Intelligent operation and maintenance system of power grid system based on artificial intelligence
By introducing artificial intelligence technology into the power grid system, data is collected and analyzed in real time, fault prediction and automated operation and maintenance are solved, and the problems of low data collection and processing efficiency and insufficient fault monitoring and recovery capabilities of traditional power grid systems are improved. Operating efficiency and stability reduce operation and maintenance costs.
Patent Information
- Application Number
- CN202410901755.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-05
- Publication Date
- 2024-11-08
AI Technical Summary
Traditional power grid systems have low data collection and processing efficiency, inability to monitor and predict faults in real time, insufficient fault isolation and recovery capabilities, and high operation and maintenance costs, resulting in low grid operation efficiency, poor stability and high maintenance costs.
Adopt an intelligent operation and maintenance system of the power grid system based on artificial intelligence, collects grid data in real time through sensors, uses deep learning algorithms to perform data analysis and fault prediction, automatically triggers fault isolation and recovery mechanisms, dynamically adjusts grid operating parameters, and realizes intelligent operation of the power grid. dimension.
It improves the operation and maintenance efficiency and reliability of the power grid system, reduces the operation and maintenance costs, realizes real-time monitoring and fault warning of the power grid, and ensures the stable operation and power supply reliability of the power grid.
Smart Images

Figure CN118920692A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of artificial intelligence technology, and in particular relates to an intelligent operation and maintenance system for a power grid system based on artificial intelligence. Background Art
[0002] With the development of economy and population growth, the load of power grid is increasing, and the traditional power system operation and maintenance methods can no longer meet the complexity and efficiency requirements of modern power grids. Therefore, the intelligent operation and maintenance system of power grid system based on artificial intelligence has emerged and has become an important development direction of modern power grid operation and maintenance.
[0003] Technical problems existing in existing technologies:
[0004] 1. Inefficient data collection and processing:
[0005] Data collection limitations: Traditional power grid systems rely on manual monitoring and decentralized sensors for data collection, resulting in limited data collection coverage and low collection frequency, making it impossible to comprehensively and real-time monitor the operating status of each node in the power grid.
[0006] Insufficient data processing capabilities: Due to the huge and complex amount of grid operation data, traditional data processing methods are difficult to efficiently clean, store and analyze massive amounts of data. The data processing efficiency is low and it is impossible to provide accurate feedback on the grid operation status in a timely manner.
[0007] 2. Unable to monitor and predict potential failures in real time:
[0008] Untimely monitoring: The monitoring methods of traditional power grid systems mainly rely on regular inspections and emergency event reports. It is difficult to achieve real-time monitoring and rapid response to the operating status of the power grid, and can often only be handled passively after a fault occurs.
[0009] Lack of intelligent prediction: The existing system lacks advanced fault prediction methods and cannot use historical data and real-time data to identify and predict potential failure modes, resulting in the inability to provide early warning and prevent failures.
[0010] 3. Insufficient fault isolation and recovery capabilities:
[0011] Fault isolation lag: When a fault occurs, traditional power grid systems require manual intervention to troubleshoot and isolate the fault, which has a slow response speed and can easily lead to the expansion of the fault scope and affect power supply stability.
[0012] Single recovery method: The existing system has limited means for fault recovery and often relies on a single backup line or manual operation to restore power supply. The recovery efficiency is low and it is difficult to ensure the continuous operation of the power grid.
[0013] 4. High operation and maintenance costs:
[0014] High dependence on manual labor: The traditional power grid system operation and maintenance method is highly dependent on manual operation and monitoring. Operation and maintenance personnel need to conduct frequent inspections and manually adjust grid parameters, resulting in large consumption of human resources and high operation and maintenance costs.
[0015] Low equipment utilization: Due to the lack of intelligent operation and maintenance methods, the operating status and fault conditions of power grid equipment are difficult to grasp in a timely manner, resulting in low equipment utilization and high maintenance and replacement costs.
[0016] In summary, existing technologies in power grid operation and maintenance have problems such as low efficiency in data collection and processing, inability to monitor and predict faults in real time, insufficient fault isolation and recovery capabilities, and high operation and maintenance costs. These problems restrict the stability and efficiency of the power grid system and urgently need to be improved and optimized by introducing artificial intelligence technology. Summary of the invention
[0017] In view of the problems existing in the prior art, the present invention provides an intelligent operation and maintenance system for a power grid system based on artificial intelligence.
[0018] The present invention is implemented as follows: an artificial intelligence-based intelligent operation and maintenance method for a power grid system comprises the following steps:
[0019] Data collection: Using sensors deployed in substations, transmission lines, and distribution equipment, the voltage, current, power, temperature and other operating data of each node in the power grid are collected in real time; smart meters installed at the user end are used to collect information on power consumption and power consumption time periods; and environmental monitoring equipment is used to collect external environmental data, such as temperature, humidity, wind speed, etc.
[0020] Data transmission: The collected data is transmitted to the central processing unit in real time through wireless transmission technology and Internet of Things technology;
[0021] Data processing and analysis: The central processing unit receives data transmitted by the data acquisition module and sends it to the data processing and analysis module. The data processing and analysis module uses big data technology to clean, store and pre-process the massive data collected, and uses deep learning algorithms to extract features and recognize patterns from the processed data to analyze the operation status of the power grid.
[0022] Fault prediction: Build a fault prediction model based on historical data and real-time data, and identify potential fault modes by training the model; when the system detects a potential fault, the prediction model outputs a fault warning signal;
[0023] Fault isolation and recovery: Based on the fault prediction results, the fault isolation mechanism is automatically triggered to isolate the fault area through circuit breakers and reclosers to prevent the fault from spreading. At the same time, the system starts the fault recovery mechanism and tries to restore power supply through backup lines or distributed power sources to ensure the continuous operation of the power grid.
[0024] Automation control: Integrate the SCADA system to realize remote monitoring and control of power grid equipment; use automation control algorithms to dynamically adjust power grid operating parameters and optimize power grid operating status.
[0025] Further, the following steps are included:
[0026] 1. Sensor data processing: collect real-time data such as voltage, current, power, and temperature from sensors deployed on substations, transmission lines, and distribution equipment, use smart meters to collect information on user-side power consumption and power consumption time periods, and collect external environmental data such as temperature, humidity, and wind speed from environmental monitoring equipment;
[0027] 2. Data cleaning and preprocessing: The data processing and analysis module cleans the massive data collected, including removing noise and abnormal data, filling missing values, etc.; the cleaned data is preprocessed, including data normalization, dimensionality reduction and other operations;
[0028] 3. Feature extraction and pattern recognition: Use deep learning algorithms (such as convolutional neural networks and recursive neural networks) to extract features from preprocessed data, identify patterns and abnormal conditions in power grid operation, and evaluate the power grid operation status by analyzing characteristic values such as voltage, current, and power.
[0029] Further, the following steps are included:
[0030] 1. Fault prediction and early warning: The fault prediction module uses historical data and real-time data to build a fault prediction model and identify potential fault modes of the power grid by training the model. When the system detects a potential fault, the prediction model outputs a fault early warning signal and sends the early warning information to the fault isolation and recovery module.
[0031] 2. Fault isolation and recovery: The fault isolation and recovery module automatically triggers the fault isolation mechanism based on the fault prediction results, isolates the fault area through circuit breakers and reclosers to prevent the fault from spreading; at the same time, it starts the fault recovery mechanism to restore power supply through backup lines or distributed power sources to ensure the continuous operation of the power grid;
[0032] 3. Operation parameter optimization: The automation control module uses the SCADA system to realize remote monitoring and control of power grid equipment, applies automation control algorithms, dynamically adjusts power grid operation parameters such as voltage, current, power, etc., optimizes the power grid operation status, and improves the power grid operation efficiency and stability.
[0033] Furthermore, the data processing and analysis module includes:
[0034] Big data technology: used to clean, store and pre-process the massive amounts of data collected to ensure the integrity and accuracy of the data;
[0035] Deep learning model: Convolutional neural network (CNN) and recurrent neural network (RNN) are used to perform feature extraction and pattern recognition on the processed data. CNN is used to extract the spatial features of data such as voltage and current, and RNN is used to capture the temporal features of time series data. Through the comprehensive analysis of these features, the operation status of the power grid can be accurately evaluated and abnormal situations can be identified.
[0036] Further, the fault prediction module includes:
[0037] Fault prediction model: Use historical data and real-time data to build a fault prediction model based on deep learning. By training the model to identify potential fault modes, a long short-term memory network (LSTM) or a bidirectional long short-term memory network (BiLSTM) is used to process power grid operation data to predict faults.
[0038] Fault warning system: When the fault prediction model detects a potential fault, it outputs a fault warning signal and triggers the automated processing flow of the fault isolation and recovery module based on the warning information. It isolates the faulty area through circuit breakers and reclosers, and attempts to restore power supply through backup lines or distributed power sources to ensure the stable operation of the power grid.
[0039] The present invention provides an intelligent operation and maintenance system for a power grid system based on artificial intelligence, the system comprising:
[0040] The data acquisition module uses various sensors, smart meters and environmental monitoring equipment to collect the voltage, current, frequency, temperature and other operating data of each node of the power grid in real time; the collected data is transmitted to the central processing unit through wireless transmission technology and Internet of Things technology;
[0041] The central processing unit is connected to the data acquisition module and is used to receive data transmitted by the data acquisition module and send it to the data processing and analysis module;
[0042] The data processing and analysis module is connected to the central processing unit and uses big data technology to clean, store and pre-process the massive amounts of data collected. It uses deep learning algorithms (such as convolutional neural networks and recursive neural networks) to extract features and recognize patterns from the processed data and analyze the operation status of the power grid.
[0043] The fault prediction module is connected with the data processing and analysis module to build a fault prediction model based on historical data and real-time data, and identify potential fault modes through training models; when the system detects a potential fault, the prediction model outputs a fault warning signal;
[0044] The fault isolation and recovery module is connected to the fault prediction module. According to the fault prediction results, it automatically triggers the fault isolation mechanism and isolates the fault area through circuit breakers and reclosers to prevent the fault from spreading. At the same time, the system starts the fault recovery mechanism and tries to restore power supply through backup lines or distributed power sources to ensure the continuous operation of the power grid.
[0045] The automation control module is connected to the central processing unit and integrated with the SCADA system to achieve remote monitoring and control of power grid equipment; it uses automation control algorithms to dynamically adjust power grid operating parameters and optimize power grid operating status.
[0046] Furthermore, the sensors are deployed in substations, transmission lines, and distribution equipment to collect real-time data such as voltage, current, power, and temperature; the smart meters are installed at the user end to collect information such as power consumption and power consumption time periods; the environmental monitoring equipment collects external environmental data such as temperature, humidity, wind speed, etc. to help evaluate the impact on power grid operation.
[0047] Furthermore, the wireless transmission technology uses wireless technologies such as 5G, WiFi, LoRa, NBIoT, etc. to transmit data to a data center; some data is preprocessed before data transmission to reduce network bandwidth requirements and improve response speed.
[0048] Furthermore, the data processing and analysis module uses big data technology to clean, store and pre-process the collected massive data, specifically including:
[0049] Data cleaning: remove noise and outliers, fill in missing data, and ensure data quality;
[0050] Data integration: Standardize data from different sources to form a unified data format.
[0051] Furthermore, the feature extraction utilizes deep learning techniques such as autoencoders to extract important features from raw data; feature selection algorithms (such as LASSO regression and recursive feature elimination (RFE)) are used to select key features, reduce data dimensions, and improve model efficiency.
[0052] Furthermore, the fault prediction module constructs a fault prediction model based on historical data and real-time data, specifically including:
[0053] S1: Convolutional Neural Network (CNN): used to process data with strong temporal and spatial correlation, such as temperature and voltage distribution;
[0054] S2: Recurrent Neural Network (RNN): such as Long Short-Term Memory Network (LSTM), used to process time series data and predict future loads and potential failures;
[0055] S3: Generative Adversarial Network (GAN): used to generate data samples and improve the generalization ability of the model.
[0056] Further, identifying potential failure modes by training the model specifically includes:
[0057] Use historical data and labels for supervised learning and training deep learning models;
[0058] The cross-validation method was used to evaluate the model performance, and the model was optimized by adjusting the hyperparameters.
[0059] Furthermore, the fault isolation and recovery module automatically triggers the fault isolation mechanism according to the fault prediction result, isolates the fault area through circuit breakers and reclosers to prevent the fault from spreading; at the same time, the system starts the fault recovery mechanism and tries to restore power supply through backup lines or distributed power sources to ensure the continuous operation of the power grid, which specifically includes:
[0060] Real-time monitoring: The model monitors real-time data and identifies potential failures when abnormal patterns are detected;
[0061] Fault classification: Based on the model output, identify the fault type (such as short circuit, overload, equipment failure, etc.).
[0062] Fault isolation: Automation system: Use the SCADA (Supervisory Control and Data Acquisition) system to remotely control circuit breakers, disconnectors and other equipment to quickly isolate the fault area and prevent the fault from spreading.
[0063] Emergency Plan: The system has a built-in emergency plan that automatically executes predefined isolation strategies based on the fault type and location.
[0064] Backup line switching: The system automatically switches to backup lines or distributed power sources to restore power supply to the affected area;
[0065] Self-healing mechanism: Utilize self-healing control technology to automatically adjust grid operating parameters and restore normal operating conditions.
[0066] Furthermore, the automation control module integrates the SCADA system to realize remote monitoring and control of power grid equipment; and utilizes the automation control algorithm to dynamically adjust the power grid operation parameters and optimize the power grid operation status, specifically including:
[0067] Automated operation: Reduce manual intervention, improve fault response speed, and reduce the risk of human operational errors;
[0068] Intelligent decision support: Use artificial intelligence technology to provide decision support to help operation and maintenance personnel make accurate judgments and respond quickly;
[0069] Centralized monitoring: Establish a unified operation and maintenance management platform to monitor the grid operation status and fault information in real time;
[0070] Data visualization: Visualization tools are used to display grid operation data, fault detection results, and recovery status, so that operation and maintenance personnel can intuitively understand the system status;
[0071] Fault recording and analysis: The system automatically records detailed information of each fault for subsequent analysis and improvement;
[0072] Continuous learning and optimization: Based on fault records and recovery situations, the fault prediction model and recovery strategy are continuously optimized to improve the intelligence level of the system.
[0073] Another object of the present invention is to provide an artificial intelligence-based power grid system intelligent operation and maintenance method for implementing the artificial intelligence-based power grid system intelligent operation and maintenance system, the method comprising:
[0074] S21: Through the data acquisition module, various sensors, smart meters and environmental monitoring equipment are used to collect the voltage, current, frequency, temperature and other operating data of each node of the power grid in real time; the collected data is transmitted to the central processing unit through wireless transmission technology and Internet of Things technology; the data transmitted by the data acquisition module is received by the central processing unit and sent to the data processing and analysis module;
[0075] S22: Through the data processing and analysis module, use big data technology to clean, store and pre-process the massive data collected; use deep learning algorithms (such as convolutional neural networks and recursive neural networks) to extract features and recognize patterns on the processed data to analyze the operation status of the power grid;
[0076] S23: Through the fault prediction module, a fault prediction model based on historical data and real-time data is constructed, and potential fault modes are identified through training models; when the system detects a potential fault, the prediction model outputs a fault warning signal;
[0077] S24: Through the fault isolation and recovery module, according to the fault prediction results, the fault isolation mechanism is automatically triggered, and the fault area is isolated through circuit breakers and reclosers to prevent the fault from spreading; at the same time, the system starts the fault recovery mechanism and tries to restore power supply through backup lines or distributed power sources to ensure the continuous operation of the power grid;
[0078] S25: Through the automation control module, the SCADA system is integrated to realize remote monitoring and control of power grid equipment; the automation control algorithm is used to dynamically adjust the power grid operation parameters and optimize the power grid operation status.
[0079] In combination with the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solutions to be protected by the present invention are as follows:
[0080] First, the present invention first collects grid operation data in real time, including power supply load, voltage, temperature and other information. The collected data is analyzed and processed using a deep learning algorithm to monitor the operation status of the grid in real time and predict potential faults. When the system detects a potential fault, it automatically triggers the fault isolation mechanism to prevent the fault from spreading. At the same time, the system will also try to automatically restore the fault to ensure the stable operation of the grid. Improve operation and maintenance efficiency and reliability: The entire process is highly automated and intelligent, significantly improving the operation and maintenance efficiency and reliability of the grid system.
[0081] The present invention can efficiently collect and process power grid operation data, monitor and predict potential faults in real time, achieve fault isolation and recovery, and significantly improve the operation and maintenance efficiency and reliability of the power grid system.
[0082] Second, the present invention solves several key problems in the prior art, especially in the intelligent operation and maintenance of power grid systems. Traditional power grid operation and maintenance methods rely on manual monitoring and manual control, which are unable to collect and analyze a large amount of operating data in real time, and it is difficult to detect and handle potential faults in a timely manner. This leads to low operating efficiency of the power grid, high maintenance costs, and difficulty in ensuring the continuity and stability of power supply. In addition, traditional methods lack intelligent means in fault prediction and recovery, and can often only take passive response measures after a fault occurs, and cannot provide early warning and active processing.
[0083] The present invention introduces artificial intelligence technology, uses big data and deep learning algorithms to collect, process and analyze power grid operation data in real time, significantly improving the intelligence level of the power grid system. The data acquisition module uses various sensors, smart meters and environmental monitoring equipment to collect the voltage, current, frequency, temperature and other operating data of each node of the power grid in real time, and transmits the data to the central processing unit through wireless transmission technology and Internet of Things technology. The central processing unit receives and processes this data, and uses big data technology to clean, store and preprocess the massive data to ensure the integrity and accuracy of the data.
[0084] Through deep learning algorithms, the data processing and analysis module can extract features and recognize patterns from processed data, accurately analyze the operating status of the power grid and identify abnormal situations. The fault prediction module builds a fault prediction model based on historical data and real-time data, identifies potential fault modes through training models, and outputs fault warning signals when potential faults are detected. The fault isolation and recovery module automatically triggers the fault isolation mechanism based on the fault prediction results, isolates the fault area through circuit breakers and reclosers, and prevents the fault from spreading. At the same time, the system starts the fault recovery mechanism and tries to restore power supply through backup lines or distributed power sources to ensure the continuous operation of the power grid.
[0085] In summary, the present invention has made significant technical progress in the intelligent operation and maintenance of power grid systems. Through the automation control module, the SCADA system is integrated to achieve remote monitoring and control of power grid equipment, and the automation control algorithm is used to dynamically adjust the power grid operation parameters and optimize the power grid operation status. Overall, the present invention not only improves the operation efficiency and stability of the power grid, but also reduces the operation and maintenance costs, improves the reliability and safety of power supply, and provides advanced technical support for power grid operation and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS
[0086] Figure 1 It is a structural diagram of an intelligent operation and maintenance system of a power grid system based on artificial intelligence provided by an embodiment of the present invention;
[0087] Figure 2 It is a flow chart of a fault prediction module provided by an embodiment of the present invention, and a method for constructing a fault prediction model based on historical data and real-time data;
[0088] Figure 3 It is a flow chart of an intelligent operation and maintenance method of a power grid system based on artificial intelligence provided by an embodiment of the present invention;
[0089] Figure 4 is a time trend graph provided by an embodiment of the present invention;
[0090] Figure 5 is a frequency statistics diagram of problems that occur in applications provided by an embodiment of the present invention;
[0091] Figure 6 It is a diagram showing that the GPR model provided by an embodiment of the present invention has the ability to output confidence information. DETAILED DESCRIPTION
[0092] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0093] like Figure 1 As shown, an embodiment of the present invention provides an intelligent operation and maintenance system for a power grid system based on artificial intelligence, the system comprising:
[0094] The data acquisition module uses various sensors, smart meters and environmental monitoring equipment to collect the voltage, current, frequency, temperature and other operating data of each node of the power grid in real time; the collected data is transmitted to the central processing unit through wireless transmission technology and Internet of Things technology;
[0095] The central processing unit is connected to the data acquisition module and is used to receive data transmitted by the data acquisition module and send it to the data processing and analysis module;
[0096] The data processing and analysis module is connected to the central processing unit and uses big data technology to clean, store and pre-process the massive amounts of data collected. It uses deep learning algorithms (such as convolutional neural networks and recursive neural networks) to extract features and recognize patterns from the processed data and analyze the operation status of the power grid.
[0097] The fault prediction module is connected with the data processing and analysis module to build a fault prediction model based on historical data and real-time data, and identify potential fault modes through training models; when the system detects a potential fault, the prediction model outputs a fault warning signal;
[0098] The fault isolation and recovery module is connected to the fault prediction module. According to the fault prediction results, it automatically triggers the fault isolation mechanism and isolates the fault area through circuit breakers and reclosers to prevent the fault from spreading. At the same time, the system starts the fault recovery mechanism and tries to restore power supply through backup lines or distributed power sources to ensure the continuous operation of the power grid.
[0099] The automation control module is connected to the central processing unit and integrated with the SCADA system to achieve remote monitoring and control of power grid equipment; it uses automation control algorithms to dynamically adjust power grid operating parameters and optimize power grid operating status.
[0100] The sensors are deployed in substations, transmission lines, and distribution equipment to collect real-time data such as voltage, current, power, and temperature. The smart meters are installed at the user end to collect information such as power consumption and power consumption time periods. The environmental monitoring equipment collects external environmental data such as temperature, humidity, wind speed, etc. to help evaluate the impact on power grid operation.
[0101] The wireless transmission technology uses wireless technologies such as 5G, WiFi, LoRa, NBIoT, etc. to transmit data to a data center; some data is preprocessed before data transmission to reduce network bandwidth requirements and improve response speed.
[0102] The data processing and analysis module uses big data technology to clean, store and pre-process the collected massive data, specifically including:
[0103] Data cleaning: remove noise and outliers, fill in missing data, and ensure data quality;
[0104] Data integration: Standardize data from different sources to form a unified data format.
[0105] The feature extraction is to extract important features from the original data using deep learning technologies such as autoencoders; feature selection algorithms (such as LASSO regression and recursive feature elimination RFE) are used to select key features, reduce data dimensions, and improve model efficiency.
[0106] Build an implementation plan for an AI-based intelligent operation and maintenance method for power grid systems, and provide a specific example for fault prediction and warning, fault isolation and recovery, and operation parameter optimization.
[0107] #1. System Architecture
[0108] The entire system consists of data acquisition layer, data transmission layer, data processing and analysis layer, fault prediction and early warning layer, fault isolation and recovery layer, and automation control layer.
[0109] #2. Data Collection Layer
[0110] *Use sensors deployed in substations, transmission lines, and distribution equipment to collect real-time operating data such as voltage, current, power, and temperature at each node of the power grid.
[0111] * Collect user-side electricity consumption and electricity consumption period information through smart meters.
[0112] *Use environmental monitoring equipment to collect external environmental data, such as temperature, humidity, wind speed, etc.
[0113] #3. Data transmission layer
[0114] *Through wireless transmission technology and Internet of Things technology, the collected data is transmitted to the central processing unit in real time.
[0115] #4. Data processing and analysis layer
[0116] *Use big data technology to clean, store and pre-process massive data.
[0117] * Use deep learning algorithms (such as convolutional neural networks and recurrent neural networks) to extract features and recognize patterns from processed data to analyze the operating status of the power grid.
[0118] #5. Fault prediction and warning layer (specific example)
[0119] * Fault prediction model construction: Based on historical data and real-time data, a fault prediction model is constructed using a long short-term memory network (LSTM). LSTM can process time series data and is suitable for power grid fault prediction scenarios.
[0120] *Model training: The LSTM model is trained using historical fault data and normal operation data to enable it to identify potential failure modes of the power grid.
[0121] * Fault warning: When the system detects a significant difference between real-time data and model prediction results, it considers that there is a potential fault and outputs a fault warning signal. The warning information includes the fault type, location, and expected impact range.
[0122] #6. Fault Isolation and Recovery Layer (Specific Examples)
[0123] *Fault isolation: When receiving a fault warning signal, the system automatically triggers the fault isolation mechanism. By controlling the opening and closing status of the circuit breaker and recloser, the fault area is isolated from the normal area to prevent the fault from spreading.
[0124] *Fault recovery: While isolating the fault, the system starts the fault recovery mechanism. First, the severity and impact of the fault area are assessed; then, a recovery strategy is selected based on the assessment results, such as restoring power supply through backup lines or distributed power sources; finally, the recovery strategy is executed to ensure the continuous operation of the power grid.
[0125] #7. Automation control layer (specific example)
[0126] *Remote monitoring and control: Integrate the SCADA system to realize remote monitoring and control of power grid equipment. Through the SCADA system, you can view the operating status, parameter settings and other information of power grid equipment in real time, and remotely control the switch status and parameter adjustment of the equipment.
[0127] *Operation parameter optimization: Utilize automated control algorithms (such as genetic algorithms, particle swarm optimization algorithms, etc.) to dynamically adjust grid operation parameters (such as voltage, current, power, etc.) based on real-time grid operation data and forecast results to optimize grid operation status. For example, during peak hours, the grid power supply capacity and voltage quality can be improved by adjusting the transformer tap position and switching capacitor banks; during off-peak hours, the grid loss and operation costs can be reduced by reducing equipment loads and shutting down some equipment.
[0128] Through the above implementation scheme, intelligent operation and maintenance management of the power grid system can be realized, the power grid operation efficiency and stability can be improved, and the failure rate and impact range can be reduced.
[0129] like Figure 2 As shown, the fault prediction module constructs a fault prediction model based on historical data and real-time data, specifically including:
[0130] S1: Convolutional Neural Network (CNN): used to process data with strong temporal and spatial correlation, such as temperature and voltage distribution;
[0131] S2: Recurrent Neural Network (RNN): such as Long Short-Term Memory Network (LSTM), used to process time series data and predict future loads and potential failures;
[0132] S3: Generative Adversarial Network (GAN): used to generate data samples and improve the generalization ability of the model.
[0133] The identifying of potential failure modes by training the model specifically includes:
[0134] Use historical data and labels for supervised learning and training deep learning models;
[0135] The cross-validation method was used to evaluate the model performance, and the model was optimized by adjusting the hyperparameters.
[0136] The fault isolation and recovery module automatically triggers the fault isolation mechanism according to the fault prediction result, isolates the fault area through circuit breakers and reclosers to prevent the fault from spreading; at the same time, the system starts the fault recovery mechanism and tries to restore power supply through backup lines or distributed power sources to ensure the continuous operation of the power grid, which specifically includes:
[0137] Real-time monitoring: The model monitors real-time data and identifies potential failures when abnormal patterns are detected;
[0138] Fault classification: Based on the model output, identify the fault type (such as short circuit, overload, equipment failure, etc.).
[0139] Fault isolation: Automation system: Use the SCADA (Supervisory Control and Data Acquisition) system to remotely control circuit breakers, disconnectors and other equipment to quickly isolate the fault area and prevent the fault from spreading.
[0140] Emergency Plan: The system has a built-in emergency plan that automatically executes predefined isolation strategies based on the fault type and location.
[0141] Backup line switching: The system automatically switches to backup lines or distributed power sources to restore power supply to the affected area;
[0142] Self-healing mechanism: Utilize self-healing control technology to automatically adjust grid operating parameters and restore normal operating conditions.
[0143] The automation control module integrates the SCADA system to realize remote monitoring and control of power grid equipment; uses the automation control algorithm to dynamically adjust the power grid operation parameters and optimize the power grid operation status, specifically including:
[0144] Automated operation: Reduce manual intervention, improve fault response speed, and reduce the risk of human operational errors;
[0145] Intelligent decision support: Use artificial intelligence technology to provide decision support to help operation and maintenance personnel make accurate judgments and respond quickly;
[0146] Centralized monitoring: Establish a unified operation and maintenance management platform to monitor the grid operation status and fault information in real time;
[0147] Data visualization: Visualization tools are used to display grid operation data, fault detection results, and recovery status, so that operation and maintenance personnel can intuitively understand the system status;
[0148] Fault recording and analysis: The system automatically records detailed information of each fault for subsequent analysis and improvement;
[0149] Continuous learning and optimization: Based on fault records and recovery situations, the fault prediction model and recovery strategy are continuously optimized to improve the intelligence level of the system.
[0150] like Figure 3 As shown, an embodiment of the present invention provides an artificial intelligence-based power grid system intelligent operation and maintenance method for implementing the artificial intelligence-based power grid system intelligent operation and maintenance system, the method comprising:
[0151] S21: Through the data acquisition module, various sensors, smart meters and environmental monitoring equipment are used to collect the voltage, current, frequency, temperature and other operating data of each node of the power grid in real time; the collected data is transmitted to the central processing unit through wireless transmission technology and Internet of Things technology; the data transmitted by the data acquisition module is received by the central processing unit and sent to the data processing and analysis module;
[0152] S22: Through the data processing and analysis module, use big data technology to clean, store and pre-process the massive data collected; use deep learning algorithms (such as convolutional neural networks and recursive neural networks) to extract features and recognize patterns on the processed data to analyze the operation status of the power grid;
[0153] S23: Through the fault prediction module, a fault prediction model based on historical data and real-time data is constructed, and potential fault modes are identified through training models; when the system detects a potential fault, the prediction model outputs a fault warning signal;
[0154] S24: Through the fault isolation and recovery module, according to the fault prediction results, the fault isolation mechanism is automatically triggered, and the fault area is isolated through circuit breakers and reclosers to prevent the fault from spreading; at the same time, the system starts the fault recovery mechanism and tries to restore power supply through backup lines or distributed power sources to ensure the continuous operation of the power grid;
[0155] S25: Through the automation control module, the SCADA system is integrated to realize remote monitoring and control of power grid equipment; the automation control algorithm is used to dynamically adjust the power grid operation parameters and optimize the power grid operation status.
[0156] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. A person of ordinary skill in the art will appreciate that the above-mentioned devices and methods can be implemented using computer executable instructions and / or contained in a processor control code, such as a carrier medium such as a disk, CD or DVDROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium. The device and its modules of the present invention can be implemented by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, and can also be implemented by a combination of the above-mentioned hardware circuits and software, such as firmware.
[0157] like Figure 4 As shown, a sequence number represents a build task, blue represents a successfully built task, and red represents a failed build task. The build time trend chart shows that after multiple continuous integration builds, the time it takes for the power grid system to complete the integrated deployment each time is getting shorter and shorter, and the number of successful continuous integration deployments is also increasing. This indicates that the power grid system adopts the artificial intelligence-based power grid system intelligent operation and maintenance system proposed by the present invention, and after a long period of frequent integration, the product quality has gradually become perfect, mature and stable.
[0158] In actual operation, the automatic operation and maintenance platform of the power control cloud ensures the convenience and efficiency of the operation and maintenance of the power control cloud. Figure 5 As shown in the figure, the frequency statistics of application problems in the power control cloud application within 3 months before and after the maintenance of this operation and maintenance platform are continuously recorded. It can be seen that by adopting this automated operation and maintenance platform, the operation and maintenance management mode has been transformed from passive to active, and the overall operation and maintenance level, quality and effect have been significantly improved. Therefore, the operation and maintenance platform can monitor the power control cloud application services in a targeted manner. Before the use of this platform, some applications on the control cloud can only be discovered after problems occur, and cannot be handled in time. After using this platform, some fault problems can be predicted in a targeted manner, and they can be handled in time before problems occur, ensuring the safety and stable operation of the system. After a long period of tracking and testing, it has been proved that the use of this operation and maintenance platform for CI / CD to release power products is practical and feasible in the power control cloud system, and after a series of steps such as integration and construction, the power products finally released have stable functions and reliable operation.
[0159] Figure 6This shows that the GPR model has the ability to output confidence information. Although existing point estimation machine learning (such as deep learning, linear regression, etc.) can achieve real-time calculations, it lacks the ability to perceive its intrinsic calculation errors. The GPR model can output the operation mode evaluation interval within a certain confidence range, which improves the reliability of the data-driven method. If the index interval of the operation mode evaluation based on the GPR model is wide, it means that the GPR model has a high uncertainty in evaluating the current operation mode. At this time, the mode department can adopt a more conservative evaluation value, such as the lower bound of the interval, the lower quantile, etc.
[0160] To realize the above-mentioned AI-based intelligent operation and maintenance solution for power grid systems, we need to combine multiple technical fields, including sensor data acquisition, data transmission, data processing and analysis, fault prediction and warning, fault isolation and recovery, and automatic control. Below I will outline the framework of a program implementation in a simplified way.
[0161] 1. Data Collection
[0162] Sensor data acquisition: Use sensor APIs or SDKs to obtain real-time data from sensors. This usually involves communication with hardware devices and requires the use of specific communication protocols (such as Modbus, MQTT, etc.).
[0163] Smart meters and environmental monitoring devices: Similarly, use the corresponding APIs or SDKs to obtain data from smart meters and environmental monitoring devices.
[0164] 2. Data Transmission
[0165] Wireless transmission and IoT technologies: Use technologies such as MQTT, HTTP, WebSocket, etc. to transmit data to a central processing unit (usually a server).
[0166] 3. Data Processing and Analysis
[0167] Big data processing: Use stream processing tools such as Apache Kafka and Flume to receive real-time data, and use Apache Spark and Hadoop for batch data processing and storage.
[0168] Deep learning algorithms: On the server side, use deep learning frameworks such as TensorFlow and PyTorch to build and train models for feature extraction and pattern recognition.
[0169] 4. Fault prediction and early warning
[0170] Model building and training: Train LSTM or other deep learning models using historical and real-time data.
[0171] Real-time prediction: For real-time incoming data, use the trained model to make predictions to determine whether there is a potential failure.
[0172] Early warning system: If a potential failure is detected, an early warning mechanism is triggered, such as sending an email, SMS or API call.
[0173] 5. Fault isolation and recovery
[0174] Fault isolation: Based on the warning information, fault isolation is performed through the API or SDK that controls the circuit breaker and recloser.
[0175] Fault recovery: Fault recovery is performed through APIs or SDKs that control backup lines or distributed power sources.
[0176] 6.Automation control
[0177] SCADA system: Integrate existing SCADA system API or SDK to achieve remote monitoring and control.
[0178] Automatic control algorithm: Use automatic control algorithms such as PID control algorithm to dynamically adjust grid operation parameters based on real-time data and prediction results.
[0179] Sample code framework
[0180] #Data Collection and Transmission
[0181] ```Python
[0182] #Assume that there is a function that can obtain data from sensors, smart meters and environmental monitoring devices
[0183] data_from_sensors=get_data_from_sensors()
[0184] data_from_meters=get_data_from_meters()
[0185] data_from_environment=get_data_from_environment()
[0186] #Send data to the central processing unit (server)
[0187] send_data_to_server(data_from_sensors,data_from_meters,data_from_environment)
[0188] ```
[0189] #Data Processing and Analysis
[0190] ```Python
[0191] #Assume that the server has received the data and stored it in the database
[0192] #Use Spark or similar tools for data processing
[0193] processed_data = process_data_with_spark(data_from_database) #Use deep learning model for feature extraction and pattern recognition predictions = predict_with_model(processed_data)
[0194] ```
[0195] #Fault prediction and warning
[0196] ```Python
[0197] #If a potential fault is predicted if is_potential_fault(predictions):
[0198] #Trigger the warning mechanism
[0199] trigger_alarm()
[0200] ```
[0201] #Fault Isolation and Recovery
[0202] ```Python
[0203] #Isolate faults based on warning information isolate_fault_area()
[0204] #Recover from fault recover_from_fault()
[0205] ```
[0206] #Automation Control
[0207] ```Python
[0208] #Remote monitoring and control of power grid equipment
[0209] monitor_and_control_grid_devices()
[0210] #Use automated control algorithms to adjust grid operating parameters
[0211] optimize_grid_parameters()
[0212] ```
[0213] The above code is only an example of a framework and pseudocode. The actual implementation will involve more details and complexity, including communication with hardware devices, database design, model training, API integration, etc. In addition, issues such as security, reliability, and performance optimization need to be considered.
[0214] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with the technical field within the technical scope disclosed by the present invention and within the spirit and principle of the present invention should be covered by the protection scope of the present invention.
Claims
1. An intelligent operation and maintenance method for a power grid system based on artificial intelligence, characterized in that: The following steps are involved: Data collection: Using sensors deployed in substations, transmission lines, and distribution equipment, the voltage, current, power, temperature and other operating data of each node in the power grid are collected in real time; smart meters installed at the user end are used to collect information on power consumption and power consumption period; Collect external environmental data through environmental monitoring equipment; Data transmission: The collected data is transmitted to the central processing unit in real time through wireless transmission technology and Internet of Things technology; Data processing and analysis: The central processing unit receives data transmitted by the data acquisition module and sends it to the data processing and analysis module. The data processing and analysis module uses big data technology to clean, store and pre-process the massive data collected, and uses deep learning algorithms to extract features and recognize patterns from the processed data to analyze the operation status of the power grid. Fault prediction: Build a fault prediction model based on historical data and real-time data, and identify potential fault modes by training the model; when the system detects a potential fault, the prediction model outputs a fault warning signal; Fault isolation and recovery: Based on the fault prediction results, the fault isolation mechanism is automatically triggered to isolate the fault area through circuit breakers and reclosers to prevent the fault from spreading. At the same time, the system starts the fault recovery mechanism and tries to restore power supply through backup lines or distributed power sources to ensure the continuous operation of the power grid. Automation control: Integrate the SCADA system to realize remote monitoring and control of power grid equipment; use automation control algorithms to dynamically adjust power grid operating parameters and optimize power grid operating status.
2. The intelligent operation and maintenance method of a power grid system based on artificial intelligence according to claim 1, characterized in that: The following steps are involved: Sensor data processing: Collect real-time data on voltage, current, power, and temperature from sensors deployed on substations, transmission lines, and distribution equipment. Use smart meters to collect information on user-side power consumption and power consumption time periods, as well as external environmental data such as temperature, humidity, and wind speed collected by environmental monitoring equipment. Data cleaning and preprocessing: The data processing and analysis module cleans the massive data collected, including removing noise and abnormal data and filling missing values; and preprocesses the cleaned data, including data normalization and dimensionality reduction operations; Feature extraction and pattern recognition: Use deep learning algorithms (such as convolutional neural networks and recursive neural networks) to extract features from preprocessed data, identify patterns and abnormal conditions in grid operation, and evaluate grid operation status by analyzing voltage, current, and power characteristic values.
3. The intelligent operation and maintenance method of a power grid system based on artificial intelligence according to claim 1, characterized in that: The following steps are involved: Fault prediction and warning: The fault prediction module uses historical data and real-time data to build a fault prediction model and identify potential fault modes of the power grid by training the model. When the system detects a potential fault, the prediction model outputs a fault warning signal and sends the warning information to the fault isolation and recovery module. Fault isolation and recovery: The fault isolation and recovery module automatically triggers the fault isolation mechanism based on the fault prediction results, isolates the fault area through circuit breakers and reclosers to prevent the fault from spreading; at the same time, it starts the fault recovery mechanism to restore power supply through backup lines or distributed power sources to ensure the continuous operation of the power grid; Operation parameter optimization: The automation control module uses the SCADA system to realize remote monitoring and control of power grid equipment, applies automation control algorithms, dynamically adjusts power grid operation parameters such as voltage, current, and power, optimizes the power grid operation status, and improves power grid operation efficiency and stability.
4. The method for intelligent operation and maintenance of a power grid system based on artificial intelligence according to claim 1, characterized in that: The data processing and analysis module includes: Big data technology: used to clean, store and pre-process the massive amounts of data collected to ensure the integrity and accuracy of the data; Deep learning model: Convolutional neural network (CNN) and recurrent neural network (RNN) are used to perform feature extraction and pattern recognition on the processed data. CNN is used to extract the spatial features of voltage and current data, and RNN is used to capture the temporal features of time series data. Through comprehensive analysis of these features, the operation status of the power grid can be accurately evaluated and abnormal situations can be identified.
5. The method for intelligent operation and maintenance of a power grid system based on artificial intelligence according to claim 1, characterized in that: The fault prediction module comprises: Fault prediction model: Use historical data and real-time data to build a fault prediction model based on deep learning. By training the model to identify potential fault modes, a long short-term memory network (LSTM) or a bidirectional long short-term memory network (BiLSTM) is used to process power grid operation data to predict faults. Fault warning system: When the fault prediction model detects a potential fault, it outputs a fault warning signal and triggers the automated processing flow of the fault isolation and recovery module based on the warning information. It isolates the faulty area through circuit breakers and reclosers, and attempts to restore power supply through backup lines or distributed power sources to ensure the stable operation of the power grid.
6. An artificial intelligence-based intelligent operation and maintenance system for power grid systems, characterized in that: The system includes: The data acquisition module uses various sensors, smart meters and environmental monitoring equipment to collect the voltage, current, frequency and temperature operation data of each node of the power grid in real time; the collected data is transmitted to the central processing unit through wireless transmission technology and Internet of Things technology; The central processing unit is connected to the data acquisition module and is used to receive data transmitted by the data acquisition module and send it to the data processing and analysis module; The data processing and analysis module is connected to the central processing unit and uses big data technology to clean, store and pre-process the massive amounts of data collected. It uses deep learning algorithms (such as convolutional neural networks and recursive neural networks) to extract features and recognize patterns from the processed data and analyze the operation status of the power grid. The fault prediction module is connected with the data processing and analysis module to build a fault prediction model based on historical data and real-time data, and identify potential fault modes through training models; when the system detects a potential fault, the prediction model outputs a fault warning signal; The fault isolation and recovery module is connected to the fault prediction module. According to the fault prediction results, it automatically triggers the fault isolation mechanism and isolates the fault area through circuit breakers and reclosers to prevent the fault from spreading. At the same time, the system starts the fault recovery mechanism and tries to restore power supply through backup lines or distributed power sources to ensure the continuous operation of the power grid. The automation control module is connected to the central processing unit and integrated with the SCADA system to achieve remote monitoring and control of power grid equipment; it uses automation control algorithms to dynamically adjust power grid operating parameters and optimize power grid operating status.
7. The artificial intelligence-based intelligent operation and maintenance system for power grid systems according to claim 1, characterized in that: The sensors are deployed in substations, transmission lines, and distribution equipment to collect real-time data on voltage, current, power, and temperature; the smart meters are installed at the user end to collect information on power consumption and power consumption periods; the environmental monitoring equipment collects external environmental data; the wireless transmission technology uses wireless technologies such as 5G, WiFi, LoRa, and NBIoT to transmit data to the data center; some data is pre-processed before data transmission to reduce network bandwidth requirements and improve response speed; The data processing and analysis module uses big data technology to clean, store and pre-process the collected massive data, specifically including: Data cleaning: remove noise and outliers, fill in missing data, and ensure data quality; Data integration: Standardize data from different sources to form a unified data format.
8. The artificial intelligence-based intelligent operation and maintenance system for power grid systems according to claim 1, characterized in that: The feature extraction is to extract important features from the original data using deep learning techniques such as autoencoders; Feature selection algorithms are used to select key features, reduce data dimensions, and improve model efficiency.
9. The power grid system intelligent operation and maintenance system based on artificial intelligence as claimed in claim 1, characterized in that: The automation control module integrates the SCADA system to realize remote monitoring and control of power grid equipment; uses the automation control algorithm to dynamically adjust the power grid operation parameters and optimize the power grid operation status, specifically including: Automated operation: Reduce manual intervention, improve fault response speed, and reduce the risk of human operational errors; Intelligent decision support: Use artificial intelligence technology to provide decision support to help operation and maintenance personnel make accurate judgments and respond quickly; Centralized monitoring: Establish a unified operation and maintenance management platform to monitor the grid operation status and fault information in real time; Data visualization: Visualization tools are used to display grid operation data, fault detection results, and recovery status, so that operation and maintenance personnel can intuitively understand the system status; Fault recording and analysis: The system automatically records detailed information of each fault for subsequent analysis and improvement; Continuous learning and optimization: Based on fault records and recovery situations, the fault prediction model and recovery strategy are continuously optimized to improve the intelligence level of the system.
10. An artificial intelligence-based intelligent operation and maintenance method for a power grid system according to any one of claims 19, characterized in that: The method includes: S21: Through the data acquisition module, various sensors, smart meters and environmental monitoring equipment are used to collect the voltage, current, frequency, temperature and other operating data of each node of the power grid in real time; the collected data is transmitted to the central processing unit through wireless transmission technology and Internet of Things technology; the data transmitted by the data acquisition module is received by the central processing unit and sent to the data processing and analysis module; S22: Through the data processing and analysis module, use big data technology to clean, store and pre-process the massive data collected; use deep learning algorithms (such as convolutional neural networks and recursive neural networks) to extract features and recognize patterns on the processed data to analyze the operation status of the power grid; S23: Through the fault prediction module, a fault prediction model based on historical data and real-time data is constructed, and potential fault modes are identified through training models; when the system detects a potential fault, the prediction model outputs a fault warning signal; S24: Through the fault isolation and recovery module, according to the fault prediction results, the fault isolation mechanism is automatically triggered, and the fault area is isolated through circuit breakers and reclosers to prevent the fault from spreading; at the same time, the system starts the fault recovery mechanism and tries to restore power supply through backup lines or distributed power sources to ensure the continuous operation of the power grid; S25: Through the automation control module, the SCADA system is integrated to realize remote monitoring and control of power grid equipment; the automation control algorithm is used to dynamically adjust the power grid operation parameters and optimize the power grid operation status.
Citation Information
Cited By
Visual intelligent monitoring system and method for power distribution cabinet, computing equipment and storage medium
CN119298395A
Integrated device for adding potential safety hazard to current collection line
CN119535095A
Safety hazard integrated device for overhead line
CN119535095B
Transparent visual power grid operation and maintenance system
CN119561255A
Monitoring method and monitoring system based on electric power operation and maintenance big data
CN119628251A
