CN-121995158-A - Power distribution network line abnormal state identification method and system

CN121995158ACN 121995158 ACN121995158 ACN 121995158ACN-121995158-A

Abstract

The invention provides a method and a system for identifying abnormal states of a power distribution network line, which relate to the technical field of power distribution network abnormal identification, realize localization and low-delay response of data processing by constructing a distributed network based on topology analysis and edge calculation, remarkably improve the real-time performance of abnormal identification, dynamically correct a key node set and set a personalized threshold value by introducing an environment sensing mechanism, enable the system to adapt to the running environment and load change of the power distribution network, effectively reduce the false alarm rate, fuse multi-source data such as voltage, current, temperature and environment, and perform correlation and judgment on discrete abnormal information by combining a space-time aggregation analysis technology, not only enhance the comprehensiveness and anti-interference capability of fault diagnosis, but also accurately identify a fault propagation path and progressive hidden danger, and finally form a set of power distribution network active defense system with optimized resource configuration, rapid response and high confidence, and greatly improve the running reliability and operation and maintenance efficiency of the power network.

Inventors

WU RONGYUAN
WANG QIANG
WU JICAI
WU RUI

Assignees

四川天灵高科电气有限公司

Dates

Publication Date: 20260508
Application Date: 20260409

Claims (7)

1. The method for identifying the abnormal state of the power distribution network line is characterized by comprising the following steps of: S1, identifying and determining an initial key node set based on distribution network topology layout analysis, distributing edge computing nodes for the initial key nodes, and constructing a distributed edge computing network; s2, an environment sensing mechanism is established based on the intelligent sensing system of the multiple sensors and the historical data, the operation environment data of the power distribution network is monitored in real time, the set of initial key nodes is dynamically corrected, and the updated key nodes are formed and the corresponding operation abnormality judgment threshold value is set; s3, collecting line operation data through an intelligent sensing system in a current monitoring period to form current period line information, wherein the current period line information comprises voltage, current, temperature and environmental parameters; S4, inputting the current period line information to a corresponding edge computing node, comparing and analyzing the operation data of each updated key node by utilizing the generated abnormality judgment threshold value, and outputting the node abnormal state identification of each updated key node in the current period; S5, collecting node abnormal state identifiers of the updated key nodes in different characteristic frequencies in the current period, and carrying out space-time aggregation analysis on discrete node abnormal information by combining a preset fluctuation deviation threshold value to generate an identification analysis result.
2. The method for identifying abnormal states of power distribution network lines according to claim 1, wherein the step S1 is as follows: s101, reading topological structure data of a power distribution network, and sorting connection relation and load distribution characteristics of line nodes; s102, screening out the matched line nodes from the line node connection relation according to the electrical distance centrality and the load importance, and sorting the line nodes into an initial key node set; S103, distributing corresponding edge computing nodes for each line node in the initial key node set, establishing a one-to-one or one-to-many mapping relation, and constructing a distributed edge computing network.
3. The method for identifying abnormal states of power distribution network lines according to claim 1, wherein the step of step S2 is: S201, establishing an environment sensing mechanism, wherein the steps comprise collecting historical sensing data and operation parameters acquired by an intelligent sensing system, and arranging the historical sensing data and the operation parameter data according to a time sequence to form time sequence data, wherein the intelligent sensing system comprises a plurality of types of sensors, analyzing the change trend of the data based on the time sequence data by a moving average method to obtain a time sequence analysis result, setting an environment data deviation threshold corresponding to a plurality of time periods in the same period according to the time sequence analysis result, and incorporating the environment data into a set of initial key nodes when the operation environment data exceeds the environment data deviation threshold in the current time period; S202, monitoring operation environment data of the power distribution network in real time, and correcting an initial key node set to an updated key node set based on an established environment sensing mechanism; S203, collecting historical normal fluctuation ranges of all key nodes in the updated key node set, and setting a corresponding operation abnormality judgment threshold value for each key node in the updated key node set, wherein the operation abnormality judgment threshold value is located in the historical normal fluctuation range.
4. The method for identifying abnormal states of power distribution network lines according to claim 1, wherein the step of step S3 is: S301, setting a fixed time interval as a monitoring period, and collecting operation data of a power distribution network line, including voltage, current, temperature and environmental parameters, through an intelligent sensing system in the current monitoring period; S302, collecting voltage, current, temperature and environmental parameters, and finishing to form current period line information.
5. The method for identifying abnormal states of power distribution network lines according to claim 1, wherein the step S4 is as follows: s401, transmitting the current period line information to a corresponding edge computing node, calling the generated abnormality judgment threshold value of the key node after updating in the edge computing node, and comparing the voltage, the current and the temperature in the current period line information with the abnormality judgment threshold value one by one to obtain a comparison analysis result; S402, outputting a node abnormal state identification in the current period for each updated key node according to the comparison analysis result, wherein if any one index of voltage, current and temperature exceeds an abnormal judgment threshold value, the node is marked as an abnormal state.
6. The method for identifying abnormal states of power distribution network lines according to claim 1, wherein the step of step S5 is: S501, aggregating node abnormal state identifiers in all current periods under different characteristic frequencies, namely different sampling frequencies, in the current period; S502, analyzing abnormal state association of adjacent line nodes in the same time period and abnormal state change trend of the same line node in different time periods based on a preset fluctuation deviation threshold and node abnormal state identification in all current periods, and finally finishing to generate an identification analysis result.
7. A distribution network line abnormal state identification system, applying a distribution network line abnormal state identification method according to any one of claims 1-6, wherein the identification system comprises: the node distribution module is used for identifying and determining an initial key node set based on distribution network topology layout analysis, distributing edge computing nodes for the initial key nodes and constructing a distributed edge computing network; The dynamic updating module is used for establishing an environment sensing mechanism based on the intelligent sensing system of the multiple sensors and the historical data, monitoring the running environment data of the power distribution network in real time, dynamically correcting the set of initial key nodes, forming updated key nodes and setting corresponding running abnormality judgment thresholds; the data acquisition module is used for acquiring line operation data through the intelligent sensing system in the current monitoring period to form current period line information, wherein the current period line information comprises voltage, current, temperature and environmental parameters; The comparison analysis module is used for inputting the current period line information to the corresponding edge computing node, comparing and analyzing the operation data of each updated key node by utilizing the generated abnormality judgment threshold value, and outputting the node abnormal state identification of each updated key node in the current period; the result generation module is used for collecting node abnormal state identifiers of the updated key nodes in different characteristic frequencies in the current period, and carrying out space-time aggregation analysis on discrete node abnormal information by combining a preset fluctuation deviation threshold value to generate an identification analysis result.

Description

Power distribution network line abnormal state identification method and system Technical Field The invention relates to the technical field of power distribution network anomaly identification, in particular to a power distribution network line anomaly state identification method and system. Background The power distribution network is used as an important component of a power system, is directly oriented to wide users, and the safe and stable operation of the power distribution network is important to guaranteeing the normal operation of socioeconomic and the quality of life of people. When the traditional power distribution network line abnormal state identification method is used for coping with the complex situations, a plurality of limitations are exposed, and the requirements of the modern power distribution network on efficient, accurate and real-time monitoring and identification are difficult to meet. Firstly, a traditional monitoring mode mostly adopts a framework of terminal acquisition and cloud centralized processing, communication bandwidth congestion and transmission delay are caused when massive high-frequency sensing data are uploaded to a main station, the requirement of rapid response of a power distribution network fault millisecond level is difficult to meet, secondly, the selection of key monitoring nodes is often based on static topology or fixed experience, dynamic perceptibility of a real-time operation environment is lacked, monitoring key points cannot be timely adjusted when the environment is suddenly changed, a fixed unified threshold is difficult to adapt to historical fluctuation characteristics of different nodes, reporting omission or false reporting is extremely easy to occur, finally, the traditional method is mostly dependent on single electric quantity indexes to carry out independent judgment, depth fusion of multi-source heterogeneous data and correlation analysis capability of time and space are lacked, instantaneous interference and real faults are difficult to effectively distinguish, progressive hidden hazards with strong concealment cannot be accurately identified, and abnormal identification accuracy and reliability are insufficient. Therefore, it is necessary to provide a method and a system for identifying abnormal states of a power distribution network line to solve the above technical problems. Disclosure of Invention In order to solve the technical problems, the invention provides a method and a system for identifying abnormal states of a power distribution network line, which are used for solving the problems that the requirement of millisecond-level quick response of power distribution network faults is difficult to meet and the selection of key monitoring nodes lacks dynamic perception capability to a real-time operation environment in the prior art. The invention provides a method for identifying abnormal states of a power distribution network line, which comprises the following steps: S1, identifying and determining an initial key node set based on distribution network topology layout analysis, distributing edge computing nodes for the initial key nodes, and constructing a distributed edge computing network; s2, an environment sensing mechanism is established based on the intelligent sensing system of the multiple sensors and the historical data, the operation environment data of the power distribution network is monitored in real time, the set of initial key nodes is dynamically corrected, and the updated key nodes are formed and the corresponding operation abnormality judgment threshold value is set; s3, collecting line operation data through an intelligent sensing system in a current monitoring period to form current period line information, wherein the current period line information comprises voltage, current, temperature and environmental parameters; S4, inputting the current period line information to a corresponding edge computing node, comparing and analyzing the operation data of each updated key node by utilizing the generated abnormality judgment threshold value, and outputting the node abnormal state identification of each updated key node in the current period; S5, collecting node abnormal state identifiers of the updated key nodes in different characteristic frequencies in the current period, and carrying out space-time aggregation analysis on discrete node abnormal information by combining a preset fluctuation deviation threshold value to generate an identification analysis result. Preferably, the step S1 comprises the steps of: s101, reading topological structure data of a power distribution network, and sorting connection relation and load distribution characteristics of line nodes; s102, screening out the matched line nodes from the line node connection relation according to the electrical distance centrality and the load importance, and sorting the line nodes into an initial key node set; S103, distributing corresponding edge compu