Lightning arrester fault diagnosis method and system based on improved K nearest neighbor algorithm

Abstract

The invention relates to a lightning arrester fault diagnosis method and system based on an improved K nearest neighbor algorithm. The method comprises the following steps of: 1, acquiring online monitoring data of a lightning arrester with a known equipment state, and obtaining equipment operation state and a multi-dimensional characteristic quantity; 2, performing standardization processing on the data to obtain data subjected to standardization processing, and performing dimension reduction processing on high-dimensional data to obtain visual data; and 3, importing obtained various types of data sets into an improved K-nearest neighbor algorithm established by software to carry out algorithm training, and carrying out algorithm optimization by taking the recognition accuracy as a target quantity to obtain an optimal solution of algorithm fault recognition; The fault diagnosis method is simple and convenient, high in operation efficiency and good in fault identification accuracy.

Description

technical field [0001] The invention belongs to the field of lightning arresters in electric power systems, and in particular relates to a fault diagnosis method and system for lightning arresters based on an improved K-nearest neighbor algorithm. Background technique [0002] Surge arresters are very important electrical equipment in power systems. As the core components to protect power equipment from overvoltage, whether they can operate safely and stably affects the reliability of the power grid to a certain extent. The arrester online monitoring can understand the operation performance of the arrester in real time without power failure, discover abnormal phenomena and accident hazards in time, make correct judgments on the detection results and take preventive measures to prevent economic losses caused by accident expansion and ensure that it is in good condition. run. [0003] Due to many interference factors, such as the influence of weather, damp valves, etc., the o...

AI Technical Summary

Problems solved by technology

[0003] Due to many interference factors, such as the influence of weather, damp valves, etc., the online monitoring system of arresters has a large number of false alarms, which makes the grass-roots staff exhausted. This not only increases the workload of the grass-roots staff, but also increases the stability of the arrester. risk of running

Due to the threshold criterion, the existing lightning arrester online monitoring and early warning rules are easily affected by interference factors and cause false alarms. Therefore, the analysis and research on the historical data of the lightning arrester online monitoring and the establishment of a reasonable and effective algorithm model are important for improving the accuracy of the lightning arrester online monitoring alarm. It is of great significance to ensure the safe and reliable operation of arresters in the power grid

There is no technical report on this aspect

Images