High-voltage circuit breaker fault diagnosis method based on multi-feature optimization fusion

Abstract

The invention discloses a high-voltage circuit breaker fault diagnosis method based on multi-feature optimization fusion. The method comprises the steps of: acquiring multiple groups of initial soundsignals YS(t) and vibration signals YZ(t) emitted in the switching-on and switching-off process of a circuit breaker; denoising, weighting and fusing each group of initial sound signals YS(t) and vibration signals YZ(t) to obtain a group of sound signals YS''(t) and vibration signals YZ''(t); extracting a feature vector of the sound signal YS''(t) by adopting a K-S test method; extracting a feature vector of the vibration signal YZ''(t) by adopting an EMD empirical mode decomposition method; fusing the feature vector of the sound signal YS''(t) and the feature vector of the vibration signal YZ''(t) into a feature matrix sample; selecting sample feature matrixes under various operation conditions of normal and fault states from the feature matrix samples in proportion; and inputting the selected sample feature matrixes to obtain a fault diagnosis result based on an FWA optimized SVM fault diagnosis model. According to the invention, through the concept of multi-sensor data fusion, the noise is effectively eliminated through weighted fusion of the data, so that the data is closer to a real value, and the fault diagnosis effect is effectively improved.

Description

technical field [0001] The invention belongs to the technical field of fault diagnosis of electrical equipment, and in particular relates to a fault diagnosis method for a high-voltage circuit breaker with multi-feature optimization fusion. Background technique [0002] As an important switching electrical equipment in the power system, the high-voltage circuit breaker ensures safe and stable operation and is a necessary condition for improving the reliability of power supply in the power system. The fault diagnosis process of a circuit breaker mainly includes three steps: signal acquisition and processing, feature vector extraction, and fault identification. Among them, selecting an appropriate method to process the operating signal of a circuit breaker and extracting feature vectors is a key part to improve fault diagnosis results. [0003] At present, the signals collected by the high-voltage circuit breaker fault diagnosis research institute are roughly divided into: ope...

AI Technical Summary

Problems solved by technology

In the combined sound-vibration fault diagnosis of high-voltage circuit breakers, the feature-level fusion of multiple energy entropy combinations is usually used, or the D-S evidence theory is used for decision-level information fusion, but for the data-level fusion of signal source information processing less research

Images