High-voltage breaker defect diagnosis based on wavelet packet energy

Abstract

The invention relates to a high-voltage breaker defect diagnosis based on wavelet packet energy. The method comprises the step shat 1) a monitored current signal is extracted coarsely to screen waveform areas with significant characteristic, and de-noising is carried out to obtain a relatively pure opening / closing current waveform; 2) wavelet packet three-layer decomposition is carried out on an original signal with characteristic information to extract an original characteristic vector composed of different frequency bands of energy; 3) dimension reduction is carried out on the original characteristic vector via major constituent analysis, a sample difference characteristic is focused, and the data is more separable; and 4) data with a label is input to a support vector machine, and a high-voltage breaker defect discrimination model is constructed. Thus, equipment defect conditions can be checked effectively, the method can meet different requirements of high-voltage breaker defect diagnosis, and the classification precision and detection precision of high-voltage breaker defects are high.

Description

technical field [0001] The invention relates to the monitoring and control technology of power transmission and distribution equipment, in particular to a high-voltage circuit breaker defect diagnosis method based on wavelet packet energy. Background technique [0002] High-voltage circuit breakers have become an indispensable and important part of my country's power grid due to their large number, wide application range, and control and protection functions for switching loads and removing faults in the power system. Therefore, in order to avoid the occurrence of various accidents, further ensure the safety and stability of the system, reduce the operation and maintenance cost of the smart grid, and promote the development of new technologies for the grid, it is necessary to broaden the new ideas of high-voltage circuit breaker feature extraction and promote fault (defect) diagnosis technology. It is very necessary to keep pace with the times and provide more reliable diagn...

AI Technical Summary

Problems solved by technology

[0004] The above research results are verified by examples and have certain theoretical and practical value. However, with the advancement of science and technology, there are many types of circuit breakers, and most of the feature extraction methods are aimed at specific types of circuit breakers. The diagnostic methods are not universal. Adaptability, unable to meet the condition monitoring of various types of circuit breakers

At the same time, in recent years, the operating mechanism, as the component with the highest mechanical failure rate in circuit breakers, is the key object of condition monitoring. The purpose of condition monitoring is not only to accurately judge mechanical defects and failures, but also to pay attention to the trend of defect state changes. Therefore, the severity judgment and fault (defect) diagnosis method for circuit breaker operating mechanism defects have long been difficult to solve and optimize.

Images