High-voltage line operation abnormal data detection method

Through the RIME-CNN-LSTM-Attention model and frost optimization algorithm, the accuracy and robustness problems in the detection of abnormal data of high-voltage line operation are solved, and the accurate identification and real-time monitoring of abnormal data of high-voltage line is realized, which improves the line Safety and stability.

CN118965155APending Publication Date: 2024-11-15BEIJING SHUNYI LIYUAN POWER SUPPLY ENG INSTALLATION CO +1
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202410974222.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-19
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The existing high-voltage line operation abnormal data detection methods have low accuracy and difficulty in adapting to complex data distribution and inaccurate feature extraction, resulting in high false alarm rate and lack of effective identification and low noise in the case of category imbalance and high noise. Detection capability affects the safety of the line system.

Method used

Using the RIME-CNN-LSTM-Attention model combined with timing prediction and deep learning, the hyperparameters are optimized to improve detection accuracy and Robust, and hyperparameter optimization of the model through frost optimization algorithm, real-time monitoring of abnormal data of high-voltage line operation.

Benefits of technology

It significantly improves the accuracy and robustness of high-voltage line operation abnormal data detection, reduces manual intervention, improves fault detection efficiency, reduces operation costs and maintenance difficulties, and enhances the safety and stability of the line.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118965155A_ABST
    Figure CN118965155A_ABST
Patent Text Reader

Abstract

The invention discloses a method for detecting abnormal operation data of a high-voltage line, and belongs to the technical field of data anomaly detection. Comprising the steps of collecting operation data traffic in real time, and performing data preprocessing; sending the processed data to a time sequence prediction model; generating a comparison flow change baseline through the time sequence prediction model; judging whether the data traffic is normal; an RIME-CNN-LSTM-Attention model is trained, and the RIME-CNN-LSTM- Detecting line operation abnormal data traffic; judging whether flow early warning is associated or not; abnormal data association early warning information is generated; generating early warning information of abnormal data of unknown reasons; unknown abnormal data analysis and diagnosis; and triggering a line abnormal data emergency measure. According to the invention, manual intervention and operation are significantly reduced, the safety and stability of high-voltage line operation are improved, and the operation cost and maintenance difficulty of the high-voltage line are reduced at the same time.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of data anomaly detection, and in particular to a method for detecting abnormal data of high-voltage line operation. Background Art

[0002] With the development and popularization of Internet of Things technology, various sensors related to high-voltage lines have been applied to production, making it possible for operation and maintenance personnel to remotely monitor the operating status of high-voltage lines, greatly reducing the burden on front-line operation and maintenance personnel. However, sensors only enrich the monitoring methods and cannot replace the decision-making of operators; in addition, even the same monitoring data, such as grounding current, has huge differences in high-voltage lines on different lines, in different environments and at different times. Therefore, it is impossible to judge the operating status of high-voltage lines by relying solely on one type of monitoring data. Even if the model trained by the most advanced machine learning (ML) algorithm is used, it cannot be actually applied because the data itself cannot fully reflect the operating status of high-voltage lines.

[0003] In order to monitor the operating status of high-voltage lines, with the help of rich IoT sensing technologies, various sensing devices are currently installed at key locations and nodes of high-voltage lines, including light temperature measurement, ground current, even partial discharge, water level, gas, etc., combined with operating data line voltage, load, etc., to form a dynamic digital world of real-time changes based on the topology of the transmission network and comprehensive perception of the operating status. On the other hand, the evaluation of the operating status of high-voltage lines, especially the discovery of defects, still requires irregular inspections by operation and maintenance personnel, supplemented by rich experience, to make correct judgments. With the rapid development of artificial intelligence technology today, especially the emergence of ChatGPT, it is possible to use big data to make a high-precision evaluation of the operating status of high-voltage lines. However, it is obvious that the rich monitoring data of high-voltage lines has not been fully and effectively utilized. There are several main reasons for this result: (1) High-voltage lines are diverse in variety and cross complex environments, making it difficult to accurately determine their operating status; (2) Monitoring data are abundant but scattered, and cannot be effectively integrated to form a unified whole; (3) High-voltage line data have many features, and their impact on the operating status varies, making it difficult to quantitatively identify correlations; and (4) It is difficult to construct an operating status recognition model.

[0004] At present, abnormal data detection technology can be divided into three categories: traditional methods, machine learning methods and deep learning methods. Traditional methods set relatively reasonable thresholds according to the characteristics of different abnormal data. Machine learning methods automatically identify abnormal data through a pre-given model, which is basically based on the characteristic data of a certain aspect of the high-voltage line for research and judgment, which is too one-sided. Deep learning has shown a stronger abnormal data recognition ability than machine learning in many fields. The application of abnormal data detection in high-voltage line operation can provide more accurate and timely fault diagnosis and prediction to improve the reliability and safety of the line system. It is a hot spot in the current research field of high-voltage line operation.

[0005] Deep learning has gradually become the main method for anomaly detection. Abnormal conditions rarely occur in my country's high-voltage lines, resulting in the fact that most of the data obtained from high-voltage line monitoring are normal data. Therefore, if these monitoring data are directly used to train and build a line abnormal data detection model using a deep learning algorithm, it may lead to problems of overfitting or underfitting of the model, which is not practical. At present, there are still a series of problems in several key aspects of abnormal data detection methods, including but not limited to low accuracy, difficulty in adapting to complex data distribution, and inaccurate feature extraction. The cumulative effect of these problems leads to a relatively high overall false alarm rate, especially in the process of judging abnormal data. This problem is particularly significant. In view of the current situation, it is urgent to find more feasible and effective methods to improve the overall performance of the abnormal data detection system. The urgency of this demand lies in the fact that the shortcomings of the current methods may pose a potential threat to the safe operation of high-voltage lines. In order to address the above problems, it is urgent to study and promote more advanced and reliable abnormal data detection methods. Summary of the invention

[0006] In order to cope with the challenges in detecting abnormal data in the operation of high-voltage lines, a powerful deep learning model is provided to detect abnormal data in the operation of high-voltage lines. Especially in the case of class imbalance and high noise, it can provide more accurate and reliable identification and detection. It can automatically detect and classify abnormal data generated in the operation of high-voltage lines, detect abnormal data and issue timely alarms, and reduce the safety risks of high-voltage lines.

[0007] The purpose of the present invention is to provide a method for detecting abnormal data of high-voltage line operation, comprising the following steps:

[0008] Step 1: Collect running data flow in real time, perform data preprocessing, and then go to step 2;

[0009] Step 2: Send the data processed in step 1 to the time series prediction model, and then go to step 3;

[0010] Step 3: Generate a comparison traffic change baseline using the time series prediction model, and then go to step 4;

[0011] Step 4: Determine whether the data flow is normal; if so, go to step 5; if not, the high-voltage line operates normally;

[0012] Step 5: Train the RIME-CNN-LSTM-Attention model, then go to step 6;

[0013] Step 6: Detect abnormal data flow in the line operation, and then go to step 7;

[0014] Step 7: Determine whether to associate traffic warning; if so, go to step 8; if not, go to step 9;

[0015] Step 8: Generate abnormal data associated warning information, and then go to step 10 and step 11 at the same time;

[0016] Step 9: Generate warning information of abnormal data with unknown causes, and then go to step 10;

[0017] Step 10: Analysis and diagnosis of unknown abnormal data;

[0018] Step 11: Trigger emergency measures for abnormal line data.

[0019] The construction process of the RIME-CNN-LSTM-Attention model in step 5 specifically includes the following steps:

[0020] Step 51: Process the data set through the Attention layer, re-evaluate the input features, and calculate new weights;

[0021] Step 52: Pass the extracted data into the CNN layer;

[0022] Step 53: Pass the output of the CNN layer to the LSTM layer;

[0023] Step 54: Use the Frost optimization algorithm to optimize the hyperparameters of the LSTM layer;

[0024] Step 55: Classify the model prediction results through the softmax classifier to obtain the prediction results of the high-voltage line abnormal data model for the operation data detection.

[0025] CNN layers include input layer, convolution layer, pooling layer and fully connected layer.

[0026] The activation function of the convolutional layer is the relu function.

[0027] When training the RIME-CNN-LSTM-Attention model in step 5, the loss function selected is Cross-entropy.

[0028] A detection system for a high-voltage line operation abnormal data detection method includes a high-voltage line operation status monitoring module, a high-voltage line operation flow monitoring module, a high-voltage line operation data flow real-time analysis module, and a high-voltage line operation abnormal data flow monitoring module, and the above modules are connected in sequence.

[0029] The high-voltage line operation status monitoring module establishes a high-voltage line operation data acquisition channel to monitor the status changes of the high-voltage line.

[0030] The high-voltage line operation flow monitoring module establishes a high-voltage line operation data monitoring channel to monitor the changes in high-voltage line operation data flow.

[0031] The high-voltage line operation data flow real-time analysis module collects high-voltage line operation data flow information and determines abnormal data flow according to the flow warning threshold configured by the system.

[0032] The high-voltage line operation abnormal data flow monitoring module analyzes the abnormal data and generates real-time high-voltage line operation abnormal data flow warning and triggers corresponding high-voltage line abnormal data emergency measures, while synchronizing the high-voltage line operation data flow information to the big data analysis platform.

[0033] The beneficial effects of the present invention are:

[0034] 1. Using the frost optimization algorithm to optimize the hyperparameters of the improved high-voltage line anomaly detection model can effectively improve the detection accuracy of the high-voltage line operation abnormal data detection model.

[0035] 2. The present invention can automatically monitor the changes in line operation data flow and conduct abnormal analysis, which improves the efficiency of fault detection and analysis compared with the traditional manual processing of abnormal data flow risk warning. This method significantly reduces manual intervention and operation, improves the safety and stability of high-voltage line operation, and reduces the operating cost and maintenance difficulty of high-voltage lines.

[0036] 3. The method of combining CNN and LSTM can effectively improve the accuracy and robustness of abnormal data detection of high-voltage line operation.

[0037] 4. This method uses a softmax classifier to process data and can solve the binary classification problem. RIME-CNN-LSTM-Attention combines the advantages of convolution and long short-term memory networks to prevent gradient vanishing and gradient explosion problems during model training, and can better capture the timing relationship when processing network traffic, thereby realizing real-time monitoring of abnormal data flows of high-voltage line operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a flow chart of the method for detecting abnormal data of high-voltage line operation of the present invention;

[0039] Figure 2 It is a module schematic diagram of the high-voltage line operation abnormal data detection system of the present invention;

[0040] Figure 3 A schematic diagram of a model of a method for detecting abnormal data of high-voltage line operation according to the present invention;

[0041] Figure 4 This is a specific structural diagram of the RIME-CNN-LSTM-Attention model of the present invention. DETAILED DESCRIPTION

[0042] The present invention provides a method for detecting abnormal data of high-voltage line operation, which will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0043] Figure 1 FIG. 1 is a flow chart of a method for detecting abnormal data of high voltage line operation according to the present invention. Figure 1 As shown, the method includes: real-time collection of operating data flow and data preprocessing operations. The processed data is sent to the time series prediction model to generate a baseline for comparison of flow changes. It is determined whether the detected data is abnormal. If the data is abnormal, the RIME-CNN-LSTM-Attention model is trained to detect abnormal data flow through the model. The correlation warning judgment of abnormal data flow is performed in the system, and whether the abnormal data flow meets the preset warning conditions is checked to determine whether further processing is required. If the abnormal data flow matches the system preset warning conditions, the system will generate corresponding abnormal data correlation warning information. For those abnormal data flows that cannot be associated with known conditions, the system will generate unknown cause abnormal data warning information to provide warnings for possible new problems. In order to deeply understand the characteristics and causes of abnormal data flow, the system starts the high-voltage line operation abnormal data analysis and diagnosis tool to provide detailed flow analysis. According to the correlation warning of abnormal data flow and the analysis results of abnormal data flow of high-voltage lines, the system triggers the emergency measures for abnormal data of the line. If no abnormal data flow is detected during monitoring, the operation status of the high-voltage line is normal.

[0044] By establishing a high-voltage line operation data collection channel, the system can monitor the changes in operation data flow in real time, and collect the current line operation data flow information in a timely manner, providing strong data support for the performance evaluation of abnormal data detection and analysis models. The system monitors the operation data flow warning threshold in real time to ensure that it can timely determine whether there is abnormal data flow in the line to prevent potential problems. Once the system detects abnormal data flow, it immediately generates a real-time abnormal data flow warning and initiates emergency measures for abnormal line data, and takes corresponding measures to respond to abnormal situations quickly and effectively to maintain the stability of the high-voltage line. The collected high-voltage line operation data flow information is sent to the big data analysis platform in real time and synchronously, providing data support for further in-depth analysis and processing, ensuring that the system has a comprehensive understanding of the high-voltage line status when dealing with abnormal situations.

[0045] By timely collecting the data flow of high-voltage line operation, the system can continuously monitor various information in the data transmission process to obtain a comprehensive high-voltage line operation status. The system performs complex calculations to evaluate the real-time data flow of each high-voltage line operation. The system is configured with an abnormal data flow warning threshold, which is a pre-set warning line. When the data flow on the line approaches or exceeds this threshold, the system will take corresponding measures to avoid potential risks and crises. Based on real-time monitoring and abnormal data flow warning thresholds, the system adopts corresponding abnormal data flow warning mechanisms to ensure the normal operation of high-voltage lines.

[0046] Using the time series prediction model RIME-CNN-LSTM-Attention trained on the big data analysis platform, the system conducts comprehensive tracking and analysis of abnormal traffic. The model generates a predicted data traffic change baseline for in-depth understanding of the abnormal operation data traffic. By comparing the predicted baseline with the actual data traffic currently collected in detail, the system deeply analyzes the difference in abnormal traffic. In the abnormal traffic tracking analysis, when the difference between the actual traffic and the predicted baseline is less than the preset threshold, the line operation status is normal. If the difference exceeds the preset threshold, it is regarded as abnormal data, and actively attempts to associate it with the warning event of abnormal data of high-voltage line operation, so as to better understand the nature of the problem. In the face of warning events that cannot be directly associated with abnormal data of high-voltage line operation, the high-voltage line operation unknown abnormal data analysis and diagnosis tool is quickly launched. By adopting a comprehensive approach, it is possible to fully understand and respond to abnormal traffic generated by various high-voltage lines.

[0047] Figure 2 This is a schematic diagram of a module of a high-voltage line operation abnormal data detection system according to the present invention; the module includes:

[0048] The high-voltage line operation status monitoring module 201 is responsible for establishing a high-voltage line operation data acquisition channel and monitoring the state changes of the high-voltage line.

[0049] The high-voltage line operation flow monitoring module 202 establishes a high-voltage line operation data monitoring channel to monitor the changes in the high-voltage line operation data flow.

[0050] The high-voltage line operation data flow real-time analysis module 203 collects high-voltage line operation data flow information and determines abnormal data flow according to the flow warning threshold configured by the system.

[0051] The high-voltage line operation abnormal data flow monitoring module 204 analyzes the abnormal data and generates a real-time high-voltage line operation abnormal data flow warning and triggers corresponding high-voltage line abnormal data emergency measures, while synchronizing the high-voltage line operation data flow information to the big data analysis platform.

[0052] The specific steps include: according to the configured abnormal data flow warning threshold, the system monitors the high-voltage line operation data flow in real time, and immediately warns and takes emergency measures when anomalies are found. Collect data and use the CNN-LSTM model for in-depth analysis to generate a flow baseline and compare it with the actual flow to detect anomalies. Once an anomaly is detected, an early warning is generated and emergency measures are triggered. The data is sent to the big data platform for analysis, and the model is trained using abnormal data to improve prediction accuracy. The process includes steps such as real-time monitoring, early warning generation, triggering emergency measures, data collection, analysis, and model training, including training the RIME-CNN-LSTM-Attention model using network traffic sensitive data. The specific steps are:

[0053] Step 1: Obtain and preprocess the original dataset;

[0054] Step 2: Process the dataset through the Attention layer, re-evaluate the input features, and calculate new weights to optimize the effective feature space and channel information while suppressing useless feature space and channel information;

[0055] Step 3: Perform model training, build a RIME-CNN-LSTM-Attention network model to extract network traffic features, and perform abnormal traffic detection through a softmax classifier;

[0056] Step 4: By evaluating the abnormal data flow detection results, the performance of the abnormal data detection model in the task can be comprehensively evaluated, including indicators such as accuracy, precision, F1 score and recall rate. These indicators comprehensively consider the performance of the model in different aspects, which helps to gain a deeper understanding of the effect of the model and its overall performance in the abnormal data detection task;

[0057] Step 5: During the training phase, Cross-entropy was used as the loss function and Adam was selected as the optimization algorithm;

[0058] Step 6: By testing the model on the test set, the performance of the model can be evaluated.

[0059] Figure 3 The model schematic diagram of the abnormal data detection method for high-voltage line operation of the present invention; with the help of the RIME-CNN-LSTM-Attention model construction method, the method cleverly integrates the convolutional neural network and the long short-term memory network to achieve comprehensive capture of network traffic characteristics and timing information. In order to enhance the ability to extract traffic features, the method introduces an attention layer at the front end of CNN. The Attention layer weights the key features in the network so that the model can focus on important information more specifically, thereby improving the efficiency and accuracy of feature extraction. Using the LSTM network, this stage focuses on capturing the timing characteristics of traffic, more comprehensively understanding the changing laws of traffic, and helping the model to better adapt to complex data traffic dynamics. The frost optimization algorithm optimizes the hyperparameters of the LSTM model to improve its detection accuracy. The extracted features are classified by the softmax classifier, and the results of abnormal data traffic detection are finally obtained. This classification stage enables the model to map the learned features to the corresponding categories, thereby achieving effective classification and identification of abnormal traffic. The specific steps are as follows:

[0060] Step 1: Process the dataset through the Attention layer, re-evaluate the input features, and calculate new weights to enhance valid information and suppress useless information.

[0061] Step 2: The extracted data is passed into a convolutional neural network (CNN), whose convolutional structure helps to mine network traffic features.

[0062] Step 3: CNN adopts a hierarchical structure, including an input layer, a convolution layer, a pooling layer, and a fully connected layer. In the convolution layer of CNN, a one-dimensional convolution operation is used to extract key features from the input data through the convolution operation. The pooling layer is introduced in the CNN structure, and its main function is to reduce the size of the feature map while retaining important feature information, thereby reducing the amount of calculation and improving the efficiency of the model. The fully connected layer maps the features extracted in the convolution and pooling stages to the output layer to achieve global integration of features. In the convolution layer, the relu activation function is selected in this embodiment. This nonlinear activation function introduces nonlinear characteristics, which helps the network to better capture and represent the complex relationships of the data and improve the network's expression and fitting capabilities.

[0063] Step 4: Pass the output of the CNN layer to the LSTM layer. Since traffic information has the sequential characteristics of the previous and next correlations, the LSTM network can extract long-term temporal features.

[0064] Step 5: Use the Frost optimization algorithm to optimize the hyperparameters of the LSTM network to improve the detection accuracy of the model.

[0065] Step 6: Classify the model prediction results by applying the softmax classifier to obtain the prediction results of the high-voltage line abnormal data model for the operation data detection. The RIME-CNN-LSTM-Attention model structure is as follows: Figure 4 shown.

[0066] The classification results of the model are used to predict the data traffic and the accuracy of the model prediction is calculated. The calculation formula is as follows:

[0067]

[0068] The classification results of the model are used to predict the data traffic and calculate the accuracy of the model prediction. The calculation formula is as follows:

[0069]

[0070] The classification results of the model are used to predict the data flow and the recall rate of the model prediction is calculated. The calculation formula is as follows:

[0071]

[0072] The classification results of the model are used to predict the data traffic and the F1 score of the model prediction is calculated. The calculation formula is as follows:

[0073]

[0074] In the above scenario, TP (True Positive) and TN (True Negative) represent the correct predictions of the model, that is, the successfully classified instances. In contrast, FP (False Positive) and FN (False Negative) represent the misclassification events of the model, that is, the instances are incorrectly judged as positive or negative.

[0075] In the detection and analysis of abnormal data of high-voltage line operation, a predicted baseline of data flow changes is generated using a trained time series prediction model. The system automatically compares the difference between the actual data flow and the predicted baseline. If the difference is less than the set threshold, it is classified as normal operation of the high-voltage line; otherwise, the system determines it as abnormal operation of the high-voltage line and attempts to initiate emergency measures for abnormal operation of the high-voltage line. For those warning events that cannot be directly associated with abnormal line data, unknown abnormal data analysis and diagnosis are introduced.

[0076] The abnormal data anomaly detection and analysis method is applicable to high-voltage line operation scenarios and can effectively detect and identify abnormal data generated during high-voltage line operation. When the high-voltage line operation is abnormal, the generated warning can be responded to quickly. In addition, the generated high-voltage line operation fault warning provides more detailed abnormal data information to the downstream system to select a more appropriate processing strategy.

[0077] In summary, the present invention introduces convolutional neural networks and attention mechanisms, which are helpful to solve the noise problem in high-voltage line data, reduce the interference of noise on abnormal flow detection, more effectively extract key features, and improve prediction accuracy. Through the long short-term memory network, the time series feature extraction of high-voltage line operation data is realized, so that the model can better understand the time evolution of high-voltage line data and further improve the detection performance. The data features of the high-voltage line are classified by the softmax classifier to accurately identify normal data and different types of abnormal data, which helps to more accurately detect potential risks and crises in the high-voltage line. In the LSTM neural network model, the setting of parameters has a crucial influence on the recognition accuracy of the model, and the existing hyperparameter selection generally adopts an empirical method. The empirical method is arbitrary and blind in parameter selection, has no universality, and its effect is unstable. The hyperparameters of the improved high-voltage line anomaly detection model are optimized using the frost optimization algorithm, which can effectively improve the detection accuracy of the high-voltage line operation abnormal data detection model.

Claims

1. A method for detecting abnormal data of high-voltage line operation, characterized in that: The following steps are involved: Step 1: Collect running data flow in real time, perform data preprocessing, and then go to step 2; Step 2: Send the data processed in step 1 to the time series prediction model, and then go to step 3; Step 3: Generate a comparison traffic change baseline using the time series prediction model, and then go to step 4; Step 4: Determine whether the data flow is normal; if so, go to step 5; if not, the high-voltage line operates normally; Step 5: Train the RIME-CNN-LSTM-Attention model, then go to step 6; Step 6: Detect abnormal data flow in the line operation, and then go to step 7; Step 7: Determine whether to associate traffic warning; if so, go to step 8; if not, go to step 9; Step 8: Generate abnormal data associated warning information, and then go to step 10 and step 11 at the same time; Step 9: Generate warning information of abnormal data with unknown causes, and then go to step 10; Step 10: Analysis and diagnosis of unknown abnormal data; Step 11: Trigger emergency measures for abnormal line data.

2. The method for detecting abnormal data of high-voltage line operation according to claim 1, characterized in that: The construction process of the RIME-CNN-LSTM-Attention model in step 5 specifically includes the following steps: Step 51: Process the data set through the Attention layer, re-evaluate the input features, and calculate new weights; Step 52: Pass the extracted data into the CNN layer; Step 53: Pass the output of the CNN layer to the LSTM layer; Step 54: Use the Frost optimization algorithm to optimize the hyperparameters of the LSTM layer; Step 55: Classify the model prediction results through the softmax classifier to obtain the prediction results of the high-voltage line abnormal data model for the operation data detection.

3. The method for detecting abnormal data of high-voltage line operation according to claim 2, characterized in that: The CNN layer includes an input layer, a convolutional layer, a pooling layer and a fully connected layer.

4. The method for detecting abnormal data of high-voltage line operation according to claim 3, characterized in that: The activation function of the convolutional layer is the relu function.

5. The method for detecting abnormal data of high-voltage line operation according to claim 2, characterized in that: When training the RIME-CNN-LSTM-Attention model in step 5, the selected loss function is Cross-entropy.

6. A detection system for the high-voltage line operation abnormal data detection method according to any one of claims 1 to 5, characterized in that: It includes a high-voltage line operation status monitoring module, a high-voltage line operation flow monitoring module, a high-voltage line operation data flow real-time analysis module, and a high-voltage line operation abnormal data flow monitoring module, and the above modules are connected in sequence.

7. The detection system of the high-voltage line operation abnormal data detection method according to claim 6 is characterized in that: The high-voltage line operation status monitoring module establishes a high-voltage line operation data acquisition channel to monitor the state changes of the high-voltage line.

8. The detection system of the high-voltage line operation abnormal data detection method according to claim 6 is characterized in that: The high-voltage line operation flow monitoring module establishes a high-voltage line operation data monitoring channel to monitor changes in high-voltage line operation data flow.

9. The detection system of the high-voltage line operation abnormal data detection method according to claim 6 is characterized in that: The high-voltage line operation data flow real-time analysis module collects high-voltage line operation data flow information and determines abnormal data flow according to the flow warning threshold configured by the system.

10. The detection system of the high-voltage line operation abnormal data detection method according to claim 6 is characterized in that: The high-voltage line operation abnormal data flow monitoring module analyzes the abnormal data and generates a real-time high-voltage line operation abnormal data flow warning and triggers corresponding high-voltage line abnormal data emergency measures, and at the same time synchronizes the high-voltage line operation data flow information to the big data analysis platform.

Citation Information

Patent Citations

  • High-voltage cable line operation state prediction method based on long-short-term memory network

    CN112070322A

  • Network traffic anomaly detection and analysis method and device

    CN114124492A

  • Attention-CNN-LSTM industrial process fault diagnosis method based on improved grey wolf algorithm optimization

    CN117421667A

  • Short-term power load prediction method

    CN118174294A

  • Power load prediction method based on decomposition strategy and deep learning

    CN118211844A