into a structured alarm information package, and transmit the alarm information package to the notification routing submodule; the input end of the notification routing submodule is connected to the output end of the information packaging submodule, used to associate the maintenance team communication terminal address according to the fault location coordinates, and push the alarm information package to the target terminal.

[0014] As a preferred embodiment of the intelligent power grid fault diagnosis and suggestion generation system of the present invention, the visualization module includes a topology rendering submodule, a fault marking submodule, and a view zooming operation submodule; the topology rendering submodule is used to load a preset power grid topology map, which is constructed based on electrical connection relationships and includes symbolic nodes and connecting edges of substations, buses, lines, and switching equipment; the fault marking submodule is used to mark the topology node number position of the fault output by the fault location module on the topology map, and associate a floating window at the topology node number position, with the first layer displaying the fault type and the second layer displaying the maintenance suggestions corresponding to the fault type; the view zooming operation submodule is used to respond to the power grid topology map zooming operation and dynamically adjust the fault information display. Granularity; the input of the topology rendering submodule is connected to a preset power grid topology database for loading the power grid topology and transmitting the rendered topology data to the fault marking submodule; the input of the fault marking submodule is connected to the output of the topology rendering submodule, the output of the fault diagnosis module, the output of the fault location module, and the output of the suggestion generation module, for marking fault locations on the topology map, associating and displaying fault types and maintenance suggestions, and transmitting the marked topology data to the view zooming operation submodule; the input of the view zooming operation submodule is connected to the output of the fault marking submodule, for responding to zooming operations, dynamically adjusting the granularity of fault information display, and outputting the final visualization interface.

[0015] This invention provides a method for intelligent diagnosis and suggestion generation of power grid faults.

[0016] To address the aforementioned technical problems, this invention provides the following technical solution: a method for intelligent diagnosis and suggestion generation of power grid faults, comprising: collecting operating data of power grid equipment, including current, voltage, active power, reactive power, temperature, protection device action signals, and equipment status alarm signals; identifying fault types based on the operating data using a pre-trained fault identification model; acquiring a power grid topology map and locating fault locations using a particle swarm optimization algorithm based on the power grid topology map and operating data; generating maintenance suggestions based on the fault type, fault location, and a preset fault maintenance strategy knowledge base; loading a preset power grid topology map and displaying the fault type, fault location, and maintenance suggestions on the power grid topology map.

[0017] The beneficial effects of this invention are as follows: This invention collects operating data of various power grid equipment through a data acquisition module, and then uses a fault identification model trained based on algorithms such as RNN to quickly identify the fault type, effectively improving the efficiency and accuracy of power grid fault diagnosis and reducing the time and manpower costs of fault diagnosis; by adopting the particle swarm optimization algorithm, combined with the power grid topology map and operating data, the fault location can be accurately located, which helps to shorten the power outage time and reduce the impact on users.

[0018] The suggestion generation module can quickly generate corresponding maintenance suggestions based on the fault type, fault location, and a preset fault maintenance strategy knowledge base. Its fault maintenance strategy knowledge base covers various fault types and corresponding three-dimensional response measures, including personnel operating procedures, equipment replacement lists, and safety isolation procedures. This provides maintenance personnel with comprehensive and specific guidance, improving the pertinence and effectiveness of maintenance work and ensuring the safety and standardization of the maintenance process.

[0019] The visualization module loads a pre-set power grid topology map and displays information such as fault type, corresponding maintenance suggestions, and fault location in a floating window at different levels on the map. It also supports view zooming and dynamically adjusts the granularity of fault information display, including global view, local view, and device view. This allows maintenance personnel to quickly understand the overall situation and detailed information of power grid faults, which helps them make more accurate decisions.

[0020] By notifying the routing unit to automatically associate the communication terminal address of the maintenance team with the power grid management responsibility area corresponding to the fault location coordinates, and pushing the structured alarm information package containing maintenance suggestions and fault location to the target terminal, the fault information can be quickly transmitted and accurately notified, ensuring that maintenance personnel can receive the fault information in a timely manner, respond quickly and carry out maintenance work, thereby improving the overall efficiency of power grid operation and maintenance. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of a power grid fault intelligent diagnosis and suggestion generation system provided in one embodiment of the present invention.

[0023] Figure 2 The above is a flowchart of the intelligent diagnosis and suggestion generation method for power grid faults provided in one embodiment of the present invention. Detailed Implementation

[0024] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0025] Example 1, referring to Figure 1 This is one embodiment of the present invention, which provides a power grid fault intelligent diagnosis and suggestion generation system, including: The system includes a data acquisition module, a fault diagnosis module, a fault location module, a suggestion generation module, and a visualization module.

[0026] The data acquisition module is used to collect operating data of power grid equipment, including current, voltage, active power, reactive power, temperature, protection device action signals, and equipment status alarm signals.

[0027] The fault diagnosis module is used to identify the fault type based on the running data using a pre-trained fault identification model.

[0028] The fault location module is used to acquire the power grid topology map and uses the particle swarm optimization algorithm to locate the fault location based on the power grid topology map and operating data.

[0029] The suggestion generation module is used to generate maintenance suggestions based on the fault type, fault location, and a preset fault maintenance strategy knowledge base.

[0030] The visualization module is used to load a preset power grid topology map and display the fault type, fault location, and maintenance suggestions on the power grid topology map.

[0031] The output of the data acquisition module is connected to the input of the fault diagnosis module, which is used to transmit the collected power grid equipment operation data to the fault diagnosis module in real time.

[0032] The output of the fault diagnosis module is connected to the input of the fault location module and the input of the suggestion generation module, and is used to transmit the identified fault types to the fault location module and the suggestion generation module in parallel.

[0033] The output of the fault location module is connected to the input of the suggestion generation module and the input of the visualization module, and is used to transmit the located fault location to the suggestion generation module and the visualization module in parallel.

[0034] It is recommended that the output of the generation module be connected to the input of the visualization module to transmit the generated maintenance suggestions to the visualization module.

[0035] The visualization module is configured to receive and integrate the fault types transmitted by the fault diagnosis module, the fault locations transmitted by the fault location module, and the maintenance suggestions transmitted by the suggestion generation module, and then visualize them on a preset power grid topology map.

[0036] The intelligent power grid fault diagnosis and suggestion generation system provided by this invention can quickly identify fault types through operational data and fault identification models, effectively improving the efficiency and accuracy of power grid fault diagnosis. By employing a particle swarm optimization algorithm combined with power grid topology maps and operational data, it can accurately locate fault positions, helping to shorten power outage time and reduce the impact on users. The suggestion generation module can quickly generate corresponding maintenance suggestions based on fault type, fault location, and a preset fault maintenance strategy knowledge base. This provides maintenance personnel with comprehensive and specific guidance, improving the pertinence and effectiveness of maintenance work and ensuring the safety and standardization of the maintenance process. The visualization module loads a preset power grid topology map and displays information such as fault type, corresponding maintenance suggestions, and fault location on the map, facilitating maintenance personnel to quickly understand the overall situation and detailed information of power grid faults, helping to make more accurate decisions.

[0037] The fault diagnosis module includes a data preprocessing submodule, a feature extraction submodule, a fault identification module training submodule, and a fault classification submodule.

[0038] The data preprocessing submodule is used to filter and normalize the collected running data to remove noise and outliers.

[0039] The feature extraction submodule is used to extract fault features from the processed operating data, including current surge features, voltage over-limit features, power anomaly features, temperature anomaly features, protection device action signal features, and equipment status alarm signal features.

[0040] The fault identification module training submodule is used to acquire historical operating data and extract historical fault features from the historical operating data. The historical fault features are used as the training sample set. A recurrent neural network (RNN) is used to input the training sample set into the untrained fault identification model. Based on the supervision labels representing faults, the model is trained to a convergent state, enabling it to identify fault types.

[0041] The fault classification submodule is used to input the fault features from the feature extraction submodule into the trained fault identification model to identify the fault type.

[0042] The input of the data preprocessing submodule is connected to the output of the data acquisition module. It is used to receive the running data, filter and normalize the running data, and transmit the processed running data to the feature extraction submodule.

[0043] The input of the feature extraction submodule is connected to the output of the data preprocessing submodule. It is used to extract fault features from the processed running data and transmit the fault features to the fault classification submodule.

[0044] The input of the fault classification submodule is connected to the output of the feature extraction submodule. It is used to input fault features into the pre-trained fault identification model, identify the fault type, and output the fault type to the output of the fault diagnosis module.

[0045] The fault identification module training submodule operates offline and is used to train the fault identification model. The trained model parameters are then transferred to the fault classification submodule for it to use.

[0046] The fault types include short circuit faults, open circuit faults, internal equipment faults, grounding faults, overload / overheating faults, malfunctions / failures of protection devices, and switchgear faults.

[0047] The fault location module includes a data acquisition submodule, a data extraction submodule, a chromosome encoding submodule, a fitness function construction submodule, a genetic evolution submodule, a convergence determination submodule, and a fault decoding submodule.

[0048] The data acquisition submodule is used to acquire the power grid topology map and retrieve the line impedance parameters from the preset power grid topology parameter library.

[0049] The data extraction submodule extracts the current surge amplitude and voltage distortion rate from the running data.

[0050] The chromosome encoding submodule is used to map line nodes in the power grid topology diagram to chromosomes composed of binary gene sequences, with each chromosome representing a potential fault location.

[0051] The fitness function construction submodule is used to construct a fitness function with the goal of minimizing the electrical quantity deviation at the fault point, based on the current surge amplitude, voltage distortion rate, and line impedance parameters.

[0052] The genetic evolution submodule is used to initialize the chromosome population. It iteratively updates the population through roulette wheel selection, single-point crossover, and probabilistic mutation operations, retaining the chromosome with the highest fitness.

[0053] The convergence determination submodule is used to determine convergence when the fitness value of the chromosome with the highest fitness remains stable within the threshold range for N consecutive generations.

[0054] The fault decoding submodule is used to decode the chromosome gene sequence with the highest fitness into a power grid topology node number and output the topology map location identifier of the fault.

[0055] The input of the data acquisition submodule is connected to the output of the data acquisition module. It is used to acquire the power grid topology map and retrieve the line impedance parameters from the preset power grid topology parameter library. The power grid topology map and line impedance parameters are then transmitted to the fitness function construction submodule and the chromosome coding submodule.

[0056] The input of the data extraction submodule is connected to the output of the data acquisition module. It is used to extract the current fluctuation amplitude and voltage distortion rate from the running data and transmit the current fluctuation amplitude and voltage distortion rate to the fitness function construction submodule.

[0057] The input of the chromosome encoding submodule is connected to the output of the data acquisition submodule. It is used to map the line nodes in the power grid topology diagram to chromosomes and transmit the chromosome data to the genetic evolution submodule.

[0058] The input of the fitness function construction submodule is connected to the output of the data acquisition submodule and the data extraction submodule. It is used to construct a fitness function based on the current mutation amplitude, voltage distortion rate and line impedance parameters, and then transmit the fitness function to the genetic evolution submodule.

[0059] The input of the genetic evolution submodule is connected to the output of the chromosome encoding submodule and the fitness function construction submodule. It is used to initialize the chromosome population and iteratively update the population through genetic operations. The chromosome with the highest fitness during the iteration process is transferred to the convergence determination submodule.

[0060] The input of the convergence determination submodule is connected to the output of the genetic evolution submodule. It is used to determine whether the population has converged and to transmit the optimal chromosome after convergence to the fault decoding submodule.

[0061] The input of the fault decoding submodule is connected to the output of the convergence determination submodule. It is used to decode the optimal chromosome into the fault location and output the fault location to the output of the fault location module.

[0062] The recommended modules include a fault repair strategy knowledge base construction submodule, a matching submodule, an information packaging submodule, and a notification routing submodule.

[0063] The Fault Repair Strategy Knowledge Base Construction Submodule is used to build a fault repair strategy knowledge base, including fault types, a set of three-dimensional response measures corresponding to each fault type, and the set of three-dimensional response measures including personnel operating procedures, equipment replacement lists, and safety isolation procedures.

[0064] The matching submodule is used to match the corresponding maintenance strategy in the fault maintenance strategy knowledge base based on the fault type output by the fault diagnosis module, and to provide maintenance suggestions.

[0065] The information packaging submodule is used to encapsulate the matched maintenance suggestions and the fault location coordinates output by the fault location module into a structured alarm information package.

[0066] The notification routing submodule is used to automatically associate the communication terminal address of the maintenance team with the power grid management responsibility area corresponding to the fault location coordinates and push the alarm information packet to the target terminal.

[0067] The fault repair strategy knowledge base construction submodule operates offline and is used to build and update the fault repair strategy knowledge base, and store the knowledge base data locally for use by matching submodules.

[0068] The input of the matching submodule is connected to the output of the fault diagnosis module and the output of the fault repair strategy knowledge base construction submodule. It is used to match the corresponding repair strategy from the knowledge base according to the fault type and transmit the matched repair strategy to the information packaging submodule.

[0069] The input of the information packaging submodule is connected to the output of the matching submodule and the output of the fault location module. It is used to encapsulate the maintenance strategy and fault location into a structured alarm information packet and transmit the alarm information packet to the notification routing submodule.

[0070] The input of the notification routing submodule is connected to the output of the information packaging submodule. It is used to associate the communication terminal address of the maintenance team with the fault location coordinates and push the alarm information packet to the target terminal.

[0071] The visualization module includes a topology rendering submodule, a fault marking submodule, and a view zoom operation submodule.

[0072] The topology rendering submodule is used to load a pre-built power grid topology. The power grid topology is built on electrical connection relationships and includes symbolic nodes and connecting edges of substations, buses, lines and switching equipment.

[0073] The fault marking submodule is used to mark the topology node number of the fault output by the fault location module on the topology map, and associate a floating window with the topology node number. The first layer displays the fault type; the second layer displays the maintenance suggestions corresponding to the fault type.

[0074] For example, when the fault location module outputs SSE-220kV-BAY-A17, the visualization module will add a red halo to node 17 in area A of the 220kV zone in the SSE (substation) to indicate a fault alarm.

[0075] The view zoom operation submodule is used to respond to the power grid topology map zoom operation and dynamically adjust the granularity of fault information display.

[0076] The granularity of fault information display includes global view, local view and device view; global view: displays the location of the fault node in the power grid topology diagram; local view: expands the wiring diagram of the bay where the fault node is located; device view: overlays the operating data and fault type of the faulty device.

[0077] The input of the topology rendering submodule is connected to a preset power grid topology database to load the power grid topology and transmit the rendered topology data to the fault marking submodule.

[0078] The input of the fault marking submodule is connected to the output of the topology rendering submodule, the fault diagnosis module, the fault location module, and the suggestion generation module. It is used to mark the fault location on the topology map, and to display the fault type and maintenance suggestions. The marked topology map data is then transmitted to the view zoom operation submodule.

[0079] The input of the view zoom operation submodule is connected to the output of the fault marker submodule. It is used to respond to zoom operations, dynamically adjust the granularity of fault information display, and output the final visualization interface.

[0080] Example 2, refer to Figure 2 As an embodiment of the present invention, based on the previous embodiment, a method for intelligent diagnosis and suggestion generation of power grid faults is provided, including: S1. Collect operating data of power grid equipment, including current, voltage, active power, reactive power, temperature, protection device action signals, and equipment status alarm signals.

[0081] S2. Based on the operational data, identify the fault type using a pre-trained fault identification model.

[0082] S3. Obtain the power grid topology map and use the particle swarm optimization algorithm to locate the fault location based on the power grid topology map and operating data.

[0083] S4. Generate maintenance suggestions based on the fault type, fault location, and preset fault maintenance strategy knowledge base.

[0084] S5. Load the preset power grid topology map and display the fault type, fault location and maintenance suggestions on the power grid topology map.

[0085] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A power grid fault intelligent diagnosis and suggestion generation system, characterized in that: include, Data acquisition module, fault diagnosis module, fault location module, suggestion generation module, and visualization module; The output of the data acquisition module is connected to the input of the fault diagnosis module, and is used to transmit the collected power grid equipment operation data to the fault diagnosis module in real time. The output of the fault diagnosis module is connected to the input of the fault location module and the input of the suggestion generation module, and is used to transmit the identified fault types to the fault location module and the suggestion generation module in parallel. The output of the fault location module is connected to the input of the suggestion generation module and the input of the visualization module, and is used to transmit the located fault location to the suggestion generation module and the visualization module in parallel. The output of the suggestion generation module is connected to the input of the visualization module, and is used to transmit the generated maintenance suggestions to the visualization module; The visualization module is configured to receive and integrate the fault type transmitted by the fault diagnosis module, the fault location transmitted by the fault location module, and the maintenance suggestions transmitted by the suggestion generation module, and to visualize them on a preset power grid topology map.

2. The intelligent power grid fault diagnosis and suggestion generation system as described in claim 1, characterized in that: The data acquisition module is used to collect operating data of power grid equipment, including current, voltage, active power, reactive power, temperature, protection device action signals, and equipment status alarm signals. The fault diagnosis module is used to identify the fault type based on the operating data using a pre-trained fault identification model; The fault location module is used to acquire a power grid topology map and use a particle swarm optimization algorithm to locate the fault location based on the power grid topology map and the operating data. The suggestion generation module is used to generate maintenance suggestions based on the fault type, the fault location, and a preset fault maintenance strategy knowledge base; The visualization module is used to load a preset power grid topology map and display the fault type, fault location, and maintenance suggestions on the power grid topology map.

3. The intelligent power grid fault diagnosis and suggestion generation system as described in claim 2, characterized in that: The fault diagnosis module includes a data preprocessing submodule, a feature extraction submodule, a fault identification module training submodule, and a fault classification submodule. The data preprocessing submodule is used to filter and normalize the collected running data to remove noise and outliers; The feature extraction submodule is used to extract fault features from the processed operating data, including current surge features, voltage over-limit features, power anomaly features, temperature anomaly features, protection device action signal features, and equipment status alarm signal features. The fault identification module training submodule is used to acquire historical operating data and extract historical fault features from the historical operating data. The historical fault features are used as training sample set. An RNN is used to input the training sample set into the untrained fault identification model. Based on the supervision label representing the fault, the model is trained to a convergent state, so that it has the ability to identify the fault type. The fault classification submodule is used to input the fault features from the feature extraction submodule into the trained fault identification model to identify the fault type.

4. The intelligent power grid fault diagnosis and suggestion generation system as described in claim 3, characterized in that: The input end of the data preprocessing submodule is connected to the output end of the data acquisition module, and is used to receive the running data, filter and normalize the running data, and transmit the processed running data to the feature extraction submodule. The input of the feature extraction submodule is connected to the output of the data preprocessing submodule, and is used to extract fault features from the processed running data and transmit the fault features to the fault classification submodule. The input end of the fault classification submodule is connected to the output end of the feature extraction submodule, which is used to input the fault features into the pre-trained fault identification model, identify the fault type, and output the fault type to the output end of the fault diagnosis module. The fault identification module training submodule operates offline to train the fault identification model and transmits the trained model parameters to the fault classification submodule for the fault classification submodule to call.

5. The intelligent power grid fault diagnosis and suggestion generation system as described in claim 4, characterized in that: The fault location module includes a data acquisition submodule, a data extraction submodule, a chromosome encoding submodule, a fitness function construction submodule, a genetic evolution submodule, a convergence determination submodule, and a fault decoding submodule. The data acquisition submodule is used to acquire the power grid topology map and retrieve the line impedance parameters from the preset power grid topology parameter library; The data extraction submodule extracts the current surge amplitude and voltage distortion rate from the operating data; The chromosome coding submodule is used to map the line nodes in the power grid topology diagram to chromosomes composed of binary gene sequences, with each chromosome representing a potential fault location; The fitness function construction submodule is used to construct a fitness function with the goal of minimizing the electrical quantity deviation at the fault point, based on the current mutation amplitude, the voltage distortion rate, and the line impedance parameters. The genetic evolution submodule is used to initialize the chromosome population, and iteratively updates the population through roulette wheel selection, single-point crossover and probabilistic mutation operations, retaining the chromosome with the highest fitness. The convergence determination submodule is used to determine convergence when the fitness value of the chromosome with the highest fitness remains stable within the threshold range for N consecutive generations. The fault decoding submodule is used to decode the chromosome gene sequence with the highest fitness into a power grid topology node number and output the topology map location identifier of the fault.

6. The intelligent power grid fault diagnosis and suggestion generation system as described in claim 5, characterized in that: The input end of the data acquisition submodule is connected to the output end of the data acquisition module, and is used to acquire the power grid topology map and retrieve the line impedance parameters from the preset power grid topology parameter library, and transmit the power grid topology map and line impedance parameters to the fitness function construction submodule and the chromosome encoding submodule. The input of the data extraction submodule is connected to the output of the data acquisition module, and is used to extract the current fluctuation amplitude and voltage distortion rate from the running data, and transmit the current fluctuation amplitude and voltage distortion rate to the fitness function construction submodule. The input of the chromosome encoding submodule is connected to the output of the data acquisition submodule, and is used to map the line nodes in the power grid topology diagram to chromosomes and transmit the chromosome data to the genetic evolution submodule. The input of the fitness function construction submodule is connected to the output of the data acquisition submodule and the data extraction submodule. It is used to construct a fitness function based on the current mutation amplitude, voltage distortion rate and line impedance parameters, and transmit the fitness function to the genetic evolution submodule. The input of the genetic evolution submodule is connected to the output of the chromosome encoding submodule and the fitness function construction submodule. It is used to initialize the chromosome population and iteratively update the population through genetic operations. The chromosome with the highest fitness during the iteration process is transmitted to the convergence determination submodule. The input of the convergence determination submodule is connected to the output of the genetic evolution submodule, and is used to determine whether the population has converged and transmit the optimal chromosome after convergence to the fault decoding submodule. The input of the fault decoding submodule is connected to the output of the convergence determination submodule, and is used to decode the optimal chromosome into a fault location and output the fault location to the output of the fault location module.

7. The intelligent power grid fault diagnosis and suggestion generation system as described in claim 6, characterized in that: The suggestion generation module includes a fault repair strategy knowledge base construction submodule, a matching submodule, an information packaging submodule, and a notification routing submodule. The fault repair strategy knowledge base construction submodule is used to build a fault repair strategy knowledge base, including fault types and a set of three-dimensional response measures corresponding to the fault types. The set of three-dimensional response measures includes personnel operation procedures, equipment replacement lists, and safety isolation steps. The matching submodule is used to match the corresponding maintenance strategy in the fault maintenance strategy knowledge base according to the fault type output by the fault diagnosis module, as a maintenance suggestion; The information packaging submodule is used to encapsulate the matched maintenance suggestions and the fault location coordinates output by the fault location module into a structured alarm information package; The notification routing submodule is used to automatically associate the communication terminal address of the maintenance team with the power grid management responsibility area corresponding to the fault location coordinates and push the alarm information packet to the target terminal.

8. The intelligent power grid fault diagnosis and suggestion generation system as described in claim 7, characterized in that: The fault repair strategy knowledge base construction submodule operates offline and is used to build and update the fault repair strategy knowledge base, and store the knowledge base data locally for the matching submodule to call. The input of the matching submodule is connected to the output of the fault diagnosis module and the output of the fault repair strategy knowledge base construction submodule. It is used to match the corresponding repair strategy from the knowledge base according to the fault type and transmit the matched repair strategy to the information packaging submodule. The input of the information packaging submodule is connected to the output of the matching submodule and the output of the fault location module, and is used to encapsulate the maintenance strategy and the fault location into a structured alarm information package, and transmit the alarm information package to the notification routing submodule. The input of the notification routing submodule is connected to the output of the information packaging submodule, and is used to associate the communication terminal address of the maintenance team according to the fault location coordinates and push the alarm information package to the target terminal.

9. The intelligent power grid fault diagnosis and suggestion generation system as described in claim 8, characterized in that: The visualization module includes a topology rendering submodule, a fault marking submodule, and a view zoom operation submodule. The topology rendering submodule is used to load a preset power grid topology diagram, which is constructed based on electrical connection relationships and includes symbolic nodes and connection edges of substations, buses, lines and switching equipment. The fault marking submodule is used to mark the topology node number position of the fault output by the fault location module on the topology map, and associate a floating window with the topology node number position, with the fault type displayed in the first layer; The second layer displays repair suggestions corresponding to the fault type; The view zoom operation submodule is used to respond to the power grid topology map zoom operation and dynamically adjust the granularity of fault information display. The input terminal of the topology rendering submodule is connected to a preset power grid topology database, which is used to load the power grid topology and transmit the rendered topology data to the fault marking submodule. The input of the fault marking submodule is connected to the output of the topology rendering submodule, the output of the fault diagnosis module, the output of the fault location module, and the output of the suggestion generation module. It is used to mark the fault location on the topology map, associate and display the fault type and maintenance suggestions, and transmit the marked topology map data to the view zoom operation submodule. The input of the view zoom operation submodule is connected to the output of the fault marker submodule, which is used to respond to the zoom operation, dynamically adjust the granularity of the fault information display, and output the final visualization interface.

10. A method for intelligent diagnosis and suggestion generation of power grid faults, using the intelligent diagnosis and suggestion generation system for power grid faults as described in any one of claims 1 to 7, characterized in that, include: Collect operational data of power grid equipment, including current, voltage, active power, reactive power, temperature, protection device action signals, and equipment status alarm signals; Based on operational data, the fault type is identified using a pre-trained fault identification model; Obtain the power grid topology map and use the particle swarm optimization algorithm to locate the fault location based on the power grid topology map and operating data; Based on the fault type, fault location, and a pre-defined fault repair strategy knowledge base, repair suggestions are generated. Load a preset power grid topology map and display the fault type, fault location, and maintenance suggestions on the power grid topology map.

Images