Circuit breaker fault diagnosis method for detecting and extracting current signal based on sliding window

Abstract

The invention relates to a circuit breaker fault diagnosis method for extracting a current signal based on sliding window detection, and the method comprises the following steps: S1, extracting a key time point and a corresponding current value in a current curve of an opening and closing coil of a circuit breaker through employing a sliding window detection method, establishing a comprehensive characteristic sample set of the opening and closing current by combining the duration time of the working stage of the coil and the overall time domain statistical characteristic of the current signal; s2, performing feature dimension reduction on the comprehensive feature sample set of the opening and closing current by using a principal component analysis method, and selecting a principal component according to the accumulated contribution degree to obtain a comprehensive feature sample set after dimension reduction; and S3, inputting the comprehensive feature sample set after dimension reduction into a PCA-SVM model, optimizing the model by adopting a K-fold cross validation method and a grid search method GS, and performing performance verification on the optimized model. The method can effectively improve the diagnosis precision and the diagnosis efficiency, and can be adaptive to fault identification of various types of circuit breakers.

Description

technical field [0001] The present application relates to the technical field of fault diagnosis of high-voltage circuit breakers, in particular to a circuit breaker fault diagnosis method based on sliding window detection and extraction of current signals. Background technique [0002] The high-voltage circuit breaker is an important key device that plays a role in protection and control in the power system. Diagnosing its operating status is of great significance for maintaining the safe operation of the power system. The current signal of the opening and closing coil of the high-voltage circuit breaker can reflect the working conditions of the circuit breaker's iron core, excessive iron core air travel, short circuit between coil turns, etc., and is often used to evaluate the working state of the circuit breaker. [0003] In the previous research on signal extraction of circuit breaker working state, the key point features of the current waveform of the circuit breaker's ...

AI Technical Summary

Problems solved by technology

However, in the above research, the circuit breaker feature extraction cannot adapt to various types of circuit breakers, and the accuracy of the traditional fault classification model is not ideal, resulting in low diagnostic accuracy and efficiency

Images