Power distribution network fault judgment system and method based on edge calculation
By collecting and processing data at the edge nodes of the power distribution network and using machine learning algorithms for intelligent analysis, the data processing delay and security problems in the power distribution network monitoring system are solved, and efficient and reliable fault detection and response are achieved, reducing the number of problems. Operating costs and enhancing data security and personalized service capabilities.
Patent Information
- Application Number
- CN202410934834.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-12
- Publication Date
- 2024-11-29
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing power distribution network monitoring system has problems such as data processing delay, bandwidth limitation, centralized risk, low resource reuse rate, high operation and maintenance costs, data security threats and scalability restrictions, especially the inconsistent distribution IoT information model. , poor data fusion quality, low synergy efficiency of edge and end resources and insufficient data security.
Adopting a distributed architecture based on edge computing, data acquisition and processing is carried out on the edge nodes of the power distribution network, and intelligent analysis and fault identification is used to realize intelligent data on-site analysis, reduce data transmission distance, and improve fault detection. and response speed, and enhance system reliability and security through redundant settings and adaptive learning capabilities.
Realize near real-time data processing capabilities, reduce fault detection and response time, improve system reliability and fault tolerance, provide more accurate fault analysis results, reduce operational costs, and enhance data security and personalization. Service capability.
Smart Images

Figure CN119044663A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field, and in particular to a power distribution network fault diagnosis system and method based on edge computing. Background Art
[0002] The existing power distribution network monitoring system mostly adopts centralized data processing, which has problems such as data processing delay, bandwidth limitation and centralization risk. In addition, the traditional fault analysis method relies on manual experience and lacks intelligence and automation. In view of the above technical difficulties, based on the engineering and technical problems caused by the different demands of user owners and power grid enterprises in the process of electricity use, with the goal of meeting the energy needs of multiple users, the joint efforts of production, learning, research and application are combined to jointly tackle the problems from the optimization of access schemes, user-side optimization control and system-level platform optimization management, and the innovative results of user-side comprehensive energy application combining theory and practice are achieved. The traditional power distribution network monitoring system mainly includes the following modules: Data acquisition: Through sensors and measuring equipment installed in various parts of the power grid, voltage, current, power, frame rate, electric energy, temperature and other data are collected in real time. Data transmission: The collected data is transmitted to the central server using the communication network. Data processing and storage: The central server processes the data, such as data cleaning and formatting, and stores the processed data in the database. Status monitoring and analysis: The monitoring software platform analyzes the data in real time, identifies the operation status of the power grid, and detects abnormalities or faults. Alarm and notification: When an abnormality or fault is detected, the system will automatically issue an alarm and notify relevant personnel. Decision support: Provide the information and tools required by operators to assist in fault diagnosis and decision making. Automated control: According to the preset control strategy, the system can automatically perform some control operations, such as disconnecting the faulty line and adjusting the load distribution. Reporting and recording: The system generates various reports and records for auditing, maintaining and improving the operation of the power grid. Disadvantages of existing technologies: The information model of the power distribution Internet of Things is not unified: The power distribution Internet of Things system involves many types of power grid terminals and different interface protocols, and lacks a unified information architecture and information model. In addition, the traditional way of professional construction results in low resource reuse rate and huge operation and maintenance costs, resulting in a lack of technical foundation for the coordination of computing resources and the fusion of Internet of Things in the power distribution Internet of Things; the quality of data fusion in the power distribution Internet of Things is poor: The power distribution Internet of Things involves the high-frequency collection of more than 300 types of data. The existing power distribution intelligent edge computing device lacks data fusion models, data correction and alignment methods, and cannot guarantee the quality of data collected by the terminal. The quality of data directly affects the effect of upper-level business applications. Low efficiency of resource coordination in the edge and cloud of power distribution: In the distribution network, edge computing has been used to effectively solve the problem that the traditional cloud platform approach does not meet the second-level and minute-level data collection and access of the distribution network. However, the edge side of the distribution network is often constrained by the solidification of multi-dimensional resources such as computing, communication, and storage, which limits the reuse and scheduling of widely distributed and large-scale distribution computing resources. Data security issues: In addition to the various threats faced by general wireless networks, such as information leakage, information tampering, replay attacks, and denial of service, the power terminal side also faces the threat that the end-side nodes are easily physically manipulated by attackers, thereby controlling part of the network.At the same time, edge devices are also characterized by large numbers and dispersed deployment. During the deployment process, it is necessary to not consider the specific location of use and installation level, so the security mechanism needs to take into account the plug-and-play characteristics of the equipment. Scalability issues: As the scale of the power grid expands, centralized systems may face scalability limitations. Summary of the invention
[0003] The present invention solves the problem of data security caused by low collaborative efficiency of distribution cloud-edge resources, and proposes a power distribution network fault judgment system and method based on edge computing. Edge computing nodes are used to process large amounts of data, and the overall data processing capability and efficiency are improved by distributing computing tasks. A distributed architecture is adopted, and even if an edge node fails, other nodes can continue to operate, thereby ensuring the continuity and reliability of the system.
[0004] In order to achieve the above objectives, the following technical solutions are proposed: A power distribution network fault judgment system based on edge computing, comprising: Data acquisition module, which obtains power grid operation data in real time and sends it to the edge computing server; The edge computing server receives the operation instructions from the central server, identifies faults in the power grid operation data, obtains fault identification results, and feeds them back to the central server; The central server sends operation instructions to the edge computing server and receives the fault identification results of the edge computing server; The edge computing servers are provided with several and deployed at various key nodes of the power distribution network. Each of the edge computing servers is electrically connected to a collection module group consisting of several data collection modules. The collection module groups of the edge computing servers of adjacent nodes have cross-repeated data collection modules.
[0005] The present invention adopts a centralized system, which usually requires data to be transmitted from the edge of the distribution network to the central server for processing. This process introduces significant transmission and processing delays. The system of the present invention collects and processes data at the edge node, realizes intelligent analysis of data on the spot, reduces the data transmission distance, and thus realizes near real-time data processing capabilities, greatly improving fault detection and response speed. The centralized system of the present invention is highly dependent on the central server. Once the central server fails, the entire monitoring system may be paralyzed. The present invention adopts a distributed architecture. Even if an edge node fails, other nodes can continue to operate. In addition, adjacent nodes are set up redundantly, thereby ensuring the continuity and reliability of the system. Centralized systems may encounter performance bottlenecks when processing large-scale data, especially under high load conditions. The edge computing nodes of the present invention can process large amounts of data in parallel, and by distributing computing tasks, the overall data processing capabilities and efficiency are improved. This invention is guided by the overall plan, uses modern information technology, relies on artificial intelligence technology, and builds an edge computing-based power distribution network data intelligent analysis and fault diagnosis system based on the unified Internet application platform of provincial-level industrial units. It integrates "Internet +" with the original traditional industries, collects real-time operation data of power terminal production equipment, and carries out intelligent analysis, intelligent diagnosis, and intelligent decision-making in parallel. It comprehensively grasps the equipment operation status and power consumption of power-consuming enterprises, analyzes and mines energy consumption characteristics, and helps users save energy and reduce consumption; through big data analysis, it accurately matches user needs and provides personalized services, saves resource investment, scientifically and efficiently operates and maintains, and creates greater value.
[0006] Preferably, the edge computing server is electrically connected to: Data preprocessing module, preprocessing the power grid operation data; Intelligent analysis engine that identifies abnormal patterns in grid operation based on preprocessed grid operation data; The fault analysis module identifies and locates the fault type based on the abnormal pattern combined with the power grid model and historical fault data to obtain the fault identification result.
[0007] The intelligent analysis engine of the present invention uses a machine learning algorithm to conduct in-depth analysis of the processed data to identify abnormal patterns in power grid operation. The fault analysis module of the present invention identifies and locates fault types based on the results of the intelligent analysis engine, combined with power grid models and historical fault data. Traditional centralized systems may rely on preset thresholds and rules, and have limited ability to analyze and judge complex faults. The present invention uses machine learning and artificial intelligence technology to automatically identify complex fault patterns and achieve deeper fault analysis and prediction.
[0008] Preferably, the preprocessing includes data cleaning, standardization and normalization of the collected power grid operation data.
[0009] Preferably, the edge computing server is electrically connected to a decision support module, and the decision support module is preset with decision suggestion information of corresponding measures to deal with various types of faults. The edge computing server sends the fault identification result and decision suggestion information to the central server.
[0010] Preferably, the edge computing server and the central server exchange data and issue commands via a communication module.
[0011] Preferably, a data security detection module is electrically connected between the central server and the communication module, and the data security detection module performs virus disinfection on the data uploaded by the edge computing server.
[0012] Preferably, the edge computing server is electrically connected to a user interaction interface, and the user interaction interface provides the user with real-time data display, fault identification result display and decision suggestion display.
[0013] A method for judging power distribution network faults based on edge computing, using the above-mentioned power distribution network fault judgment system based on edge computing, includes the following steps: S1, real-time collection of power grid operation data of various terminal devices; S2, preprocessing the power grid operation data; S3, judging whether there is an abnormal mode in the operation of the power grid according to the pre-processed power grid operation data, if so, proceeding to S4, if not, cyclically judging the pre-processed power grid operation data; S4, combining the power grid model and historical fault data, identifying and locating the fault type to obtain the fault identification result; S5, the edge computing server sends the fault identification result to the central server.
[0014] By deploying computing resources at the edge of the power grid, the present invention can perform real-time data processing near the data source, reduce data transmission time, and quickly respond to power grid events. Compared with traditional centralized systems, the data processing delay of the present invention is significantly reduced because it eliminates the step of sending data to a remote central server. The distributed architecture of the present invention reduces the dependence on a single central point, and the system can continue to operate even if an edge node fails, thereby improving the overall fault tolerance. The edge nodes of the present invention can work independently or collaboratively to process large amounts of data, reduce the burden on the central server, and improve the efficiency and speed of data processing. The present invention is intelligent in fault analysis and judgment, and utilizes advanced data analysis technology and machine learning algorithms. The present invention can automatically identify and classify power grid faults and provide more accurate and objective analysis results than manual experience. The present invention has adaptive learning capabilities, and the system can continuously learn and optimize based on real-time data and historical fault cases to improve the accuracy of fault prediction and analysis.
[0015] Preferably, S5 includes performing a data security check on the fault identification result, and the central server receives the fault identification result after passing the data security check.
[0016] The S5 includes performing data security detection on the fault identification result to determine whether the fault identification result passes the data security detection. If so, the central server receives the fault identification result. If not, the central server does not receive the fault identification result and generates virus prompt information to the central server.
[0017] Preferably, S5 further includes obtaining decision suggestion information for coping with the corresponding fault according to the fault identification result, and sending the fault identification result and the decision suggestion information to the central server.
[0018] The S5 also includes obtaining decision suggestion information for dealing with corresponding faults according to the fault identification results. The decision support module is preset with decision suggestion information for corresponding measures to deal with various types of faults. The decision suggestion information is output according to the fault identification results. The edge computing server sends the decision suggestion information and the fault identification results to the central server.
[0019] The beneficial effects of the present invention are: 1. Edge computing capabilities: By deploying computing resources at the edge of the power grid, the present invention can perform real-time data processing near the data source, reduce data transmission time, and quickly respond to power grid events.
[0020] 2. Low latency: Compared with traditional centralized systems, the data processing latency of the present invention is significantly reduced because it eliminates the step of sending data to a remote central server.
[0021] 3. High reliability: The distributed architecture reduces the reliance on a single central point. Even if an edge node fails, the system can continue to operate, improving the overall fault tolerance.
[0022] 4. Distributed data processing: The edge nodes of the present invention can work independently or collaboratively to process large amounts of data, reduce the burden on the central server, and improve the efficiency and speed of data processing.
[0023] 5. Intelligent fault analysis: Using advanced data analysis technology and machine learning algorithms, the present invention can automatically identify and classify power grid faults, providing more accurate and objective analysis results than manual experience.
[0024] 6. Adaptive learning capability: The system can continuously learn and optimize based on real-time data and historical failure cases to improve the accuracy of fault prediction and analysis.
[0025] 7. Customized service: Users can customize analysis parameters and judgment logic according to specific needs, and the system provides personalized decision support.
[0026] 8. Data security and privacy protection: Edge computing reduces the transmission of sensitive data in the network, reduces the risk of data leakage, and enhances data security.
[0027] 9. Easy integration and expansion: The system design of the present invention takes modularity into consideration and can be easily integrated into the existing power distribution network architecture and expanded as needed.
[0028] 10. Cost-effectiveness: By reducing the reliance on central servers and optimizing the data processing flow, the present invention helps to reduce operating costs.
[0029] 11. Environmental adaptability: Edge computing nodes can adapt to different power grid environments, including remote areas or areas with limited network connectivity. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] Figure 1 It is a schematic diagram of the system structure of the present invention.
[0031] Figure 2 It is a flow chart of the method of the present invention. DETAILED DESCRIPTION
[0032] Embodiment 1: This embodiment proposes a power distribution network fault diagnosis system based on edge computing, referring to Figure 1, including several data acquisition modules, several edge computing servers and a central server. The several edge computing servers are deployed at various key nodes of the power distribution network and are responsible for the standardized access and preliminary data processing of the data of various terminal devices of the power distribution network. One edge computing server is electrically connected to a collection module group, and a collection module group is composed of several data acquisition modules. The data acquisition modules of the collection module groups of the edge computing servers of adjacent nodes can be cross-set repeatedly. Even if one edge computing node is damaged, another node can also perform redundant calculations to ensure the stability of the system. The data acquisition module is used to collect the power grid operation data of various terminal devices in real time, including current, voltage, power quality, equipment status and other information. And the power grid operation data is sent to the edge computing server; the edge computing server is used to receive the operation instructions of the central server, use the power grid operation data to perform fault identification to obtain the fault identification result, and feed back the fault identification result to the central server; the central server is used to send the operation instructions to the edge computing server, and at the same time to receive the fault identification result sent by the edge computing server. The central server and the edge computing server exchange data and send commands through the communication module. The edge computing server is electrically connected to a user interaction interface that provides real-time data display, decision suggestion display and fault identification result display for users.
[0033] The edge computing server is electrically connected to a data preprocessing module, an intelligent analysis engine and a fault analysis module. The data preprocessing module is used to preprocess the power grid operation data; the data preprocessing module cleans, standardizes and normalizes the collected raw data.
[0034] The intelligent analysis engine is used to identify abnormal patterns in power grid operation using pre-processed power grid operation data; the fault analysis module is used to identify and locate fault types based on abnormal patterns, combined with historical fault data and power grid models, and then obtain fault identification results.
[0035] The intelligent analysis engine of the present invention uses a machine learning algorithm to conduct in-depth analysis of the processed data to identify abnormal patterns in power grid operation. The fault analysis module of the present invention identifies and locates fault types based on the results of the intelligent analysis engine, combined with power grid models and historical fault data. Traditional centralized systems may rely on preset thresholds and rules, and have limited ability to analyze and judge complex faults. The present invention uses machine learning and artificial intelligence technology to automatically identify complex fault patterns and achieve deeper fault analysis and prediction.
[0036] The present invention adopts a centralized system, which usually requires data to be transmitted from the edge of the distribution network to the central server for processing. This process introduces significant transmission and processing delays. The system of the present invention collects and processes data at the edge node, realizes intelligent analysis of data on the spot, reduces the data transmission distance, and thus realizes near real-time data processing capabilities, greatly improving fault detection and response speed. The centralized system of the present invention is highly dependent on the central server. Once the central server fails, the entire monitoring system may be paralyzed. The present invention adopts a distributed architecture. Even if an edge node fails, other nodes can continue to operate. In addition, adjacent nodes are set up redundantly, thereby ensuring the continuity and reliability of the system. Centralized systems may encounter performance bottlenecks when processing large-scale data, especially under high load conditions. The edge computing nodes of the present invention can process large amounts of data in parallel, and by distributing computing tasks, the overall data processing capabilities and efficiency are improved.
[0037] This embodiment also proposes a power distribution network fault judgment method based on edge computing, using the above-mentioned power distribution network fault judgment system based on edge computing, referring to Figure 2 , including the following steps: S1, the data acquisition module collects the grid operation data of various terminal devices in real time, including current, voltage, power quality, equipment status and other information; S2, the data preprocessing module preprocesses the power grid operation data and cleans, standardizes and normalizes the collected raw data; S3, intelligent analysis engine: using machine learning algorithms, conducts in-depth analysis of the processed data to identify abnormal patterns in the operation of the power grid and determine whether there are abnormal patterns in the operation of the power grid. If so, proceed to S4. If not, perform cyclic judgment on the pre-processed power grid operation data; S4, fault analysis module: Based on the results of the intelligent analysis engine, combined with the power grid model and historical fault data, the fault type is identified and located to obtain the fault identification result; S5: The edge computing server sends the obtained fault identification result to the central server.
[0038] By deploying computing resources at the edge of the power grid, the present invention can perform real-time data processing near the data source, reduce data transmission time, and quickly respond to power grid events. Compared with traditional centralized systems, the data processing delay of the present invention is significantly reduced because it eliminates the step of sending data to a remote central server. The distributed architecture of the present invention reduces the dependence on a single central point, and the system can continue to operate even if an edge node fails, thereby improving the overall fault tolerance. The edge nodes of the present invention can work independently or collaboratively to process large amounts of data, reduce the burden on the central server, and improve the efficiency and speed of data processing. The present invention is intelligent in fault analysis and judgment, and utilizes advanced data analysis technology and machine learning algorithms. The present invention can automatically identify and classify power grid faults and provide more accurate and objective analysis results than manual experience. The present invention has adaptive learning capabilities, and the system can continuously learn and optimize based on real-time data and historical fault cases to improve the accuracy of fault prediction and analysis.
[0039] Embodiment 2: This embodiment adds data security detection on the basis of embodiment 1, and proposes a power distribution network fault judgment system based on edge computing. Figure 1 , including several data acquisition modules, several edge computing servers and a central server. The several edge computing servers are deployed at various key nodes of the power distribution network and are responsible for the standardized access and preliminary data processing of data of various terminal devices in the power distribution network. One edge computing server is electrically connected to a collection module group, and a collection module group is composed of several data acquisition modules. The data acquisition modules of the collection module groups of the edge computing servers of adjacent nodes can be cross-set repeatedly. Even if one edge computing node is damaged, another node can also perform redundant calculations to ensure the stability of the system. The data acquisition module is used to collect the power grid operation data of various terminal devices in real time, including current, voltage, power quality, equipment status and other information. And the power grid operation data is sent to the edge computing server; the edge computing server is used to receive the operation instructions of the central server, use the power grid operation data to perform fault identification to obtain the fault identification result, and feed back the fault identification result to the central server; the central server is used to send the operation instructions to the edge computing server, and at the same time to receive the fault identification result sent by the edge computing server. The central server and the edge computing server exchange data and issue commands through the communication module. The edge computing server is electrically connected to a user interaction interface that provides users with real-time data display, decision suggestion display and fault identification result display. A data security detection module for performing virus disinfection on data uploaded by the edge computing server is electrically connected between the communication module and the central server of the present invention.
[0040] The edge computing server is electrically connected to a data preprocessing module, an intelligent analysis engine and a fault analysis module. The data preprocessing module is used to preprocess the power grid operation data; the data preprocessing module cleans, standardizes and normalizes the collected raw data.
[0041] The intelligent analysis engine is used to identify abnormal patterns in power grid operation using pre-processed power grid operation data; the fault analysis module is used to identify and locate fault types based on abnormal patterns, combined with historical fault data and power grid models, and then obtain fault identification results.
[0042] The intelligent analysis engine of the present invention uses a machine learning algorithm to conduct in-depth analysis of the processed data to identify abnormal patterns in power grid operation. The fault analysis module of the present invention identifies and locates fault types based on the results of the intelligent analysis engine, combined with power grid models and historical fault data. Traditional centralized systems may rely on preset thresholds and rules, and have limited ability to analyze and judge complex faults. The present invention uses machine learning and artificial intelligence technology to automatically identify complex fault patterns and achieve deeper fault analysis and prediction.
[0043] The present invention adopts a centralized system, which usually requires data to be transmitted from the edge of the distribution network to the central server for processing. This process introduces significant transmission and processing delays. The system of the present invention collects and processes data at the edge node, realizes intelligent analysis of data on the spot, reduces the data transmission distance, and thus realizes near real-time data processing capabilities, greatly improving fault detection and response speed. The centralized system of the present invention is highly dependent on the central server. Once the central server fails, the entire monitoring system may be paralyzed. The present invention adopts a distributed architecture. Even if an edge node fails, other nodes can continue to operate. In addition, adjacent nodes are set up redundantly, thereby ensuring the continuity and reliability of the system. Centralized systems may encounter performance bottlenecks when processing large-scale data, especially under high load conditions. The edge computing nodes of the present invention can process large amounts of data in parallel, and by distributing computing tasks, the overall data processing capabilities and efficiency are improved.
[0044] This embodiment also proposes a power distribution network fault judgment method based on edge computing, using the above-mentioned power distribution network fault judgment system based on edge computing, referring to Figure 2 , including the following steps: S1, the data acquisition module collects the grid operation data of various terminal devices in real time, including current, voltage, power quality, equipment status and other information; S2, the data preprocessing module preprocesses the power grid operation data and cleans, standardizes and normalizes the collected raw data; S3, intelligent analysis engine: using machine learning algorithms, conducts in-depth analysis of the processed data to identify abnormal patterns in the operation of the power grid and determine whether there are abnormal patterns in the operation of the power grid. If so, proceed to S4. If not, perform cyclic judgment on the pre-processed power grid operation data; S4, fault analysis module: Based on the results of the intelligent analysis engine, combined with the power grid model and historical fault data, the fault type is identified and located to obtain the fault identification result; S5: The edge computing server sends the obtained fault identification result to the central server.
[0045] The S5 includes performing data security detection on the fault identification result to determine whether the fault identification result passes the data security detection. If so, the central server receives the fault identification result. If not, the central server does not receive the fault identification result and generates virus prompt information to the central server.
[0046] By deploying computing resources at the edge of the power grid, the present invention can perform real-time data processing near the data source, reduce data transmission time, and quickly respond to power grid events. Compared with traditional centralized systems, the data processing delay of the present invention is significantly reduced because it eliminates the step of sending data to a remote central server. The distributed architecture of the present invention reduces the dependence on a single central point, and the system can continue to operate even if an edge node fails, thereby improving the overall fault tolerance. The edge nodes of the present invention can work independently or collaboratively to process large amounts of data, reduce the burden on the central server, and improve the efficiency and speed of data processing. The present invention is intelligent in fault analysis and judgment, and utilizes advanced data analysis technology and machine learning algorithms. The present invention can automatically identify and classify power grid faults and provide more accurate and objective analysis results than manual experience. The present invention has adaptive learning capabilities, and the system can continuously learn and optimize based on real-time data and historical fault cases to improve the accuracy of fault prediction and analysis.
[0047] Embodiment 3: This embodiment adds decision suggestions on the basis of embodiment 2, and proposes a power distribution network fault judgment system based on edge computing. Figure 1, including several data acquisition modules, several edge computing servers and a central server. The several edge computing servers are deployed at various key nodes of the power distribution network and are responsible for the standardized access and preliminary data processing of data of various terminal devices in the power distribution network. One edge computing server is electrically connected to a collection module group, and a collection module group is composed of several data acquisition modules. The data acquisition modules of the collection module groups of the edge computing servers of adjacent nodes can be cross-set repeatedly. Even if one edge computing node is damaged, another node can also perform redundant calculations to ensure the stability of the system. The data acquisition module is used to collect the power grid operation data of various terminal devices in real time, including current, voltage, power quality, equipment status and other information. And the power grid operation data is sent to the edge computing server; the edge computing server is used to receive the operation instructions of the central server, use the power grid operation data to perform fault identification to obtain the fault identification result, and feed back the fault identification result to the central server; the central server is used to send the operation instructions to the edge computing server, and at the same time to receive the fault identification result sent by the edge computing server. The central server and the edge computing server exchange data and issue commands through the communication module. The edge computing server is electrically connected to a user interaction interface that provides users with real-time data display, decision suggestion display and fault identification result display. A data security detection module for performing virus disinfection on data uploaded by the edge computing server is electrically connected between the communication module and the central server of the present invention.
[0048] The edge computing server is electrically connected to a decision support module with decision suggestion information preset for corresponding measures to deal with various types of faults, and the edge computing server sends the fault identification result and decision suggestion information to the central server. The decision support module provides suggestions for maintenance, adjustment or preventive measures based on the fault analysis results.
[0049] The edge computing server is electrically connected to a data preprocessing module, an intelligent analysis engine and a fault analysis module. The data preprocessing module is used to preprocess the power grid operation data; the data preprocessing module cleans, standardizes and normalizes the collected raw data.
[0050] The intelligent analysis engine is used to identify abnormal patterns in power grid operation using pre-processed power grid operation data; the fault analysis module is used to identify and locate fault types based on abnormal patterns, combined with historical fault data and power grid models, and then obtain fault identification results.
[0051] The intelligent analysis engine of the present invention uses a machine learning algorithm to conduct in-depth analysis of the processed data to identify abnormal patterns in power grid operation. The fault analysis module of the present invention identifies and locates fault types based on the results of the intelligent analysis engine, combined with power grid models and historical fault data. Traditional centralized systems may rely on preset thresholds and rules, and have limited ability to analyze and judge complex faults. The present invention uses machine learning and artificial intelligence technology to automatically identify complex fault patterns and achieve deeper fault analysis and prediction.
[0052] The present invention adopts a centralized system, which usually requires data to be transmitted from the edge of the distribution network to the central server for processing. This process introduces significant transmission and processing delays. The system of the present invention collects and processes data at the edge node, realizes intelligent analysis of data on the spot, reduces the data transmission distance, and thus realizes near real-time data processing capabilities, greatly improving fault detection and response speed. The centralized system of the present invention is highly dependent on the central server. Once the central server fails, the entire monitoring system may be paralyzed. The present invention adopts a distributed architecture. Even if an edge node fails, other nodes can continue to operate. In addition, adjacent nodes are set up redundantly, thereby ensuring the continuity and reliability of the system. Centralized systems may encounter performance bottlenecks when processing large-scale data, especially under high load conditions. The edge computing nodes of the present invention can process large amounts of data in parallel, and by distributing computing tasks, the overall data processing capabilities and efficiency are improved.
[0053] This embodiment also proposes a power distribution network fault judgment method based on edge computing, using the above-mentioned power distribution network fault judgment system based on edge computing, referring to Figure 2 , including the following steps: S1, the data acquisition module collects the grid operation data of various terminal devices in real time, including current, voltage, power quality, equipment status and other information; S2, the data preprocessing module preprocesses the power grid operation data and cleans, standardizes and normalizes the collected raw data; S3, intelligent analysis engine: using machine learning algorithms, conducts in-depth analysis of the processed data to identify abnormal patterns in the operation of the power grid and determine whether there are abnormal patterns in the operation of the power grid. If so, proceed to S4. If not, perform cyclic judgment on the pre-processed power grid operation data; S4, fault analysis module: Based on the results of the intelligent analysis engine, combined with the power grid model and historical fault data, the fault type is identified and located to obtain the fault identification result; S5: The edge computing server sends the obtained fault identification result to the central server.
[0054] The S5 includes performing data security detection on the fault identification result to determine whether the fault identification result passes the data security detection. If so, the central server receives the fault identification result. If not, the central server does not receive the fault identification result and generates virus prompt information to the central server.
[0055] The S5 also includes obtaining decision suggestion information for dealing with corresponding faults according to the fault identification results. The decision support module is preset with decision suggestion information for corresponding measures to deal with various types of faults. The decision suggestion information is output according to the fault identification results. The edge computing server sends the decision suggestion information and the fault identification results to the central server.
[0056] By deploying computing resources at the edge of the power grid, the present invention can perform real-time data processing near the data source, reduce data transmission time, and quickly respond to power grid events. Compared with traditional centralized systems, the data processing delay of the present invention is significantly reduced because it eliminates the step of sending data to a remote central server. The distributed architecture of the present invention reduces the dependence on a single central point, and the system can continue to operate even if an edge node fails, thereby improving the overall fault tolerance. The edge nodes of the present invention can work independently or collaboratively to process large amounts of data, reduce the burden on the central server, and improve the efficiency and speed of data processing. The present invention is intelligent in fault analysis and judgment, and utilizes advanced data analysis technology and machine learning algorithms. The present invention can automatically identify and classify power grid faults and provide more accurate and objective analysis results than manual experience. The present invention has adaptive learning capabilities, and the system can continuously learn and optimize based on real-time data and historical fault cases to improve the accuracy of fault prediction and analysis.
Claims
1. A power distribution network fault judgment system based on edge computing, characterized in that: include: Data acquisition module, which obtains power grid operation data in real time and sends it to the edge computing server; The edge computing server receives the operation instructions from the central server, identifies faults in the power grid operation data, obtains fault identification results, and feeds them back to the central server; The central server sends operation instructions to the edge computing server and receives the fault identification results of the edge computing server; The edge computing servers are provided with several and deployed at various key nodes of the power distribution network. Each of the edge computing servers is electrically connected to a collection module group consisting of several data collection modules. The collection module groups of the edge computing servers of adjacent nodes have cross-repeated data collection modules.
2. According to the power distribution network fault diagnosis system based on edge computing according to claim 1, it is characterized in that: The edge computing server is electrically connected to: Data preprocessing module, preprocessing the power grid operation data; Intelligent analysis engine that identifies abnormal patterns in grid operation based on preprocessed grid operation data; The fault analysis module identifies and locates the fault type based on the abnormal pattern combined with the power grid model and historical fault data to obtain the fault identification result.
3. According to the power distribution network fault judgment system based on edge computing according to claim 2, it is characterized in that: The preprocessing includes data cleaning, standardization and normalization of the collected power grid operation data.
4. According to the power distribution network fault judgment system based on edge computing according to claim 2, it is characterized in that: The edge computing server is electrically connected to a decision support module, and the decision support module is preset with decision suggestion information of corresponding measures to deal with various types of faults. The edge computing server sends the fault identification result and decision suggestion information to the central server.
5. A power distribution network fault diagnosis system based on edge computing according to any one of claims 1 to 4, characterized in that: The edge computing server and the central server exchange data and issue commands through the communication module.
6. The power distribution network fault diagnosis system based on edge computing according to claim 5 is characterized in that: A data security detection module is electrically connected between the central server and the communication module, and the data security detection module performs virus disinfection on the data uploaded by the edge computing server.
7. A power distribution network fault diagnosis system based on edge computing according to any one of claims 1 to 4, characterized in that: The edge computing server is electrically connected to a user interaction interface, which provides users with real-time data display, fault identification result display and decision suggestion display.
8. A method for judging power distribution network faults based on edge computing, using a power distribution network fault judgment system based on edge computing as claimed in claim 2, characterized in that: The following steps are involved: S1, real-time collection of power grid operation data of various terminal devices; S2, preprocessing the power grid operation data; S3, judging whether there is an abnormal mode in the operation of the power grid according to the pre-processed power grid operation data, if so, proceeding to S4, if not, cyclically judging the pre-processed power grid operation data; S4, combining the power grid model and historical fault data, identifying and locating the fault type to obtain the fault identification result; S5, the edge computing server sends the fault identification result to the central server.
9. The method for determining power distribution network faults based on edge computing according to claim 8 is characterized in that: S5 includes performing a data security check on the fault identification result, and the central server receives the fault identification result after the data security check.
10. The method for determining power distribution network faults based on edge computing according to claim 8, characterized in that: The S5 also includes obtaining decision suggestion information for coping with the corresponding fault according to the fault identification result, and sending the fault identification result and the decision suggestion information to the central server.
Citation Information
Patent Citations
Distributed power distribution fault processing system and method based on edge calculation
CN115800520A
Power dispatching fault detection system based on big data
CN116186600A
Production power transformation fault monitoring method and system
CN117353444A
Power distribution network measurement and control system and method
CN117674140A
Power distribution network fault positioning method, system and device based on visual platform, and storage medium
CN117783769A
Cited By
Photovoltaic module online fault diagnosis system based on edge calculation
CN120763502A
