GIS equipment operation fault analysis method and system based on three-dimensional modeling
By cleaning and three-dimensional modeling of GIS equipment data, combined with deep neural network analysis, the traditional fault detection method has solved the shortcomings of complex fault identification and prediction, and achieved efficient and accurate fault detection and analysis to meet the needs of modern power systems. need.
Patent Information
- Application Number
- CN202410959992.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-17
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-07-17
AI Technical Summary
Traditional GIS equipment fault detection methods have low accuracy in detecting complex faults, cannot monitor and respond quickly in real time, and cannot effectively use operating data for fault trend analysis and prediction, resulting in low information utilization and difficult to meet the efficiency of modern power systems. Accurate fault detection requirements.
Using a three-dimensional modeling method, data cleaning, denoising and missing value filling of GIS equipment operating status data, environment data and historical fault data is used to build a high-precision three-dimensional model of GIS equipment, and fault characteristics are combined with deep neural networks. Extraction and analysis to achieve efficient identification and prediction of complex failures.
It significantly improves the real-time, accuracy and comprehensiveness of fault detection, can accurately identify and predict complex faults of GIS equipment, improves the efficiency and accuracy of fault analysis, and meets the needs of modern power systems for efficient fault detection.
Smart Images

Figure CN118918254A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for analyzing GIS equipment operation failure based on three-dimensional modeling. Background Art
[0002] Gas-insulated switchgear (GIS) equipment is widely used in power systems, and its operating stability and reliability are crucial to the safety of power systems. At present, the fault detection and analysis methods of GIS equipment mainly include infrared imaging, ultrasonic detection, and partial discharge detection. These methods rely on manual experience and traditional instruments to monitor the operating status of equipment and diagnose faults. To a certain extent, they can detect equipment faults and provide maintenance references.
[0003] Disadvantages: Traditional fault detection methods have certain limitations. First, these methods are not accurate in detecting complex faults, and it is difficult to fully capture the characteristics of various faults generated during the operation of GIS equipment. Secondly, the existing technology performs poorly in terms of real-time performance and cannot monitor faults in real time and respond quickly. In addition, these methods cannot effectively use a large amount of operating data for fault trend analysis and prediction, resulting in low information utilization, which makes it difficult to meet the modern power system's demand for efficient and accurate fault detection. Summary of the invention
[0004] In view of the above problems existing in the prior art, the present invention is proposed.
[0005] Therefore, the existing technical problem to be solved by the present invention is that the detection accuracy of complex faults is not high, and it is difficult to fully capture the various fault characteristics generated during the operation of GIS equipment.
[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a GIS equipment operation fault analysis method based on three-dimensional modeling, which includes performing data cleaning processing on the collected GIS equipment operation status data, environmental data and historical fault data to obtain a cleaned data set;
[0007] Performing data denoising on the cleaned data set to obtain a denoised data set;
[0008] Filling missing values in the denoised data set by a variational autoencoder to obtain a filled data set;
[0009] Perform three-dimensional modeling according to the filled data set to obtain a three-dimensional model of the GIS equipment;
[0010] Collecting real-time operation data of the GIS equipment, and extracting fault features of the three-dimensional model of the GIS equipment through the real-time operation data to obtain a fault feature vector;
[0011] The fault feature vector is input into a deep neural network model to perform fault analysis to obtain fault analysis data.
[0012] As a preferred solution of the GIS equipment operation fault analysis method based on three-dimensional modeling described in the present invention, the step of performing data cleaning on the collected GIS equipment operation status data, environmental data and historical fault data to obtain a cleaned data set includes:
[0013] Perform outlier detection and processing on the GIS equipment operation status data to obtain the processed operation status data;
[0014] Performing data fusion according to the processed operating status data and environmental data to obtain fused data;
[0015] Performing feature engineering on the fused data to extract key feature vectors to obtain a feature vector data set;
[0016] Constructing a label data set according to the feature vector data set and the historical fault data;
[0017] Performing data segmentation on the labeled data set to obtain a training data set and a verification data set;
[0018] Training a classification model based on the training data set to obtain a trained classification model;
[0019] Using the validation data set to evaluate the trained classification model to obtain an evaluation result;
[0020] Optimizing the trained classification model according to the evaluation result to obtain an optimized classification model;
[0021] Applying the optimized classification model to the feature vector data set to classify the data to obtain classified data;
[0022] The classified data is cleaned to obtain a cleaned data set.
[0023] As a preferred solution of the GIS equipment operation fault analysis method based on three-dimensional modeling described in the present invention, the step of denoising the cleaned data set to obtain the denoised data set includes:
[0024] Constructing a Gaussian mixture model according to the cleaned data set to obtain Gaussian mixture model parameters;
[0025] Performing anomaly detection on each data point in the cleaned data set, calculating its anomaly score according to the Gaussian mixture model parameters, and obtaining an anomaly score set;
[0026] An anomaly threshold is set according to the anomaly score set, and data points with an anomaly score higher than the anomaly threshold are marked as anomalies, so as to obtain a marked data set;
[0027] Performing principal component analysis on the labeled data set, extracting a principal component loading matrix, and obtaining a principal component loading matrix;
[0028] Performing a projection transformation on the labeled data set according to the principal component loading matrix to obtain a projected data set;
[0029] Performing a kernel discriminant analysis on the projected data set, constructing a discriminant model, and obtaining a discriminant model;
[0030] Using the discriminant model to discriminate abnormal points in the projection data set to obtain abnormal point discrimination results;
[0031] According to the outlier identification result, data identified as outliers are removed from the projection data set to obtain a denoised projection data set;
[0032] Performing a back-projection transformation on the denoised projection data set to obtain a back-projection data set;
[0033] The back-projection data set is converted into a data format to obtain the denoised data set.
[0034] As a preferred solution of the GIS equipment operation fault analysis method based on three-dimensional modeling described in the present invention, the step of filling missing values of the denoised data set by a variational autoencoder to obtain a filled data set includes:
[0035] Performing standardization processing on the denoised data set to obtain a standardized data set;
[0036] Constructing a variational autoencoder model according to the standardized data set, including an encoder and a decoder network, to obtain a variational autoencoder model;
[0037] Applying an encoder network in the variational autoencoder model to the standardized dataset to obtain a latent space distribution;
[0038] Sampling latent variables from the latent space distribution to obtain a set of latent variables;
[0039] Inputting the latent variable set into the decoder network of the variational autoencoder model to obtain a reconstructed data set;
[0040] Calculate the reconstruction loss based on the standardized data set and the reconstructed data set;
[0041] Calculate the KL divergence loss according to the latent space distribution;
[0042] Add the reconstruction loss and the KL divergence loss to get the total loss;
[0043] Based on the total loss, the variational autoencoder model is updated with parameters to obtain an updated variational autoencoder model;
[0044] The updated variational autoencoder model is used to fill missing values in the denoised data set to obtain the filled data set.
[0045] As a preferred solution of the GIS equipment operation fault analysis method based on three-dimensional modeling described in the present invention, the step of performing three-dimensional modeling according to the filled data set to obtain a three-dimensional model of the GIS equipment includes:
[0046] Performing feature extraction on the filled data set to obtain a feature data set;
[0047] Constructing three-dimensional point cloud data according to the feature data set to obtain original point cloud data;
[0048] Performing point cloud filtering on the original point cloud data to remove outliers and obtain filtered point cloud data;
[0049] Performing point cloud segmentation on the filtered point cloud data, extracting key point cloud clusters, and obtaining segmented point cloud data;
[0050] Performing three-dimensional reconstruction on the segmented point cloud data, constructing a three-dimensional mesh model, and obtaining a rough three-dimensional model;
[0051] Performing mesh optimization on the rough three-dimensional model, including vertex refinement and facet subdivision, to obtain an optimized three-dimensional model;
[0052] Performing texture mapping on the optimized three-dimensional model, giving the model a real texture according to the feature data set, and obtaining a textured three-dimensional model;
[0053] Performing lighting settings on the textured three-dimensional model, including setting ambient light and point light sources, to obtain an illuminated three-dimensional model;
[0054] Generate a rendering image according to the lighting three-dimensional model to obtain a three-dimensional rendering image of the GIS device;
[0055] The three-dimensional rendering image of the GIS device is combined with the feature data set to obtain the three-dimensional model of the GIS device.
[0056] As a preferred solution of the GIS equipment operation fault analysis method based on three-dimensional modeling described in the present invention, the step of collecting real-time operation data of the GIS equipment and extracting fault features of the three-dimensional model of the GIS equipment through the real-time operation data to obtain the fault feature vector includes:
[0057] Preprocessing the real-time operation data, including data standardization and feature scaling, to obtain processed real-time data;
[0058] Mapping the processed real-time data onto the three-dimensional model of the GIS device to obtain a three-dimensional model with data;
[0059] Performing model segmentation on the three-dimensional model with data to obtain a segmented model data set;
[0060] Construct a model graph according to the segmented model data set to obtain model graph data;
[0061] Performing a graph convolution operation on the model graph data to extract graph features and obtain graph feature data;
[0062] Inputting the graph feature data into a graph attention network to obtain attention-weighted graph features;
[0063] Performing a pooling operation on the attention weighted graph features to obtain a pooled feature tensor;
[0064] Inputting the pooled feature tensor into a fully connected layer to perform feature transformation to obtain a potential fault feature vector;
[0065] Normalizing the potential fault feature vector to obtain a normalized fault feature vector;
[0066] The normalized fault feature vector is input into the classification layer to obtain the fault feature vector.
[0067] As a preferred solution of the GIS equipment operation failure analysis method based on three-dimensional modeling described in the present invention, wherein:
[0068] The step of inputting the fault feature vector into a deep neural network model for fault analysis to obtain fault analysis data comprises:
[0069] Constructing an input tensor according to the fault feature vector to obtain input tensor data;
[0070] Inputting the input tensor data into a convolutional neural network model to obtain a convolutional feature map;
[0071] Performing a maximum pooling operation on the convolutional feature map to obtain a pooled feature map;
[0072] Flattening the pooled feature map into a one-dimensional vector to obtain a one-dimensional feature vector;
[0073] Inputting the one-dimensional feature vector into a long short-term memory network to obtain time series feature data;
[0074] Input the time series feature data into the attention layer, perform feature weighting, and obtain weighted feature data;
[0075] Inputting the weighted feature data into a fully connected layer and performing nonlinear transformation to obtain a fault probability distribution;
[0076] Perform classification according to the fault probability distribution to obtain a classification result;
[0077] Mapping the classification result into a fault code to obtain fault code data;
[0078] The fault code data is combined with corresponding fault description information to obtain the fault analysis data.
[0079] A GIS equipment operation fault analysis system based on three-dimensional modeling, including:
[0080] The acquisition module is used to perform data cleaning on the collected GIS equipment operating status data, environmental data and historical fault data to obtain a cleaned data set;
[0081] A denoising module, used for denoising the cleaned data set to obtain a denoised data set;
[0082] A filling module, used for filling missing values of the denoised data set by using a variational autoencoder to obtain a filled data set;
[0083] A modeling module, used for performing three-dimensional modeling according to the filled data set to obtain a three-dimensional model of the GIS equipment;
[0084] A collection module is used to collect real-time operation data of GIS equipment, and extract fault features of the three-dimensional model of the GIS equipment through the real-time operation data to obtain a fault feature vector;
[0085] The analysis module is used to input the fault feature vector into a deep neural network model to perform fault analysis and obtain fault analysis data.
[0086] A preferred solution of a GIS equipment operation fault analysis system based on three-dimensional modeling, wherein: the acquisition module further includes:
[0087] It is used to clean the collected GIS equipment operation status data, environmental data and historical fault data to obtain a clean data set;
[0088] Used to perform data denoising on the cleaned data set to obtain a denoised data set;
[0089] Used to fill missing values of the denoised data set through a variational autoencoder to obtain a filled data set;
[0090] Used to perform three-dimensional modeling according to the filled data set to obtain a three-dimensional model of the GIS equipment;
[0091] Used to collect real-time operation data of GIS equipment, and extract fault features of the three-dimensional model of the GIS equipment through the real-time operation data to obtain a fault feature vector;
[0092] It is used to input the fault feature vector into a deep neural network model to perform fault analysis and obtain fault analysis data.
[0093] The GIS equipment operation fault analysis system based on three-dimensional modeling is characterized in that the analysis module further includes:
[0094] Used to construct an input tensor according to the fault feature vector to obtain input tensor data;
[0095] Used to input the input tensor data into a convolutional neural network model to obtain a convolutional feature map;
[0096] Used to perform a maximum pooling operation on the convolution feature map to obtain a pooled feature map;
[0097] Used to flatten the pooled feature map into a one-dimensional vector to obtain a one-dimensional feature vector;
[0098] Used to input the one-dimensional feature vector into a long short-term memory network to obtain time series feature data;
[0099] Used to input the time series feature data into the attention layer, perform feature weighting, and obtain weighted feature data;
[0100] Used to input the weighted feature data into a fully connected layer, perform nonlinear transformation, and obtain a fault probability distribution;
[0101] Used to perform classification according to the fault probability distribution to obtain a classification result;
[0102] Used to map the classification result into a fault code to obtain fault code data;
[0103] Used to combine the fault code data with corresponding fault description information to obtain fault analysis data.
[0104] Beneficial effects of the present invention: In the technical solution provided by the present invention, by performing data cleaning on the collected GIS equipment operating status data, environmental data and historical fault data, the noise and outliers in the data are effectively removed, ensuring the quality and reliability of the data. This step includes sub-steps such as outlier detection and processing, data fusion, feature engineering, label data set construction, data segmentation, classification model training and optimization, and each sub-step is carefully designed to maximize the effect and efficiency of data cleaning. Next, the data is denoised by a Gaussian mixture model to effectively remove outliers in the data, and the purity of the data is further improved by principal component analysis and kernel discriminant analysis to ensure that the data processed subsequently is more accurate and reliable. The multi-level processing method of this step enables the denoised data set to not only retain the main features of the data, but also effectively reduce the interference of noise on the analysis results. Subsequently, the missing values of the denoised data set are filled by a variational autoencoder, further enhancing the integrity and accuracy of the data. The variational autoencoder maps the standardized data set to the latent space through the dual processing of the encoder and decoder networks, and reconstructs the data by sampling latent variables, ultimately achieving the filling of missing values. The comprehensive calculation and optimization of reconstruction loss and KL divergence loss ensure the quality of the filled data set, so that the data still maintains a high degree of consistency and integrity in the filled state. On this basis, three-dimensional modeling is carried out according to the filled data set, and a high-precision three-dimensional model of GIS equipment is constructed through steps such as feature extraction, point cloud data construction, point cloud filtering, point cloud segmentation, three-dimensional reconstruction, mesh optimization, texture mapping and lighting setting. This model not only has real physical properties, but also can intuitively display the structure and status of the equipment through rendered images, providing a solid foundation for fault analysis. The collection and processing of real-time operation data is an important part of fault analysis. By preprocessing the real-time operation data, it is mapped to the three-dimensional model, and model segmentation, graph convolution operation, graph attention network weighting, pooling operation and feature transformation are performed to extract the potential fault feature vector. This process makes full use of the combination of real-time data and three-dimensional models, making the extraction of fault features more accurate and comprehensive. After the fault feature vector is extracted, the fault analysis is further performed through the deep neural network model. The combination of convolutional neural network, long short-term memory network and attention mechanism fully explores the spatiotemporal correlation of fault features and ensures the accuracy and real-time performance of fault analysis. Finally, the fault feature vector undergoes multi-level feature transformation and classification to generate fault probability distribution and fault code data, and combined with the corresponding fault description information, the fault analysis data is obtained. This series of technical features and steps, through the organic combination of deep learning and 3D modeling, realizes the efficient analysis and prediction of GIS equipment operation faults, significantly improves the real-time, accuracy and comprehensiveness of fault detection, and solves the problem that traditional fault detection methods are difficult to accurately identify and predict faults in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS
[0105] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0106] Figure 1 It is a flow chart of a GIS equipment operation failure analysis method based on three-dimensional modeling in an embodiment of the present invention;
[0107] Figure 2 It is a schematic diagram of a GIS equipment operation fault analysis system based on three-dimensional modeling in an embodiment of the present invention. DETAILED DESCRIPTION
[0108] In order to make the above-mentioned objects, features and advantages of the present invention more clearly understood, the specific implementation modes of the present invention are described in detail below in conjunction with the accompanying drawings.
[0109] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.
[0110] Example 1
[0111] Reference Figure 1 , Figure 1 FIG. 1 is a flow chart of a method for analyzing GIS equipment operation failure based on three-dimensional modeling according to an embodiment of the present invention. Figure 1 As shown, the following steps are included:
[0112] S101, performing data cleaning on the collected GIS equipment operation status data, environmental data and historical fault data to obtain a cleaned data set;
[0113] S102, performing data denoising on the cleaned data set to obtain a denoised data set;
[0114] S103, filling missing values in the denoised data set by using a variational autoencoder to obtain a filled data set;
[0115] S104, performing three-dimensional modeling according to the filled data set to obtain a three-dimensional model of the GIS equipment;
[0116] S105, collecting real-time operation data of the GIS equipment, and extracting fault features of the GIS equipment three-dimensional model through the real-time operation data to obtain a fault feature vector;
[0117] S106. Input the fault feature vector into a deep neural network model to perform fault analysis to obtain fault analysis data.
[0118] It should be noted that data cleaning of the collected GIS equipment operating status data, environmental data and historical fault data is the basis of the entire method. Specifically, data cleaning includes outlier detection and processing of operating status data, and ensures the validity and accuracy of the data by identifying and eliminating values that are obviously deviated from the normal range. For example, during the operation of a certain GIS device, if the reading of the temperature sensor suddenly appears to be an extremely high value, this may be a misreading caused by a sensor failure. The outlier detection algorithm can identify this anomaly and eliminate or replace it with a reasonable value. Next, the processed operating status data is fused with the environmental data, and information from multiple data sources is combined to form a more comprehensive data set. After data fusion, feature engineering is performed to extract key feature vectors, which can better reflect the operating status of the GIS equipment. For example, the features that best reflect the health status of the equipment can be extracted from multiple sensor data such as temperature, humidity, current, and voltage. After completing feature engineering, these feature vectors are used with historical fault data to construct a label data set, and each feature vector is matched with the corresponding fault type. Next, the label data set is segmented to obtain a training data set and a validation data set. By training a classification model based on a training data set, such as a random forest or support vector machine model, a fault classification system can be preliminarily established. The classification model is evaluated using a validation data set, and the model is optimized based on the evaluation results to ensure that it has high accuracy and robustness. The optimized classification model is applied to the feature vector data set to classify the data, obtain the classified data, and clean it to form a cleaned data set. After the cleaned data set is obtained, the data is denoised. By constructing a Gaussian mixture model, anomaly detection is performed on each data point, and its anomaly score is calculated. An anomaly threshold is set, and data points with anomaly scores higher than the threshold are marked as anomalies. The principal component load matrix is extracted through principal component analysis, and the marked data set is projected and transformed and kernel discriminant analyzed. Finally, the anomalies are removed to obtain a denoised data set. For example, when analyzing the vibration data of GIS equipment, some data points may be found to be abnormal due to external interference. Through the above process, these anomalies can be effectively identified and removed to ensure the purity of the data. Based on the denoised data set, a variational autoencoder is used to fill missing values. The variational autoencoder maps data to a latent space by building a neural network model that includes an encoder and a decoder, samples latent variables from it, and reconstructs the data through the decoder. By calculating the reconstruction loss and KL divergence loss, the variational autoencoder model is optimized so that it can accurately fill in missing values. For example, some sensor data is missing due to sensor failure. Through the processing of the variational autoencoder, reasonable filling values can be generated to ensure the integrity and consistency of the data.Next, three-dimensional modeling is performed based on the filled data set, and three-dimensional point cloud data is constructed through feature extraction. Point cloud filtering and segmentation are performed to remove outliers and noise, extract key point cloud clusters, and finally complete three-dimensional reconstruction and mesh optimization to obtain a three-dimensional model of GIS equipment with real texture and lighting effects. For example, by performing three-dimensional reconstruction on multiple sensor data of GIS equipment, the internal structure and operating status of the equipment can be intuitively displayed, providing an important reference for fault analysis. The collection and processing of real-time operation data is a key link in fault analysis. The real-time data is standardized and mapped to the three-dimensional model through preprocessing, and model segmentation and graph convolution operations are performed to extract graph feature data. The pooled feature tensor is obtained through weighting and pooling operations of the graph attention network. The pooled feature tensor is input into the fully connected layer for feature transformation to obtain the potential fault feature vector. For the actual application of GIS equipment, for example, when the temperature of a part of the equipment suddenly rises, combined with the temperature distribution map of the three-dimensional model, the fault location can be accurately located and related features can be extracted. Finally, the fault feature vector is input into the deep neural network model for fault analysis. The convolutional neural network extracts feature maps, the long short-term memory network analyzes time series features, the attention mechanism weights features, and the fully connected layer performs nonlinear transformation to generate fault probability distribution, and finally classifies the fault code. The fault code is combined with the description information to generate fault analysis data. For example, by performing deep learning analysis on the fault feature vector of a certain device, the possible fault type of the device, such as arc fault or ground fault, can be accurately identified, and a detailed fault description and recommended treatment measures can be provided. This method not only improves the accuracy and real-time performance of fault detection, but also provides maintenance personnel with intuitive and detailed fault analysis results, greatly improving the maintenance and management efficiency of GIS equipment.
[0119] By executing the above steps, the noise and outliers in the data are effectively removed by cleaning the collected GIS equipment operation status data, environmental data and historical fault data, ensuring the quality and reliability of the data. This step includes sub-steps such as outlier detection and processing, data fusion, feature engineering, label data set construction, data segmentation, classification model training and optimization. Each sub-step has been carefully designed to maximize the effect and efficiency of data cleaning. Next, the data is denoised by using the Gaussian mixture model to effectively remove outliers in the data, and the purity of the data is further improved by principal component analysis and kernel discriminant analysis to ensure that the data processed later is more accurate and reliable. The multi-level processing method of this step not only retains the main features of the data in the denoised data set, but also effectively reduces the interference of noise on the analysis results. Subsequently, the missing values of the denoised data set are filled by the variational autoencoder, which further enhances the integrity and accuracy of the data. The variational autoencoder maps the standardized data set to the latent space through the dual processing of the encoder and decoder networks, and reconstructs the data by sampling latent variables, finally achieving the filling of missing values. The comprehensive calculation and optimization of reconstruction loss and KL divergence loss ensure the quality of the filled data set, so that the data still maintains a high degree of consistency and integrity in the filled state. On this basis, three-dimensional modeling is carried out according to the filled data set, and a high-precision three-dimensional model of GIS equipment is constructed through steps such as feature extraction, point cloud data construction, point cloud filtering, point cloud segmentation, three-dimensional reconstruction, mesh optimization, texture mapping and lighting setting. This model not only has real physical properties, but also can intuitively display the structure and status of the equipment through rendered images, providing a solid foundation for fault analysis. The collection and processing of real-time operation data is an important part of fault analysis. By preprocessing the real-time operation data, it is mapped to the three-dimensional model, and model segmentation, graph convolution operation, graph attention network weighting, pooling operation and feature transformation are performed to extract the potential fault feature vector. This process makes full use of the combination of real-time data and three-dimensional models, making the extraction of fault features more accurate and comprehensive. After the fault feature vector is extracted, the fault analysis is further performed through the deep neural network model. The combination of convolutional neural network, long short-term memory network and attention mechanism fully explores the spatiotemporal correlation of fault features and ensures the accuracy and real-time performance of fault analysis. Finally, the fault feature vector undergoes multi-level feature transformation and classification to generate fault probability distribution and fault code data, and combined with the corresponding fault description information, the fault analysis data is obtained. This series of technical features and steps, through the organic combination of deep learning and 3D modeling, realizes the efficient analysis and prediction of GIS equipment operation faults, significantly improves the real-time, accuracy and comprehensiveness of fault detection, and solves the problem that traditional fault detection methods are difficult to accurately identify and predict faults in complex environments.
[0120] In a specific embodiment, the process of executing step S101 may specifically include the following steps:
[0121] (1) Detect and process outliers in the GIS equipment operating status data to obtain processed operating status data;
[0122] (2) performing data fusion based on the processed operation status data and environmental data to obtain fused data;
[0123] (3) Perform feature engineering on the fused data, extract key feature vectors, and obtain a feature vector data set;
[0124] (4) Construct a label dataset based on the feature vector dataset and historical fault data;
[0125] (5) Segment the labeled data set to obtain a training data set and a validation data set;
[0126] (6) training a classification model based on the training data set to obtain a trained classification model;
[0127] (7) Use the validation data set to evaluate the trained classification model and obtain the evaluation results;
[0128] (8) Optimizing the trained classification model according to the evaluation results to obtain an optimized classification model;
[0129] (9) applying the optimized classification model to the feature vector data set to classify the data and obtain classified data;
[0130] (10) Perform data cleaning on the classified data to obtain a cleaned data set.
[0131] Specifically, first of all, it is necessary to detect and process outliers in the operating status data of GIS equipment. During the operation of GIS equipment, sensors may generate outliers for various reasons, which will affect subsequent data processing and analysis. Therefore, it is necessary to identify these outliers through an outlier detection algorithm. There are many methods for outlier detection, such as the 3σ principle based on statistics, the box plot method, or the isolation forest algorithm based on machine learning. For example, when a current sensor is operating normally, its data fluctuates within a certain range, but at a certain moment the data of the sensor suddenly soars to an extremely high value, which is obviously beyond the normal range. It can be identified as an outlier by the isolation forest algorithm. After identifying the outliers, it can be processed by elimination, interpolation or replacement to obtain the processed operating status data. Next, data fusion is performed based on the processed operating status data and environmental data to form a more comprehensive data set. The operating status data of GIS equipment includes parameters such as voltage, current, and temperature, while the environmental data includes external temperature, humidity, etc. The fusion of these data can more comprehensively reflect the operating status of the equipment. Data fusion can be performed by weighted average, principal component analysis (PCA) or feature fusion. For example, in a high temperature environment in summer, the temperature data of the equipment may be affected by the ambient temperature. The data fusion method can more accurately separate the temperature changes inside the equipment, so as to obtain a more realistic equipment operation status. After data fusion, feature engineering is performed on the fused data to extract key feature vectors. Feature engineering is to transform the original data into features that can better reflect the essence of the problem, thereby improving the performance of the model. Feature extraction can use statistical features, frequency domain features, or time series-based features. For example, statistical features such as mean, standard deviation, and kurtosis can be extracted from current data, frequency components can be extracted from frequency domain data, and lag features can be extracted from time series data. Through feature engineering, the original data is converted into feature vectors to obtain a feature vector data set. Then, a label data set is constructed based on the feature vector data set and historical fault data. The historical fault data records the fault conditions of GIS equipment under different operating conditions and can be paired with feature vectors as label data. When constructing a label data set, it is necessary to ensure that each feature vector corresponds to a clear fault label. For example, a feature vector represents the operating status of a device under high temperature and high humidity conditions. If an arc fault occurs in the device under this state, the label of the feature vector is arc fault. After the label dataset is constructed, it is segmented to obtain the training dataset and the validation dataset. Data segmentation usually uses the random segmentation method to divide the dataset into a training set and a validation set. The training set is used for model training, and the validation set is used for model evaluation. When segmenting data, it is necessary to ensure that the distribution of the training set and the validation set is consistent to improve the credibility of the model evaluation results.Based on the training data set, the classification model can be trained by selecting classification algorithms such as random forest, support vector machine (SVM) or deep neural network (DNN). Through multiple iterations of training data sets, the model gradually learns the patterns and rules in the data. For example, using the random forest algorithm, multiple decision tree models can be constructed and their results can be voted on to improve the accuracy and stability of classification. After the model training is completed, the trained classification model is evaluated using the validation data set to obtain the evaluation results. The evaluation indicators can include accuracy, precision, recall and F1 value. For example, the evaluation result of a certain evaluation shows that the accuracy of the classification model is 90%, indicating that the model can correctly classify the fault type in 90% of cases. According to the evaluation results, the model can be optimized, such as adjusting model parameters, using more complex algorithms or increasing the amount of training data, so as to obtain an optimized classification model. The optimized classification model is applied to the feature vector data set to classify the data and obtain classified data. For example, a feature vector can be classified as an arc fault, a ground fault or other fault type through the optimized random forest model. The classified data represents the fault type corresponding to each feature vector, providing a basis for subsequent data cleaning and fault analysis. Finally, the classified data is cleaned to obtain a cleaned data set. The purpose of data cleaning is to remove noise and outliers in the classified data and improve the quality and reliability of the data. For example, some misclassified results may be generated during the classification process. Through data cleaning, these misclassified data can be eliminated to obtain a purer cleaned data set.
[0132] In a specific embodiment, the process of executing step S102 may specifically include the following steps:
[0133] (1) constructing a Gaussian mixture model based on the cleaned data set and obtaining Gaussian mixture model parameters;
[0134] (2) Perform anomaly detection on each data point in the cleaned data set, calculate its anomaly score based on the Gaussian mixture model parameters, and obtain an anomaly score set;
[0135] (3) Setting an anomaly threshold according to the anomaly score set, marking data points with an anomaly score higher than the anomaly threshold as anomalies, and obtaining a marked data set;
[0136] (4) Performing principal component analysis on the labeled data set, extracting the principal component loading matrix, and obtaining the principal component loading matrix;
[0137] (5) Performing projection transformation on the labeled data set according to the principal component loading matrix to obtain a projected data set;
[0138] (6) Performing kernel discriminant analysis on the projected data set, constructing a discriminant model, and obtaining a discriminant model;
[0139] (7) Using the discriminant model to discriminate the outliers in the projected data set, and obtaining the outlier discrimination result;
[0140] (8) According to the outlier identification result, the data identified as outliers are removed from the projection data set to obtain a denoised projection data set;
[0141] (9) performing back-projection transformation on the denoised projection data set to obtain a back-projection data set;
[0142] (10) Convert the data format of the back-projection data set to obtain a denoised data set.
[0143] It should be noted that a Gaussian mixture model is constructed for the cleaned data set. The Gaussian mixture model (GMM) is a probabilistic model for clustering and density estimation. By assuming that the data is composed of multiple Gaussian distributions, the potential structure of the data can be captured. When constructing a GMM, the parameters of the model need to be determined, including the mean, covariance matrix, and mixing coefficient of each Gaussian distribution. The expectation maximization (EM) algorithm can be used for parameter estimation, and the model parameters can be repeatedly optimized until convergence. For example, for the current data of GIS equipment, the parameters of each Gaussian distribution can be obtained through GMM modeling to reflect the different modes of the current data. After obtaining the parameters of the Gaussian mixture model, anomaly detection is then performed on each data point in the cleaned data set. According to the GMM parameters, the anomaly score of each data point can be calculated. The calculation method of the anomaly score is based on the probability density value of the data point under each Gaussian distribution. The lower the probability density value, the more likely the data point is an anomaly point. For example, the calculation result of the anomaly score of a current data point shows that its probability density value under all Gaussian distributions is very low, indicating that the data point may be an anomaly. The anomaly scores of all data points are collected to form an anomaly score set. According to the anomaly score set, an anomaly threshold is set to distinguish normal points from anomalies. The setting of the threshold can be determined based on experience or by statistical methods, for example, the 99th percentile of the anomaly score set is taken as the threshold. Data points with anomaly scores higher than the anomaly threshold are marked as anomalies to obtain a marked data set. The marked data set contains the marking information of normal points and anomalies, which provides a basis for subsequent processing. Next, the marked data set is subjected to principal component analysis (PCA) to extract the principal component load matrix. PCA is a dimensionality reduction method that projects the original data into a low-dimensional space through linear transformation so that the projected data retains the variance information of the original data as much as possible. The principal component load matrix is the core of the PCA transformation, which represents the direction of each principal component in the original feature space. For example, for multidimensional current data, PCA can extract the main change mode, reduce the dimension of the data, and improve the computational efficiency. According to the principal component load matrix, the marked data set is subjected to projection transformation to obtain a projected data set. Projection transformation is to project the original data into the principal component space to obtain the reduced-dimensional data. For example, for three-dimensional current data, PCA can be used to project it into two-dimensional or one-dimensional space, reducing the complexity of the data and facilitating subsequent analysis. Perform kernel discriminant analysis (KDA) on the projected data set to build a discriminant model. KDA is a classification method that combines kernel methods and linear discriminant analysis. It maps data into a high-dimensional space through nonlinear transformation, making data of different categories more linearly separable in the high-dimensional space. The discriminant model is used to distinguish normal points from abnormal points. For example, KDA can be used to separate normal current data from abnormal current data in the projected data set, thereby improving the accuracy of anomaly detection. The discriminant model is used to discriminate the abnormal points in the projected data set to obtain the abnormal point discrimination result.The discrimination result indicates whether each data point is a normal point or an abnormal point. For example, through the KDA model, it is possible to determine whether a certain projected current data point is abnormal and obtain the discrimination result of the abnormal point. According to the discrimination result of the abnormal point, the data identified as an abnormal point is removed from the projected data set to obtain a denoised projected data set. After removing the abnormal points, only normal points remain in the projected data set, and the purity and reliability of the data are greatly improved. For example, by removing the abnormal current data points, a more accurate and reliable current data set is obtained. The denoised projected data set is back-projected to restore it to the original feature space to obtain a back-projected data set. Back-projection transformation is to remap the reduced-dimensional data back to the original high-dimensional space. For example, the two-dimensional current data is restored to the three-dimensional space through back-projection transformation to ensure the integrity of the data. Finally, the back-projected data set is converted to a denoised data set. Data format conversion is to convert the processed data into a standard format to facilitate subsequent analysis and processing. For example, the back-projected current data is converted into a standard time series format to facilitate further fault analysis and modeling.
[0144] In a specific embodiment, the process of executing step S103 may specifically include the following steps:
[0145] (1) Standardize the denoised data set to obtain a standardized data set;
[0146] (2) constructing a variational autoencoder model according to the standardized data set, including an encoder and a decoder network, to obtain a variational autoencoder model;
[0147] (3) Apply the encoder network in the variational autoencoder model to the standardized dataset to obtain the latent space distribution;
[0148] (4) sampling latent variables from the latent space distribution to obtain a set of latent variables;
[0149] (5) Input the latent variable set into the decoder network of the variational autoencoder model to obtain a reconstructed data set;
[0150] (6) Calculate the reconstruction loss based on the standardized dataset and the reconstructed dataset;
[0151] (7) Calculate the KL divergence loss based on the latent space distribution;
[0152] (8) Add the reconstruction loss and KL divergence loss to get the total loss;
[0153] (9) updating the parameters of the variational autoencoder model based on the total loss to obtain an updated variational autoencoder model;
[0154] (10) The updated variational autoencoder model is used to fill the missing values of the denoised dataset to obtain a filled dataset.
[0155] Specifically, the denoised data set is standardized. The purpose of standardization is to scale different features of the data to the same scale to eliminate the dimensional differences between the features. Standardization usually uses the mean-standard deviation standardization method to subtract the mean from the value of each feature and then divide it by the standard deviation, so that the processed data has a distribution with a mean of 0 and a standard deviation of 1. For example, when processing data such as current and temperature of GIS equipment, standardization can eliminate the impact of different dimensions, making subsequent modeling more stable and effective. After obtaining the standardized data set, a variational autoencoder (VAE) model is constructed based on these data. VAE is a generative model that can generate data and fill missing values by learning the potential distribution of data. The VAE model consists of an encoder and a decoder network. The encoder maps the input data to the latent space, and the decoder reconstructs the variables in the latent space into the original data. The encoder and decoder are usually implemented using multi-layer neural networks. For example, when processing the operating data of GIS equipment, the encoder network can compress high-dimensional sensor data into a low-dimensional latent space, while the decoder reconstructs high-dimensional data based on these low-dimensional representations. Applying the encoder network in the VAE model to the standardized data set can obtain the latent space distribution. The latent space distribution is the probability distribution of the latent variables output by the encoder network, which is usually assumed to be a Gaussian distribution. In this process, the encoder network maps the data to the latent space by learning the characteristics of the data. For example, after the current and temperature data of GIS equipment are processed by the encoder network, the latent variable distribution obtained reflects the potential structure and relationship of these data. Sampling latent variables from the latent space distribution can obtain a set of latent variables. The sampling process is usually implemented through the reparameterization technique, which introduces random noise into the generation process of the latent variables so that the model can be trained by gradient descent. For example, sampling latent variables from the Gaussian distribution of the latent space can obtain multiple latent variables, which can be used for data reconstruction and generation. The latent variable set is input into the decoder network of the VAE model to obtain a reconstructed dataset. The decoder network maps the latent variables back to the high-dimensional space and reconstructs the approximation of the original data. For example, the current and temperature data generated by the decoder network based on the latent variables should be very close to the original data, so as to achieve data reconstruction. Based on the standardized dataset and the reconstructed dataset, the reconstruction loss can be calculated. The reconstruction loss usually uses the mean square error (MSE) or cross entropy loss to measure the difference between the reconstructed data and the original data. The smaller the reconstruction loss, the closer the data generated by the decoder is to the original data. For example, the size of the reconstruction loss can reflect the accuracy of the decoder in reconstructing the running data of the GIS device. At the same time, the KL divergence loss is calculated based on the latent space distribution. The KL divergence loss measures the difference between the latent distribution and the standard normal distribution, and encourages the latent distribution to be close to the standard normal distribution, so that the model has better generation ability and robustness. The reconstruction loss and the KL divergence loss are added to get the total loss.The total loss comprehensively considers the accuracy of data reconstruction and the rationality of the potential distribution, and is the optimization target of model training. The parameters of the VAE model are updated based on the total loss, and the parameters of the encoder and decoder networks are adjusted through the back-propagation algorithm, so that the model can better reconstruct the data and learn the potential distribution. The optimized VAE model has higher accuracy and reliability in reconstructing data and handling missing values. Using the updated VAE model, the missing values of the denoised dataset can be filled. The denoised dataset is input into the encoder network to obtain the latent space distribution, and the latent variables are sampled from it, and then the filling values are generated through the decoder network. For example, in the operating data of the GIS equipment, if the temperature data at certain times is missing due to sensor failure, the VAE model can generate reasonable filling values through the learned latent distribution to obtain a filled dataset.
[0156] In a specific embodiment, the process of executing step S104 may specifically include the following steps:
[0157] (1) Extract features from the filled data set to obtain a feature data set;
[0158] (2) constructing three-dimensional point cloud data based on the feature data set to obtain original point cloud data;
[0159] (3) Perform point cloud filtering on the original point cloud data to remove outliers and obtain filtered point cloud data;
[0160] (4) performing point cloud segmentation on the filtered point cloud data, extracting key point cloud clusters, and obtaining segmented point cloud data;
[0161] (5) Perform three-dimensional reconstruction on the segmented point cloud data, construct a three-dimensional mesh model, and obtain a rough three-dimensional model;
[0162] (6) Mesh optimization of the rough three-dimensional model, including vertex refinement and facet subdivision, to obtain an optimized three-dimensional model;
[0163] (7) performing texture mapping on the optimized three-dimensional model, giving the model a real texture according to the feature data set, and obtaining a textured three-dimensional model;
[0164] (8) Setting the lighting of the textured three-dimensional model, including setting the ambient light and the point light source, to obtain an illuminated three-dimensional model;
[0165] (9) Generate a rendering image based on the lighting three-dimensional model to obtain a three-dimensional rendering image of the GIS device;
[0166] (10) The 3D rendering image of the GIS equipment is combined with the feature data set to obtain a 3D model of the GIS equipment.
[0167] Specifically, feature extraction is performed from the filling data set. The purpose of feature extraction is to convert raw data into features that can better reflect the essence of the problem, thereby providing high-quality data support for subsequent three-dimensional modeling. Feature extraction can be achieved through methods such as statistical analysis, signal processing, and machine learning. For example, for the operating data of GIS equipment, key features such as temperature, current, and voltage can be extracted, which can accurately reflect the operating status and health of the equipment. After obtaining the feature data set, the next step is to construct three-dimensional point cloud data based on the feature data set. Point cloud data is a commonly used three-dimensional data representation method, consisting of a large number of points, each of which has three-dimensional coordinates and other attributes (such as color, intensity, etc.). By spatially transforming and mapping the feature data, raw point cloud data can be generated. For example, when processing the spatial layout data of GIS equipment, the location information of temperature sensors, current sensors, etc. can be mapped into three-dimensional space according to the physical size and position of the equipment to form a complete raw point cloud data. Raw point cloud data often contains noise and outliers, which will affect the accuracy of subsequent three-dimensional modeling, so point cloud filtering is required. The purpose of point cloud filtering is to remove noise points and outliers in point cloud data and improve the purity of data. Common point cloud filtering methods include statistical filtering, radius filtering, and conditional filtering. For example, through the statistical filtering method, the neighborhood density of each point can be calculated, and the points with a density below a certain threshold can be regarded as outliers and removed, thereby obtaining the filtered point cloud data. The filtered point cloud data still contains a large number of points, which need to be further segmented to extract key point cloud clusters. The purpose of point cloud segmentation is to segment the point cloud data into multiple independent parts, each of which corresponds to an object or an area. Common point cloud segmentation methods include region growing-based segmentation, clustering-based segmentation, and graph-based segmentation. For example, when processing point cloud data of GIS equipment, a region growing algorithm can be used to segment the point cloud data into different components, such as circuit breakers, busbars, and disconnectors, according to the spatial position and attributes of the points, thereby obtaining segmented point cloud data. Next, the segmented point cloud data is 3D reconstructed to construct a 3D mesh model. The purpose of 3D reconstruction is to convert point cloud data into a 3D mesh representation to better display the shape and structure of the object. Common 3D reconstruction methods include Delaunay triangulation, Poisson reconstruction, and surface reconstruction. For example, through the Poisson reconstruction method, a continuous 3D surface can be generated based on the segmented point cloud data, thus obtaining a rough 3D model. Rough 3D models often contain some irregular and rough parts, which require mesh optimization. The purpose of mesh optimization is to refine the vertices and faces of the 3D model to make the model smoother and more detailed. Commonly used mesh optimization methods include vertex refinement, face subdivision, and mesh simplification.For example, through the vertex refinement method, the number of vertices of the three-dimensional model can be increased to make the model surface smoother; through the patch subdivision method, the patch structure of the three-dimensional model can be refined to improve the detail expression of the model, thereby obtaining an optimized three-dimensional model. The optimized three-dimensional model needs to be texture mapped to give the model a real texture according to the feature data set. The purpose of texture mapping is to map a two-dimensional image to the surface of the three-dimensional model so that the model looks more realistic. Common texture mapping methods include UV mapping, projection mapping, and procedural textures. For example, through the UV mapping method, the color and intensity information in the feature data can be mapped to the surface of the three-dimensional model so that the model has a real texture, thereby obtaining a textured three-dimensional model. In order to make the textured three-dimensional model look more realistic, it is also necessary to set the light. The purpose of the light setting is to add ambient light and point light sources to the three-dimensional model so that the model has shadow and highlight effects. Common light setting methods include ambient light setting, point light source setting, and illumination model. For example, by setting ambient light and point light sources, the lighting effect in reality can be simulated to make the three-dimensional model more realistic, thereby obtaining an illuminated three-dimensional model. Generating a rendered image based on an illuminated 3D model is the process of converting a 3D model into a 2D image. The generation of rendered images is usually achieved through technologies such as ray tracing, rasterization, and global illumination. For example, through the ray tracing method, the propagation and reflection of light in 3D space can be simulated to generate high-quality 2D rendered images, thereby obtaining 3D rendered images of GIS equipment.
[0168] In a specific embodiment, the process of executing step S105 may specifically include the following steps:
[0169] (1) Preprocessing the real-time operation data, including data standardization and feature scaling, to obtain processed real-time data;
[0170] (2) Mapping the processed real-time data onto the GIS equipment three-dimensional model to obtain a three-dimensional model with data;
[0171] (3) performing model segmentation on the three-dimensional model with data to obtain a segmented model data set;
[0172] (4) constructing a model graph according to the segmented model data set to obtain model graph data;
[0173] (5) Perform graph convolution operation on the model graph data to extract graph features and obtain graph feature data;
[0174] (6) Input the graph feature data into the graph attention network to obtain the attention-weighted graph features;
[0175] (7) Performing a pooling operation on the attention weighted graph features to obtain a pooled feature tensor;
[0176] (8) Input the pooled feature tensor into the fully connected layer to perform feature transformation to obtain the potential fault feature vector;
[0177] (9) Normalizing the potential fault feature vector to obtain a normalized fault feature vector;
[0178] (10) Input the normalized fault feature vector into the classification layer to obtain the fault feature vector.
[0179] Specifically, the real-time operation data is preprocessed. The preprocessing step includes data standardization and feature scaling, the purpose of which is to convert data of different scales to the same scale to eliminate dimensional differences and improve the consistency and comparability of the data. Data standardization usually adopts the mean-standard deviation standardization method, which subtracts the mean value of each feature and divides it by its standard deviation, so that the processed data has a distribution with a mean of 0 and a standard deviation of 1. For example, when processing the current, temperature and humidity data of GIS equipment, these data can be converted to the same scale through standardization, so that they have the same importance in subsequent processing and analysis. After completing data standardization and feature scaling, the processed real-time data is mapped to the three-dimensional model of the GIS equipment to obtain a three-dimensional model with data. This step is to combine the real-time data with the three-dimensional model so that each data point has a clear physical position in the three-dimensional space. For example, when processing the temperature sensor data of the GIS equipment, the temperature value of each sensor can be mapped to the corresponding position in the three-dimensional model of the equipment, so that the three-dimensional model not only shows the geometric structure of the equipment, but also can intuitively show the temperature distribution. Next, the three-dimensional model with data is segmented to obtain a segmented model data set. The purpose of model segmentation is to segment a three-dimensional model into multiple independent parts, each of which corresponds to a specific component or area. Common model segmentation methods include segmentation based on spatial position and segmentation based on attributes. For example, in the three-dimensional model of GIS equipment, the model can be segmented into different components such as circuit breakers, busbars, disconnectors, etc. according to the structure of the equipment, thereby obtaining a segmented model data set. A model graph is constructed based on the segmented model data set to obtain model graph data. A model graph is a method of representing a three-dimensional model as a graph structure, in which each node represents a segmented model part and each edge represents the connection relationship between the parts. For example, a circuit breaker and a busbar can be represented as nodes in a model graph, and the connection between them is represented as an edge, thereby constructing a model graph of the entire GIS equipment. A graph convolution operation is performed on the model graph data to extract graph features and obtain graph feature data. A graph convolution network (GCN) is a neural network specifically used to process graph structure data. By performing convolution operations on a graph, the feature information of the graph can be effectively extracted. For example, through GCN, the connection features and spatial relationship features between the various components of GIS equipment can be extracted, which can reflect the operating status and potential faults of the equipment. The graph feature data is input into the graph attention network to obtain the attention-weighted graph features. The graph attention network (GAT) is a network that adds an attention mechanism to GCN. By calculating the attention weight of each node, the contribution of different nodes to the features can be dynamically adjusted, thereby extracting graph features more accurately. For example, when processing the graph feature data of GIS equipment, GAT can assign different weights according to the importance of each component, so that the features of key components receive more attention, thereby obtaining more accurate fault features.The attention weighted graph features are pooled to obtain a pooled feature tensor. The purpose of the pooling operation is to reduce the dimension and compress the feature data, retain key features, and remove redundant information. Common pooling methods include maximum pooling and average pooling. For example, through maximum pooling, the maximum value can be selected on each feature dimension to obtain a representative pooled feature tensor, which can better reflect the operating status of the GIS equipment. The pooled feature tensor is input into the fully connected layer for feature transformation to obtain a potential fault feature vector. The fully connected layer is a neural network layer that maps input features to output feature space through linear transformation and nonlinear activation function. For example, through the fully connected layer, the pooled feature tensor can be converted into a potential fault feature vector, which can reflect the potential failure mode and characteristics of the equipment. The potential fault feature vector is normalized to obtain a normalized fault feature vector. The purpose of normalization is to scale the value of the feature vector to a specific range, usually between 0 and 1, to improve the stability and comparability of the feature. For example, through normalization processing, the value of the potential fault feature vector can be limited to between 0 and 1, so that different features have the same dimension, which is convenient for subsequent classification and analysis. The normalized fault feature vector is input into the classification layer to obtain the fault feature vector. The classification layer is a neural network layer used for fault classification. Through the trained classification model, the input feature vector can be mapped to the fault category. For example, in the fault detection of GIS equipment, the normalized fault feature vector can be classified into different fault types, such as arc fault, ground fault, etc., through the classification layer, so as to obtain the fault feature vector. These fault feature vectors can provide the fault type and fault probability of the equipment, and provide an important reference for the maintenance and management of the equipment.
[0180] In a specific embodiment, the process of executing step S106 may specifically include the following steps:
[0181] (1) constructing an input tensor according to the fault feature vector to obtain input tensor data;
[0182] (2) Inputting the input tensor data into the convolutional neural network model to obtain a convolutional feature map;
[0183] (3) Perform a maximum pooling operation on the convolution feature map to obtain a pooled feature map;
[0184] (4) Flatten the pooled feature map into a one-dimensional vector to obtain a one-dimensional feature vector;
[0185] (5) Input the one-dimensional feature vector into the long short-term memory network to obtain time series feature data;
[0186] (6) Input the time series feature data into the attention layer, perform feature weighting, and obtain weighted feature data;
[0187] (7) Input the weighted feature data into the fully connected layer and perform nonlinear transformation to obtain the fault probability distribution;
[0188] (8) Classify according to the fault probability distribution and obtain the classification result;
[0189] (9) Mapping the classification results into fault codes to obtain fault code data;
[0190] (10) Combine the fault code data with the corresponding fault description information to obtain fault analysis data.
[0191] It should be noted that the input tensor is constructed based on the fault feature vector. The fault feature vector is extracted from the GIS equipment operation data, and these feature vectors need to be converted into a tensor form suitable for neural network processing. The input tensor is a multidimensional array that can represent the spatial structure and attributes of the fault feature. For example, if the fault feature vector includes multiple parameters such as temperature, current, and voltage, these parameters can be combined into a two-dimensional or three-dimensional tensor to represent the changes in the fault feature at different time points. The constructed input tensor data is input into the convolutional neural network (CNN) model, which is a deep learning model that is good at processing image and spatial data. Through convolution operations, CNN can extract local features of the input data and gradually extract higher-level features through multiple layers of convolution. For example, when processing the fault features of GIS equipment, CNN can extract the spatial relationship and change pattern in the feature vector through the convolution layer to obtain a convolution feature map. The maximum pooling operation is performed on the convolution feature map to reduce the dimension and compress the feature map, retain key features, and remove redundant information. Maximum pooling reduces the size of the feature map and enhances the robustness of the feature by selecting the maximum value in the local area of the feature map. For example, through the maximum pooling operation, the most significant fault feature in the convolution feature map can be extracted to obtain the pooled feature map. The purpose of flattening the pooled feature map into a one-dimensional vector is to convert a two-dimensional or multi-dimensional feature map into a one-dimensional feature vector for subsequent processing and analysis. The flattening operation forms a long vector by expanding the elements of the feature map by rows or columns. For example, through the flattening operation, all the feature values in the pooled feature map can be combined into a one-dimensional feature vector for easy input to the subsequent neural network layer. The one-dimensional feature vector is input into the long short-term memory network (LSTM), which is a recurrent neural network that is good at processing time series data and can capture long-term and short-term dependencies in the data. Through the LSTM network, the input fault feature vector can be analyzed in time series to extract the time pattern and change trend in the feature data. For example, when processing the time series fault data of GIS equipment, LSTM can analyze the changes of the feature vector at different time points to obtain time series feature data. The time series feature data is input into the attention layer. The purpose of the attention layer is to weight the feature data, highlight key features, and suppress irrelevant features. The attention mechanism calculates the importance weight of each feature and weights the features according to the weight, thereby enhancing the model's attention to important features. For example, through the attention mechanism, the fault features of GIS equipment can be weighted to highlight the most important features for fault identification and obtain weighted feature data. The weighted feature data is input into the fully connected layer and nonlinearly transformed to map the weighted feature data to the fault probability distribution. The fully connected layer combines and transforms the input data through linear transformation and nonlinear activation function to obtain the fault probability distribution.For example, through the fully connected layer, the weighted feature data can be converted into the probability distribution of various types of faults, reflecting the possibility of each fault. Classification is performed according to the fault probability distribution to obtain the classification result. The classification process is to take the fault type corresponding to the maximum probability in the fault probability distribution as the final classification result. For example, if the probability of arc fault is the largest in the fault probability distribution of a certain GIS equipment, it is classified as an arc fault, thereby obtaining the classification result. The classification result is mapped to a fault code to obtain fault code data. The fault code is a concise fault identifier. By mapping the classification result to a predefined fault code, the fault information can be easily recorded and transmitted. For example, the arc fault is mapped to the fault code "E01" and the ground fault is mapped to the fault code "G02", thereby forming the fault code data. Finally, the fault code data is combined with the corresponding fault description information to obtain the fault analysis data. The fault description information includes a detailed explanation of the fault type, possible causes and recommended treatment measures. By combining the fault code with these description information, a complete fault analysis report can be generated. For example, for the fault code "E01", the corresponding description information can be provided: "Arc fault, which may be caused by insulation damage or excessive voltage. It is recommended to check the insulation status and voltage level of the equipment." In this way, detailed fault analysis data can be provided to maintenance personnel to help them quickly locate and handle equipment faults.
[0192] The embodiment of the present invention further provides a GIS equipment operation fault analysis system based on three-dimensional modeling, and the GIS equipment operation fault analysis system based on three-dimensional modeling specifically includes:
[0193] The collection module 201 is used to perform data cleaning on the collected GIS equipment operation status data, environmental data and historical fault data to obtain a cleaned data set;
[0194] A denoising module 202, configured to perform data denoising on the cleaned data set to obtain a denoised data set;
[0195] A filling module 203 is used to fill missing values in the denoised data set by using a variational autoencoder to obtain a filled data set;
[0196] A modeling module 204 is used to perform three-dimensional modeling according to the filled data set to obtain a three-dimensional model of the GIS device;
[0197] The acquisition module 205 is used to acquire real-time operation data of the GIS device, and extract fault features of the three-dimensional model of the GIS device through the real-time operation data to obtain a fault feature vector;
[0198] The analysis module 206 is used to input the fault feature vector into a deep neural network model to perform fault analysis and obtain fault analysis data.
[0199] Through the collaborative work of the above modules, the noise and outliers in the data are effectively removed by cleaning the collected GIS equipment operation status data, environmental data and historical fault data, ensuring the quality and reliability of the data. This step includes sub-steps such as outlier detection and processing, data fusion, feature engineering, label data set construction, data segmentation, classification model training and optimization. Each sub-step has been carefully designed to maximize the effect and efficiency of data cleaning. Next, the data is denoised by using the Gaussian mixture model to effectively remove outliers in the data, and the purity of the data is further improved by principal component analysis and kernel discriminant analysis to ensure that the data processed later is more accurate and reliable. The multi-level processing method of this step not only retains the main features of the data in the denoised data set, but also effectively reduces the interference of noise on the analysis results. Subsequently, the missing values of the denoised data set are filled by the variational autoencoder, which further enhances the integrity and accuracy of the data. The variational autoencoder maps the standardized data set to the latent space through the dual processing of the encoder and decoder networks, and reconstructs the data by sampling latent variables, finally achieving the filling of missing values. The comprehensive calculation and optimization of reconstruction loss and KL divergence loss ensure the quality of the filled data set, so that the data still maintains a high degree of consistency and integrity in the filled state. On this basis, three-dimensional modeling is carried out according to the filled data set, and a high-precision three-dimensional model of GIS equipment is constructed through steps such as feature extraction, point cloud data construction, point cloud filtering, point cloud segmentation, three-dimensional reconstruction, mesh optimization, texture mapping and lighting setting. This model not only has real physical properties, but also can intuitively display the structure and status of the equipment through rendered images, providing a solid foundation for fault analysis. The collection and processing of real-time operation data is an important part of fault analysis. By preprocessing the real-time operation data, it is mapped to the three-dimensional model, and model segmentation, graph convolution operation, graph attention network weighting, pooling operation and feature transformation are performed to extract the potential fault feature vector. This process makes full use of the combination of real-time data and three-dimensional models, making the extraction of fault features more accurate and comprehensive. After the fault feature vector is extracted, the fault analysis is further performed through the deep neural network model. The combination of convolutional neural network, long short-term memory network and attention mechanism fully explores the spatiotemporal correlation of fault features and ensures the accuracy and real-time performance of fault analysis. Finally, the fault feature vector undergoes multi-level feature transformation and classification to generate fault probability distribution and fault code data, and combined with the corresponding fault description information, the fault analysis data is obtained. This series of technical features and steps, through the organic combination of deep learning and 3D modeling, realizes the efficient analysis and prediction of GIS equipment operation faults, significantly improves the real-time, accuracy and comprehensiveness of fault detection, and solves the problem that traditional fault detection methods are difficult to accurately identify and predict faults in complex environments.
[0200] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the invention, which should be included in the scope of the claims of the present invention.
Claims
1. A GIS equipment operation failure analysis method based on 3D modeling, Features: include, Perform data cleaning on the collected GIS equipment operating status data, environmental data and historical fault data to obtain a cleaned data set; Performing data denoising on the cleaned data set to obtain a denoised data set; Filling missing values in the denoised data set by a variational autoencoder to obtain a filled data set; Perform three-dimensional modeling according to the filled data set to obtain a three-dimensional model of the GIS equipment; Collecting real-time operation data of the GIS equipment, and extracting fault features of the three-dimensional model of the GIS equipment through the real-time operation data to obtain a fault feature vector; The fault feature vector is input into a deep neural network model to perform fault analysis to obtain fault analysis data.
2. A GIS equipment operation failure analysis method based on three-dimensional modeling as claimed in claim 1, Features: The step of performing data cleaning on the collected GIS equipment operating status data, environmental data and historical fault data to obtain a cleaned data set includes: Perform outlier detection and processing on the GIS equipment operation status data to obtain the processed operation status data; Performing data fusion according to the processed operating status data and environmental data to obtain fused data; Performing feature engineering on the fused data to extract key feature vectors to obtain a feature vector data set; Constructing a label data set according to the feature vector data set and the historical fault data; Performing data segmentation on the labeled data set to obtain a training data set and a verification data set; Training a classification model based on the training data set to obtain a trained classification model; Using the validation data set to evaluate the trained classification model to obtain an evaluation result; Optimizing the trained classification model according to the evaluation result to obtain an optimized classification model; Applying the optimized classification model to the feature vector data set to classify the data to obtain classified data; The classified data is cleaned to obtain a cleaned data set.
3. A GIS equipment operation failure analysis method based on three-dimensional modeling as claimed in claim 1, Features: The step of performing data denoising on the cleaned data set to obtain a denoised data set includes: Constructing a Gaussian mixture model according to the cleaned data set to obtain Gaussian mixture model parameters; Performing anomaly detection on each data point in the cleaned data set, calculating its anomaly score according to the Gaussian mixture model parameters, and obtaining an anomaly score set; An anomaly threshold is set according to the anomaly score set, and data points with an anomaly score higher than the anomaly threshold are marked as anomalies, so as to obtain a marked data set; Performing principal component analysis on the labeled data set, extracting a principal component loading matrix, and obtaining a principal component loading matrix; Performing a projection transformation on the labeled data set according to the principal component loading matrix to obtain a projected data set; Performing a kernel discriminant analysis on the projected data set, constructing a discriminant model, and obtaining a discriminant model; Using the discriminant model to discriminate abnormal points in the projection data set to obtain abnormal point discrimination results; According to the outlier identification result, data identified as outliers are removed from the projection data set to obtain a denoised projection data set; Performing a back-projection transformation on the denoised projection data set to obtain a back-projection data set; The back-projection data set is converted into a data format to obtain the denoised data set.
4. A GIS equipment operation failure analysis method based on three-dimensional modeling as claimed in claim 1, Features: The step of filling missing values of the denoised data set by using a variational autoencoder to obtain a filled data set comprises: Performing standardization processing on the denoised data set to obtain a standardized data set; Constructing a variational autoencoder model according to the standardized data set, including an encoder and a decoder network, to obtain a variational autoencoder model; Applying an encoder network in the variational autoencoder model to the standardized dataset to obtain a latent space distribution; Sampling latent variables from the latent space distribution to obtain a set of latent variables; Inputting the latent variable set into the decoder network of the variational autoencoder model to obtain a reconstructed data set; Calculate the reconstruction loss based on the standardized data set and the reconstructed data set; Calculate the KL divergence loss according to the latent space distribution; Add the reconstruction loss and the KL divergence loss to get the total loss; Based on the total loss, the variational autoencoder model is updated with parameters to obtain an updated variational autoencoder model; The updated variational autoencoder model is used to fill missing values in the denoised data set to obtain the filled data set.
5. A GIS equipment operation failure analysis method based on three-dimensional modeling as claimed in claim 1, Features: The step of performing three-dimensional modeling according to the filled data set to obtain a three-dimensional model of the GIS device includes: Performing feature extraction on the filled data set to obtain a feature data set; Constructing three-dimensional point cloud data according to the feature data set to obtain original point cloud data; Performing point cloud filtering on the original point cloud data to remove outliers and obtain filtered point cloud data; Performing point cloud segmentation on the filtered point cloud data, extracting key point cloud clusters, and obtaining segmented point cloud data; Performing three-dimensional reconstruction on the segmented point cloud data, constructing a three-dimensional mesh model, and obtaining a rough three-dimensional model; Performing mesh optimization on the rough three-dimensional model, including vertex refinement and facet subdivision, to obtain an optimized three-dimensional model; Performing texture mapping on the optimized three-dimensional model, giving the model a real texture according to the feature data set, and obtaining a textured three-dimensional model; Performing lighting settings on the textured three-dimensional model, including setting ambient light and point light sources, to obtain an illuminated three-dimensional model; Generate a rendering image according to the lighting three-dimensional model to obtain a three-dimensional rendering image of the GIS device; The three-dimensional rendering image of the GIS device is combined with the feature data set to obtain the three-dimensional model of the GIS device.
6. A GIS equipment operation failure analysis method based on three-dimensional modeling as claimed in claim 5, Features: The step of collecting real-time operation data of the GIS device and extracting fault features of the three-dimensional model of the GIS device through the real-time operation data to obtain a fault feature vector includes: Preprocessing the real-time operation data, including data standardization and feature scaling, to obtain processed real-time data; Mapping the processed real-time data onto the three-dimensional model of the GIS device to obtain a three-dimensional model with data; Performing model segmentation on the three-dimensional model with data to obtain a segmented model data set; Construct a model graph according to the segmented model data set to obtain model graph data; Performing a graph convolution operation on the model graph data to extract graph features and obtain graph feature data; Inputting the graph feature data into a graph attention network to obtain attention-weighted graph features; Performing a pooling operation on the attention weighted graph features to obtain a pooled feature tensor; Inputting the pooled feature tensor into a fully connected layer to perform feature transformation to obtain a potential fault feature vector; Normalizing the potential fault feature vector to obtain a normalized fault feature vector; The normalized fault feature vector is input into the classification layer to obtain the fault feature vector.
7. A GIS equipment operation failure analysis method based on three-dimensional modeling as claimed in claim 6, Features: The step of inputting the fault feature vector into a deep neural network model for fault analysis to obtain fault analysis data comprises: Constructing an input tensor according to the fault feature vector to obtain input tensor data; Inputting the input tensor data into a convolutional neural network model to obtain a convolutional feature map; Performing a maximum pooling operation on the convolutional feature map to obtain a pooled feature map; Flattening the pooled feature map into a one-dimensional vector to obtain a one-dimensional feature vector; Inputting the one-dimensional feature vector into a long short-term memory network to obtain time series feature data; Input the time series feature data into the attention layer, perform feature weighting, and obtain weighted feature data; Inputting the weighted feature data into a fully connected layer and performing nonlinear transformation to obtain a fault probability distribution; Perform classification according to the fault probability distribution to obtain a classification result; Mapping the classification result into a fault code to obtain fault code data; The fault code data is combined with corresponding fault description information to obtain the fault analysis data.
8. A GIS equipment operation fault analysis system based on three-dimensional modeling, used to execute the GIS equipment operation fault analysis method based on three-dimensional modeling as claimed in any one of claims 1 to 7, It is characterized in that include: The acquisition module is used to perform data cleaning on the collected GIS equipment operating status data, environmental data and historical fault data to obtain a cleaned data set; A denoising module, used for denoising the cleaned data set to obtain a denoised data set; A filling module, used for filling missing values of the denoised data set by using a variational autoencoder to obtain a filled data set; A modeling module, used for performing three-dimensional modeling according to the filled data set to obtain a three-dimensional model of the GIS equipment; A collection module is used to collect real-time operation data of GIS equipment, and extract fault features of the three-dimensional model of the GIS equipment through the real-time operation data to obtain a fault feature vector; The analysis module is used to input the fault feature vector into a deep neural network model to perform fault analysis and obtain fault analysis data.
9. The GIS equipment operation failure analysis system based on three-dimensional modeling according to claim 8, It is characterized in that The acquisition module further comprises: It is used to clean the collected GIS equipment operation status data, environmental data and historical fault data to obtain a clean data set; Used to perform data denoising on the cleaned data set to obtain a denoised data set; Used to fill missing values of the denoised data set through a variational autoencoder to obtain a filled data set; Used to perform three-dimensional modeling according to the filled data set to obtain a three-dimensional model of the GIS equipment; Used to collect real-time operation data of GIS equipment, and extract fault features of the three-dimensional model of the GIS equipment through the real-time operation data to obtain a fault feature vector; It is used to input the fault feature vector into a deep neural network model to perform fault analysis and obtain fault analysis data.
10. The GIS equipment operation failure analysis system based on three-dimensional modeling according to claim 9, It is characterized in that The analysis module further comprises: Used to construct an input tensor according to the fault feature vector to obtain input tensor data; Used to input the input tensor data into a convolutional neural network model to obtain a convolutional feature map; Used to perform a maximum pooling operation on the convolution feature map to obtain a pooled feature map; Used to flatten the pooled feature map into a one-dimensional vector to obtain a one-dimensional feature vector; Used to input the one-dimensional feature vector into a long short-term memory network to obtain time series feature data; Used to input the time series feature data into the attention layer, perform feature weighting, and obtain weighted feature data; Used to input the weighted feature data into a fully connected layer, perform nonlinear transformation, and obtain a fault probability distribution; Used to perform classification according to the fault probability distribution to obtain a classification result; Used to map the classification result into a fault code to obtain fault code data; Used to combine the fault code data with corresponding fault description information to obtain fault analysis data.
Citation Information
Patent Citations
Power distribution network initial fault identification method of time-frequency memory neural network based on adaptive wavelet and attention mechanism
CN114781458A
High-speed train bogie fault diagnosis method based on residual neural network
CN115545101A
Power grid data abnormal point detection method, system, equipment and medium
CN116756594A
Target area prediction method and device, computer equipment and storage medium
CN117010537A
Three-dimensional reconstruction method and equipment for power transmission line project
CN117372610A
Cited By
Mine network security operation system
CN119728294A
Mine network security operation system
CN119728294B
Intelligent denoising and missing value filling method for distributed optical fiber strain data
CN120011720A
High-voltage switchgear temperature field monitoring method and system based on digital twinning
CN120805014A
Automobile bus test AI diagnosis system and closed-loop feedback and self-learning method thereof
CN121477846A
