Power grid fault diagnosis method and device, storage medium and equipment

Abstract

The invention provides a power grid fault diagnosis method and device, a storage medium and equipment. The method comprises the steps that power grid operation parameters of power grid key nodes are acquired through a power grid node sensor; sending the power grid operation parameters to an edge computing node to judge whether the power grid operation parameters are abnormal or not; if the power grid operation parameters are abnormal, marking the power grid operation parameters to be abnormal and then sending the power grid operation parameters to a cloud platform; and the cloud platform classifies the power grid operation parameters marked to be abnormal in combination with historical power grid fault data, and positions the power grid operation parameters marked to be abnormal in combination with a GIS and a GPS to obtain a power grid fault type and position. Through combination of the edge computing node and the cloud platform, power grid fault abnormity is identified, analyzed and diagnosed in stages, rapid fault response to power grid abnormal operation parameters is realized, power grid fault features can be accurately identified and analyzed in combination with a trained model, the power grid fault diagnosis efficiency and accuracy are improved, and the fault diagnosis efficiency and accuracy are improved. Powerful guarantee is provided for safe and stable operation of a power grid.

Description

Technical Field

[0001] The present invention relates to the field of electric power, and in particular to a method, device, storage medium and equipment for diagnosing faults in a power grid. Background Art

[0002] As the core component of the power system, the power grid is responsible for transmitting the electricity generated by power plants to end users through transmission lines, substations and distribution networks. It covers multiple links such as power generation, transmission, substations, distribution and power consumption, and is a complex and huge system. The stable operation of the power grid is crucial to the normal development of the social economy and the stability of people's lives.

[0003] A power grid failure refers to an abnormal state that occurs during the operation of the power grid, which may cause the interruption of power supply or degradation of quality. These failures are caused by a variety of reasons, including extreme weather conditions such as lightning strikes, storms, ice and snow, which damage power grid equipment or cause short circuits. Equipment in the power grid, such as transformers, lines, switches, etc., may experience performance degradation or damage over time, leading to failures. Sudden changes in the grid load, such as a large number of users turning on electrical equipment at the same time, may cause grid voltage fluctuations or frequency deviations, which in turn cause failures.

[0004] Traditional power grid fault diagnosis and location solutions mainly rely on methods such as manual inspection, fault recording analysis, and expert experience judgment. However, manual inspection takes a lot of time and manpower, and it is difficult to cover all power grid equipment in real time. Fault recording analysis may be interfered by various factors, resulting in inaccurate diagnosis results. Expert experience judgment is limited by the expert's personal knowledge level and experience accumulation, and it is difficult to respond to power grid faults quickly, which may cause the fault to expand or the scope of impact to increase. It lacks the support of intelligent means and cannot automatically analyze and predict power grid faults. It is necessary to propose a power grid fault diagnosis and location system and method supported by cloud technology to solve the above problems. Summary of the invention

[0005] Based on this, the present invention provides a power grid fault diagnosis method, device, storage medium and equipment. Through the combination of edge computing nodes and cloud platforms, power grid fault anomalies are identified and analyzed and diagnosed in stages. Through the high-bandwidth and low-latency communication capabilities of the cloud platform, rapid response to abnormal power grid operating parameters is achieved. In combination with the trained model, the characteristics of power grid faults can be accurately identified and analyzed, thereby improving the efficiency and accuracy of power grid fault diagnosis and providing strong guarantees for the safe and stable operation of the power grid.

[0006] In a first aspect, the present invention provides a power grid fault diagnosis method, which is applied to a power grid fault diagnosis system, wherein the power grid fault diagnosis system includes a power grid node sensor, an edge computing node, and a cloud platform, wherein the power grid node sensor is communicatively connected to the edge computing node, and the edge computing node is communicatively connected to the cloud platform, and the power grid fault diagnosis method includes:

[0007] Obtain grid operation parameters of key grid nodes through grid node sensors;

[0008] Sending the power grid operation parameters to the edge computing node to determine whether there are any abnormalities in the power grid operation parameters;

[0009] If the power grid operation parameters are abnormal, the power grid operation parameters are marked as abnormal and sent to the cloud platform;

[0010] The cloud platform classifies the abnormally marked power grid operation parameters in combination with historical power grid fault data, and locates the abnormally marked power grid operation parameters in combination with GIS and GPS to obtain the type and location of the power grid fault.

[0011] Furthermore, the power grid fault diagnosis method further includes:

[0012] The cloud platform obtains a solution to the power grid failure in combination with historical power grid failure data, and sends the solution to the power grid failure.

[0013] Furthermore, the power grid node sensor is also connected to the cloud platform for communication, and the power grid fault diagnosis method further includes:

[0014] The cloud platform receives the grid operation parameters sent by grid node sensors;

[0015] The cloud platform inputs the power grid operation parameters into the trained LSTM model to obtain a power grid fault prediction, and sends the power grid fault prediction.

[0016] Further, the key nodes of the power grid include power grid substations, distribution transformers, transmission line connection points, cable connectors and switchgear;

[0017] The grid operation parameters include at least one of current parameters, voltage parameters, power factor, active power, reactive power, frequency, harmonic content, phase angle, temperature, humidity and vibration parameters of key nodes of the grid.

[0018] Furthermore, the edge computing node determines whether there are any abnormalities in the power grid operation parameters, including:

[0019] The sliding window technology is used to process the power grid operation parameters in sections to obtain real-time power grid operation parameters;

[0020] Inputting the real-time power grid operation parameters into a fast Fourier transform model for spectrum analysis to obtain power grid operation characteristics;

[0021] The power grid operation characteristics are input into the trained isolation forest model for anomaly detection to obtain abnormal markers of power grid operation parameters.

[0022] Furthermore, the cloud platform classifies the abnormally marked power grid operation parameters in combination with the historical power grid fault data, specifically:

[0023] The cloud platform trains the convolutional neural network model in combination with historical power grid fault data until the loss of the trained convolutional neural network model converges and outputs the trained convolutional neural network model;

[0024] The abnormally marked power grid operation parameters are input into the convolutional neural network model to obtain the power grid fault type.

[0025] Furthermore, the cloud platform combines historical grid fault data to obtain solutions to grid faults, including:

[0026] The cloud platform performs structured processing on historical power grid fault data to obtain historical fault resolution feature information;

[0027] The cloud platform extracts fault feature information of abnormal power grid operation parameters;

[0028] Calculating the similarity between the fault feature information and any historical fault solution feature information until the similarity between the fault feature information and all historical fault solution feature information is obtained;

[0029] Selecting historical power grid fault data corresponding to historical fault resolution feature information with a similarity greater than a first threshold as a reference case;

[0030] A historical fault solution of the reference case is extracted as a solution to the power grid fault.

[0031] In a second aspect, the present invention further provides a power grid fault diagnosis device, which is applied to a power grid fault diagnosis system, wherein the power grid fault diagnosis system includes a power grid node sensor, an edge computing node, and a cloud platform, wherein the power grid node sensor is communicatively connected to the edge computing node, and the edge computing node is communicatively connected to the cloud platform, and the power grid fault diagnosis device includes:

[0032] A parameter acquisition module is used to acquire grid operation parameters of key grid nodes through grid node sensors;

[0033] An abnormality judgment module, used to send the power grid operation parameters to the edge computing node to determine whether there are abnormalities in the power grid operation parameters;

[0034] An abnormality uploading module is used to mark the abnormality of the power grid operation parameters and send them to the cloud platform if there are abnormalities in the power grid operation parameters;

[0035] The abnormality analysis module is used for the cloud platform to classify the abnormal power grid operation parameters combined with historical power grid fault data, and to locate the abnormal power grid operation parameters combined with GIS and GPS to obtain the type and location of the power grid fault.

[0036] In a third aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any one of the power grid fault diagnosis methods in the first aspect.

[0037] In a fourth aspect, the present invention further provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it executes any one of the power grid fault diagnosis methods in the first aspect.

[0038] The beneficial effects of adopting the above technical solution are as follows: this embodiment, through the combination of edge computing nodes and cloud platforms, identifies and analyzes power grid fault anomalies in stages, and uses the high-bandwidth, low-latency communication capabilities of the cloud platform to achieve rapid response to abnormal power grid operating parameters. In combination with the trained model, it can accurately identify and analyze power grid fault characteristics, improve the efficiency and accuracy of power grid fault diagnosis, and provide strong guarantees for the safe and stable operation of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art are briefly introduced below.

[0040] Figure 1 This is a schematic diagram of a power grid fault diagnosis method in one embodiment of the present application;

[0041] Figure 2 This is a schematic diagram of a power grid fault diagnosis device in one embodiment of the present application. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. In order to explain the present invention in more detail, the power grid fault diagnosis method, device, storage medium and equipment provided by the present invention are specifically described below in combination with the drawings.

[0043] Unless otherwise defined, the technical terms or scientific terms used in the disclosure of this application should be understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, similar words such as "one", "one" or "the" do not indicate a quantitative limit, but indicate that there is at least one. Similar words such as "include" or "include" mean that the elements or objects appearing in front of the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Similar words such as "connect" or "connected" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship can also change accordingly.

[0044] As the core component of the power system, the power grid is responsible for transmitting the electricity generated by power plants to end users through transmission lines, substations and distribution networks. It covers multiple links such as power generation, transmission, substations, distribution and power consumption, and is a complex and huge system. The stable operation of the power grid is crucial to the normal development of the social economy and the stability of people's lives.

[0045] A power grid failure refers to an abnormal state that occurs during the operation of the power grid, which may cause the interruption of power supply or degradation of quality. These failures are caused by a variety of reasons, including extreme weather conditions such as lightning strikes, storms, ice and snow, which damage power grid equipment or cause short circuits. Equipment in the power grid, such as transformers, lines, switches, etc., may experience performance degradation or damage over time, leading to failures. Sudden changes in the grid load, such as a large number of users turning on electrical equipment at the same time, may cause grid voltage fluctuations or frequency deviations, which in turn cause failures.

[0046] In this regard, the present invention provides a power grid fault diagnosis method, which identifies and analyzes power grid fault anomalies in stages through the combination of edge computing nodes and cloud platforms, and achieves rapid response to abnormal power grid operation parameters through the high bandwidth and low latency communication capabilities of the cloud platform. In addition, the trained model can accurately identify and analyze power grid fault characteristics, improve the efficiency and accuracy of power grid fault diagnosis, and provide strong protection for the safe and stable operation of the power grid. This method is applied to terminal equipment as an example for illustration. Figure 1 A schematic diagram of a power grid fault diagnosis method is shown.

[0047] The embodiment of the present application provides an application scenario of the power grid fault diagnosis method, and the application scenario includes the terminal device provided in the embodiment, and the terminal device includes but is not limited to a smart phone and a computer device, wherein the computer device can be at least one of a desktop computer, a portable computer, a laptop computer, a mainframe computer, a tablet computer, etc. The user operates the terminal device to obtain the power grid fault diagnosis result, including but not limited to the power grid fault type and fault location. For the specific process, please refer to the embodiment of the power grid fault diagnosis method.

[0048] The power grid fault diagnosis method implemented in this embodiment is applied to a power grid fault diagnosis system, which includes a power grid node sensor, an edge computing node and a cloud platform, wherein the power grid node sensor is communicatively connected to the edge computing node, and the edge computing node is communicatively connected to the cloud platform.

[0049] Among them, grid node sensors are deployed at each key grid node, including but not limited to high-precision current sensors, voltage sensors, temperature sensors, humidity sensors, vibration sensors and GPS modules. Grid node sensors are connected to edge computing nodes by wired or wireless means to collect real-time grid operation status and location data of data collection points; in addition, in order to record the specific grid operation parameters of each key grid node, grid node sensors can also upload specific grid operation parameters to the cloud platform in real time through wireless communication protocols such as 4G / 5G, LoRa, and NB-IoT for centralized processing and analysis.

[0050] Edge computing nodes migrate computing resources and services from centralized cloud platforms to the edge of the network, closer to the data source or end user. They are devices with certain computing and storage capabilities. Edge computing nodes are used to process and analyze data close to the source of data generation, thereby reducing data transmission delays and alleviating bandwidth pressure on the core network, which can greatly improve the efficiency and security of data processing.

[0051] Based on the above-mentioned power grid fault diagnosis system, the power grid fault diagnosis method provided in this embodiment includes the following specific steps:

[0052] Step S101, obtaining grid operation parameters of key grid nodes through grid node sensors.

[0053] Specifically, the key nodes of the power grid include but are not limited to power grid substations, distribution transformers, transmission line connection points, cable joints and switchgear;

[0054] The grid operation parameters include at least one of current parameters, voltage parameters, power factor, active power, reactive power, frequency, harmonic content, phase angle, temperature, humidity and vibration parameters of key nodes of the grid.

[0055] It should be noted that in order to ensure the reliability of the acquired power grid operation parameters, in this embodiment, the power grid node sensors continuously acquire the power grid operation parameters at a set sampling frequency, and the set sampling frequency is relatively high, so that the obtained power grid operation parameters can be presented in the form of continuous data.

[0056] Furthermore, the grid node sensor of this embodiment is equipped with a processor, which can automatically adjust the sampling frequency according to the real-time data fluctuation of the collected grid operation parameters, so as to ensure the high accuracy and low redundancy requirements of the collected grid operation parameters. The specific expression for automatically adjusting the sampling frequency according to the real-time data fluctuation of the collected grid operation parameters is:

[0057]

[0058] in, is the sampling time interval of the power grid operation parameters, is the basic sampling frequency, To adjust the parameters, It is the standard deviation of the signal in the most recent sampling time interval, reflecting the fluctuation of the power grid.

[0059] Step S102: Send the power grid operation parameters to the edge computing node to determine whether there are any abnormalities in the power grid operation parameters.

[0060] Specifically, in this embodiment, the grid node sensor has a built-in communication module, and the MQTT protocol can be used to transmit the data collected by the grid node sensor to the edge computing node or cloud platform in real time, thereby ensuring the timeliness and integrity of the grid operation parameters.

[0061] Furthermore, the edge computing node determines whether there are any abnormalities in the power grid operation parameters, including:

[0062] Step S301, using sliding window technology to process the power grid operation parameters in sections to obtain real-time power grid operation parameters.

[0063] Specifically, the sliding window technology is used to divide the continuous power grid operation parameter data stream into multiple overlapping or non-overlapping segments, and each segment of data is processed independently. The size and step size of the sliding window can be adjusted according to the characteristics and processing requirements of the power grid operation data. The sliding window technology is implemented on the edge computing node to reduce data transmission delays and improve real-time performance.

[0064] Step S302: input the real-time grid operation parameters into a fast Fourier transform (FFT) model for spectrum analysis to obtain grid operation characteristics.

[0065] Specifically, the grid operation characteristics include key features in the spectrum of grid operation parameters, such as amplitude and phase, for subsequent analysis. In addition, features useful for fault diagnosis are selected from the spectrum features obtained by FFT and optimized. Feature selection can be based on historical data analysis and domain knowledge, and optimization can include feature scaling, dimensionality reduction, etc.

[0066] Step S303: input the power grid operation characteristics into the trained isolation forest model for anomaly detection to obtain anomaly marks of power grid operation parameters.

[0067] The training process of the isolation forest model in step S303 is as follows:

[0068] Initialize the parameters of the isolation forest model, including but not limited to the number of trees and the maximum height of the trees;

[0069] The isolation forest model is trained using the historical power grid operation characteristics until the difference between the training prediction results and the historical abnormal results of the historical power grid operation characteristics is less than the set training threshold, and the trained isolation forest model is output.

[0070] It should be noted that, before the edge computing node of the present embodiment determines whether there is an abnormality in the power grid operation parameters, it is also configured to: perform denoising and normalization processing on the power grid operation parameters.

[0071] The interquartile range (IQR) method is used to identify and remove noise and outliers. The grid operation parameters are scaled to interval, determine the minimum and maximum values ​​in the data set, traverse the data set of power grid operation parameters, apply the normalization formula to calculate the normalized value of each power grid operation parameter data point, and update the power grid operation parameters to the normalized power grid operation parameters.

[0072] Step S103: If there is an abnormality in the power grid operation parameter, the power grid operation parameter is marked as abnormal and then sent to the cloud platform.

[0073] Specifically, real-time processing of power grid operation parameters, feature extraction, and anomaly detection are implemented on edge computing nodes to reduce data transmission delays and improve the real-time performance of power grid fault diagnosis.

[0074] It should be noted that when an abnormality is identified in the grid operation parameters in step S103, after the grid operation parameters are marked as abnormal, an alarm signal can also be generated and sent to the cloud platform, thereby immediately activating the cloud platform to achieve a rapid response to the marked abnormal grid operation parameters.

[0075] In step S104, the cloud platform classifies the abnormally marked power grid operation parameters in combination with the historical power grid fault data, and locates the abnormally marked power grid operation parameters in combination with GIS and GPS to obtain the type and location of the power grid fault.

[0076] Specifically, the cloud platform classifies the abnormally marked power grid operation parameters in combination with the historical power grid fault data, specifically:

[0077] In step S401, the cloud platform trains a convolutional neural network model in combination with historical power grid fault data until the loss of the trained convolutional neural network model converges and outputs the trained convolutional neural network model.

[0078] Step S402: Input the abnormally marked power grid operation parameters into a convolutional neural network model to obtain the power grid fault type.

[0079] It should be noted that the convolutional neural network model of this embodiment is carried on the TensorFlow or PyTorch architecture. The convolutional neural network model (CNN) includes a conventional convolution layer, a pooling layer, and a fully connected layer. After considering the characteristics of the input data, the network depth, the convolution kernel size, and the pooling method, training is performed. The specific steps are:

[0080] a. Initialize network parameters, convolution kernel, bias, and fully connected layer weights;

[0081] b. Use the labeled data as the training set and calculate the output and loss through forward propagation;

[0082] c. Update network parameters through back propagation algorithm to minimize the loss function;

[0083] d. Repeat steps b and c until the predetermined number of training rounds is reached or the loss converges. During the training process, the cross entropy loss function and learning rate hyperparameters need to be selected;

[0084] e. Use the validation set to evaluate the performance of the trained CNN model, such as accuracy, recall, and F1 score, and select the model with the best performance for subsequent applications based on the evaluation results.

[0085] In addition, since the grid operation parameters collected by grid node sensors contain specific location information, the geographic information system (GIS) and the global positioning system (GPS) are combined on this basis to achieve accurate coordinate positioning of the fault location, and map data is updated regularly to ensure the accuracy and timeliness of positioning information. On this basis, through Kalman filtering or multi-source information fusion algorithm, multiple data sources are integrated to improve positioning accuracy, and data from different sensors and devices are integrated to optimize the accuracy of fault positioning; specifically: applying Kalman filtering algorithm to fuse multi-source sensor data to improve positioning accuracy; preprocessing and standardizing data from different sensors and devices, and building a multi-source information fusion framework through Kalman filtering algorithm to integrate data from different sources to improve the reliability of positioning results; implementing real-time calibration mechanism to ensure that the fused data is consistent with the actual situation and improve positioning accuracy.

[0086] Furthermore, in order to more fully diagnose the power grid fault, the power grid fault diagnosis method of this embodiment further includes:

[0087] Step S105: The cloud platform obtains a solution to the power grid failure by combining historical power grid failure data, and sends the solution to the power grid failure.

[0088] Specifically, in step S105, the cloud platform obtains a solution to the power grid failure by combining historical power grid failure data, including:

[0089] Step S501: The cloud platform performs structured processing on historical power grid fault data to obtain historical fault resolution feature information.

[0090] Specifically, the collected historical power grid fault data is structured and key information including but not limited to fault type, fault characteristics, diagnostic steps and solutions is extracted to form a standardized case template, which is helpful for subsequent case retrieval and matching.

[0091] Step S502: The cloud platform extracts fault characteristic information of the power grid operation parameters marked as abnormal.

[0092] Specifically, the fault feature information includes but is not limited to the fault type, fault features, and related parameters, ensuring that the extracted feature information is comprehensive and accurate.

[0093] Step S503: Calculate the similarity between the fault feature information and any historical fault solution feature information until the similarity between the fault feature information and all historical fault solution feature information is obtained.

[0094] Specifically, the expression of the similarity is:

[0095]

[0096] in, is the similarity between the fault feature information and any historical fault solution feature information, The feature vector representing the fault characteristic information, A feature vector representing the characteristic information of historical fault resolution.

[0097] Step S504 , selecting historical power grid fault data corresponding to historical fault solution feature information with a similarity greater than a first threshold as a reference case.

[0098] It should be noted that the first threshold selection for similarity can be set according to the matching requirements between the real-time power grid fault characteristics and the historical fault resolution characteristics according to specific circumstances.

[0099] Step S505 : extracting the historical fault solution of the reference case as a solution to the power grid fault.

[0100] In addition, in addition to the above method, the solution acquisition method for power grid faults in this embodiment can also realize complex fault diagnosis reasoning by combining a rule engine, design and implement a rule-based reasoning engine, and integrate expert experience rules for power grid fault diagnosis. The rule engine should support dynamic addition, modification and deletion of rules to adapt to changes in power grid fault diagnosis requirements, specifically:

[0101] The current fault characteristics are input into the rule engine; the rule engine matches and infers according to the preset rules to determine the fault type and cause; if the rule engine can directly draw a diagnostic conclusion, it outputs the diagnostic result.

[0102] Furthermore, when the above two methods cannot find completely matching historical data or cannot directly diagnose a conclusion, the two methods can be used in combination for comprehensive analysis and judgment to obtain a more accurate diagnosis conclusion.

[0103] In addition, power grid failures usually have traces to follow, and the possible failure conditions of different time lengths can be predicted based on the real-time collected power grid operation parameters, and the operation and maintenance personnel can be reminded to avoid the occurrence of power grid failures. In this regard, the power grid fault diagnosis method of this embodiment also includes:

[0104] Step S201: The cloud platform receives power grid operation parameters sent by power grid node sensors.

[0105] In step S202, the cloud platform inputs the power grid operation parameters into the trained LSTM model to obtain a power grid fault prediction, and sends the power grid fault prediction.

[0106] Among them, the LSTM model consists of multiple LSTM units. Based on the Keras or TensorFlow architecture, the LSTM model processes sequence data through time steps. The LSTM model can capture long-term dependencies in time series data and is suitable for power grid fault prediction. The LSTM model training and prediction steps are as follows:

[0107] f. Initialize LSTM network parameters, such as weights and biases;

[0108] g. Use historical data as a training set and calculate output and loss through forward propagation;

[0109] h. Update network parameters through back propagation algorithm and time back propagation algorithm (BPTT);

[0110] i. Repeat steps g and h until the predetermined number of training rounds is reached or the loss converges.

[0111] The historical data includes historical power grid operation parameters, which are sorted into sequences in chronological order as the input of the LSTM model. The input data includes fault type, occurrence time, duration and related power grid operation parameters.

[0112] The trained LSTM model is used to predict power grid failures in the future. The prediction results provide decision support for power grid operation and maintenance personnel, such as arranging maintenance plans and adjusting operation strategies in advance. At the same time, the prediction results are applied to subsequent anomaly detection and potential fault identification. The new cloud platform stores and analyzes historical fault data, provides fault warning information, and enables them to prepare for fault judgment and analysis in advance.

[0113] In addition, considering the confidentiality of data transmission between the edge computing stage and the cloud platform, the power grid fault diagnosis method of this embodiment also includes:

[0114] The AES encryption algorithm combined with the TLS protocol is used to encrypt the abnormally marked power grid operation parameters. Specifically, the AES-256 encryption algorithm is used to encrypt the power grid operation parameters; the TLS protocol is used during the transmission of power grid operation parameters to encrypt the communication content and prevent data leakage; and strict access control policies are implemented to limit unauthorized access and protect data privacy.

[0115] In addition, in this embodiment, the cloud platform uses a distributed database to store massive power grid operation data, specifically: select MongoDB or Cassandra NoSQL database to support large-scale concurrent reading and writing and horizontal expansion; design reasonable data partitioning and sharding strategies to improve data query and storage efficiency; establish efficient indexes to accelerate data retrieval speed and improve user experience.

[0116] In addition, the power grid fault diagnosis method of this embodiment further includes:

[0117] Generate a formatted fault diagnosis report and display it visually based on the power grid fault type, power grid fault location and power grid fault solution.

[0118] The specific generation process of the fault diagnosis report includes: designing a configurable fault diagnosis report template to facilitate rapid report generation; integrating the output results of the fault diagnosis module, including but not limited to the type of power grid fault, power grid fault location and power grid fault solution, and filling the output results into the report template to generate a formatted fault diagnosis report. Furthermore, an automatic report sending mechanism is established to send the generated report to relevant personnel in a timely manner to facilitate rapid response.

[0119] An interactive monitoring platform is constructed using WebGIS technology and data visualization library to provide a graphical display of real-time data and fault information, facilitating rapid response by operation and maintenance personnel. The visualization specifically includes: constructing an interactive monitoring platform using WebGIS technology to provide a graphical display of real-time data and fault information; selecting D3.js or ECharts data visualization library to realize dynamic display and interactive functions of data; designing an intuitive and easy-to-use monitoring interface to facilitate operation and maintenance personnel to quickly grasp the operating status and fault conditions of the power grid.

[0120] In addition, to facilitate data query, this embodiment also includes the construction of a fault case database. The specific implementation process is: collecting historical fault cases, organizing and classifying them, and building a fault case library; applying case-based reasoning (CBR) technology to quickly retrieve similar cases and assist in fault diagnosis; establishing a case update mechanism, regularly adding new cases, and maintaining the timeliness and accuracy of the knowledge base.

[0121] Combined with the rule engine to realize the diagnostic reasoning of complex faults, the expert experience rules are integrated, fault diagnosis and decision support are carried out through the reasoning engine, and the expert system framework is constructed. The specific implementation process is: constructing the rule engine, integrating the expert experience rules, and realizing the diagnostic reasoning of complex faults; optimizing the reasoning algorithm, improving the reasoning speed and accuracy, and meeting the needs of real-time fault diagnosis; constructing the expert system framework, integrating the rule engine and the knowledge base, and providing comprehensive fault diagnosis and decision support.

[0122] In addition, this embodiment also includes a health monitoring and performance optimization resource scheduling part for the power grid, wherein the health monitoring part of the power grid monitors the operating status of the system by adopting heartbeat detection and log analysis technology, promptly discovers and handles system abnormalities, and ensures stable operation of the system; the performance optimization resource scheduling part uses containerization technology and cloud resource management technology to achieve dynamic scheduling and optimization of resources, automatically adjust resource configuration according to system load, and optimize system performance.

[0123] The implementation process of the health monitoring part of the power grid includes: implementing a heartbeat detection mechanism to regularly detect the operating status of system components and promptly detect anomalies; using log analysis tools to perform real-time analysis of system logs to identify potential problems; establishing an exception handling process and immediately initiating an emergency response mechanism once a problem is discovered to ensure stable system operation.

[0124] The implementation process of the performance optimization resource scheduling part includes: using Docker containerization technology to achieve rapid deployment and migration of applications; using Kubernetes cloud resource management tools to achieve dynamic scheduling and optimization of resources; designing a reasonable load balancing strategy to automatically adjust resource configuration according to system load and optimize system performance.

[0125] In this embodiment, the trained CNN and LSTM models are deployed on the cloud platform to provide fault diagnosis and prediction services. The cloud platform can provide elastic computing resources, data storage and access interface support to facilitate the application and expansion of the model.

[0126] This embodiment uses the high-bandwidth, low-latency communication capabilities of the cloud platform to collect, transmit and process power grid data in real time and respond to faults quickly. Combined with the algorithm model, it can accurately identify and analyze power grid fault characteristics and improve the accuracy of diagnosis. By integrating expert experience rules and reasoning engines, intelligent fault diagnosis and decision support are achieved, providing scientific and reasonable processing suggestions for operation and maintenance personnel. At the same time, the cloud platform supports dynamic expansion and high-availability deployment of the system, ensuring that the system can be smoothly upgraded as the scale of the power grid grows and maintain stable operation. It also uses encrypted storage and transmission technology to ensure the security and privacy of power grid data.

[0127] It should be understood that although the Figure 1 The steps in the flowchart are shown in sequence according to the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least part of the steps in the drawings may include multiple sub-steps or sub-stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0128] The above embodiments of the present invention describe in detail the power grid fault diagnosis method. The above methods disclosed in the present invention can be implemented by various devices. Therefore, the present invention also discloses a power grid fault diagnosis device. Figure 2 , specific embodiments are given below to explain in detail.

[0129] The parameter acquisition module 601 is used to acquire the grid operation parameters of the key grid nodes through the grid node sensors;

[0130] An abnormality judgment module 602 is used to send the power grid operation parameters to the edge computing node to determine whether there are abnormalities in the power grid operation parameters;

[0131] The abnormality uploading module 603 is used to mark the abnormality of the power grid operation parameter and send it to the cloud platform if there is an abnormality in the power grid operation parameter;

[0132] The abnormality analysis module 604 is used for the cloud platform to classify the abnormally marked power grid operation parameters in combination with the historical power grid fault data, and to locate the abnormally marked power grid operation parameters in combination with GIS and GPS to obtain the type and location of the power grid fault.

[0133] For the power grid fault diagnosis device, all the above-mentioned methods can be referred to, and no further details will be given here. Each module in the above-mentioned device can be implemented in whole or in part by software, hardware, and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor of the terminal device in the form of hardware, or can be stored in the memory of the terminal device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0134] In one embodiment, the present invention further provides a computer-readable storage medium having a computer program stored thereon, and the computer program implements the steps of the above-mentioned power grid fault diagnosis method when executed by a processor.

[0135] The computer-readable storage medium may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read-only memory), an EPROM (erasable programmable read-only memory), a hard disk or a ROM. Optionally, the computer-readable storage medium includes a non-transitory computer-readable storage medium. The computer-readable storage medium has a storage space for program codes for executing any method step of the above method. These program codes may be read from or written into one or more computer program products, and the program codes may be compressed in an appropriate form.

[0136] In one embodiment, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the above-mentioned power grid fault diagnosis method when executing the computer program.

[0137] The computer device includes a memory, a processor, and one or more computer programs, wherein the one or more computer programs can be stored in the memory and configured to be executed by one or more processors, and the one or more application programs are configured to execute the above-mentioned power grid fault diagnosis method.

[0138] The processor may include one or more processing cores. The processor uses various interfaces and lines to connect the various parts of the entire computer device, and executes various functions of the computer device and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory, and calling data stored in the memory. Optionally, the processor can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor can integrate one or more combinations of a central processing unit (Central Processing Unit, CPU), a reporting verifier (Graphics Processing Unit, GPU) for buried data, and a modem. Among them, the CPU mainly processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing display content; and the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor, but may be implemented separately through a communication chip.

[0139] The memory may include a random access memory (RAM) or a read-only memory (ROM). The memory may be used to store instructions, programs, codes, code sets or instruction sets. The memory may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc. The data storage area may also store data created by the terminal device during use, etc.

[0140] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A power grid fault diagnosis method, the power grid fault diagnosis method is applied to a power grid fault diagnosis system, the power grid fault diagnosis system includes a power grid node sensor, an edge computing node and a cloud platform, the power grid node sensor is communicatively connected to the edge computing node, the edge computing node is communicatively connected to the cloud platform, characterized in that: The power grid fault diagnosis method comprises: Obtain grid operation parameters of key grid nodes through grid node sensors; Sending the power grid operation parameters to the edge computing node to determine whether there are any abnormalities in the power grid operation parameters; If the power grid operation parameters are abnormal, the power grid operation parameters are marked as abnormal and sent to the cloud platform; The cloud platform classifies the abnormally marked power grid operation parameters in combination with historical power grid fault data, and locates the abnormally marked power grid operation parameters in combination with GIS and GPS to obtain the type and location of the power grid fault.

2. The power grid fault diagnosis method according to claim 1, characterized in that: Also includes: The cloud platform obtains a solution to the power grid failure in combination with historical power grid failure data, and sends the solution to the power grid failure.

3. The power grid fault diagnosis method according to claim 2, characterized in that: The power grid node sensor is also connected to the cloud platform for communication, and the power grid fault diagnosis method further includes: The cloud platform receives the grid operation parameters sent by grid node sensors; The cloud platform inputs the power grid operation parameters into the trained LSTM model to obtain a power grid fault prediction, and sends the power grid fault prediction.

4. The power grid fault diagnosis method according to claim 3, characterized in that: The key nodes of the power grid include power grid substations, distribution transformers, transmission line connection points, cable joints and switchgear; The grid operation parameters include at least one of current parameters, voltage parameters, power factor, active power, reactive power, frequency, harmonic content, phase angle, temperature, humidity and vibration parameters of key nodes of the grid.

5. The power grid fault diagnosis method according to claim 4, characterized in that: The edge computing node determines whether there are any abnormalities in the grid operation parameters, including: The sliding window technology is used to process the power grid operation parameters in sections to obtain real-time power grid operation parameters; Inputting the real-time power grid operation parameters into a fast Fourier transform model for spectrum analysis to obtain power grid operation characteristics; The power grid operation characteristics are input into the trained isolation forest model for anomaly detection to obtain abnormal markers of power grid operation parameters.

6. The power grid fault diagnosis method according to claim 5, characterized in that: The cloud platform classifies the abnormally marked power grid operation parameters in combination with the historical power grid fault data, specifically: The cloud platform trains the convolutional neural network model in combination with historical power grid fault data until the loss of the trained convolutional neural network model converges and outputs the trained convolutional neural network model; The abnormally marked power grid operation parameters are input into the convolutional neural network model to obtain the power grid fault type.

7. The power grid fault diagnosis method according to claim 6, characterized in that: The cloud platform combines historical grid fault data to obtain solutions to grid faults, including: The cloud platform performs structured processing on historical power grid fault data to obtain historical fault resolution feature information; The cloud platform extracts fault feature information of abnormal power grid operation parameters; Calculating the similarity between the fault feature information and any historical fault solution feature information until the similarity between the fault feature information and all historical fault solution feature information is obtained; Selecting historical power grid fault data corresponding to historical fault resolution feature information with a similarity greater than a first threshold as a reference case; A historical fault solution of the reference case is extracted as a solution to the power grid fault.

8. A power grid fault diagnosis device, the power grid fault diagnosis device is applied to a power grid fault diagnosis system, the power grid fault diagnosis system comprises a power grid node sensor, an edge computing node and a cloud platform, the power grid node sensor is communicatively connected to the edge computing node, the edge computing node is communicatively connected to the cloud platform, characterized in that: The power grid fault diagnosis device comprises: A parameter acquisition module is used to acquire grid operation parameters of key grid nodes through grid node sensors; An abnormality judgment module, used to send the power grid operation parameters to the edge computing node to determine whether there are abnormalities in the power grid operation parameters; An abnormality uploading module is used to mark the abnormality of the power grid operation parameters and send them to the cloud platform if there are abnormalities in the power grid operation parameters; The abnormality analysis module is used for the cloud platform to classify the abnormal power grid operation parameters combined with historical power grid fault data, and to locate the abnormal power grid operation parameters combined with GIS and GPS to obtain the type and location of the power grid fault.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the power grid fault diagnosis method according to any one of claims 1 to 7 are implemented.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it performs the power grid fault diagnosis method according to any one of claims 1 to 7.

Images