Research on denoising method of mechanical vibration signal of circuit breaker based on sparse decomposition

摘要

In the process of collecting the vibration signal of the circuit breaker, a certain degree of noise is mixed and the quality of the signal is reduced. In this paper, a method of denoising mechanical vibration signal of circuit breaker based on sparse decomposition is proposed. Firstly, the Gabor over complete dictionary is constructed. And then the matching pursuit (MP) algorithm based on fast fourier transform (FFT) is used to sparse decomposition of the signal containing noise. The useful information part of the signal is extracted and the residual component in the signal is discarded as the noise part. Then the signal is reconstructed using the linear combination of the best atoms, so that the denoising signal approximates the original signal to achieve the purpose of signal denoising. Through the analysis of the simulation signal and the measured signal, the method proposed in this paper realizes the ideal denoising effect and is helpful to improve the fault diagnosis precision of the circuit breaker.