基于光纤传感检测的局部放电信号去噪方法

Denoising Method of Partial Discharge Signals Based on Optical Fiber Sensing Detection

针对局部放电超声信号检测过程中存在噪声干扰的问题,提出一种基于变分模态分解和改进小波去噪的噪声抑制方法。首先对局部放电信号进行变分模态分解,得到若干频带分布不同的本征模态分量,结合模态的峭度剔除窄带干扰主导的分量;然后利用改进的小波去噪法,将其余有效模态分量中残留的白噪声进一步滤除;最后将剩余分量进行信号重构实现噪声抑制。对仿真局部放电信号去噪的结果表明,所提方法可提升信噪比约12.1 dB,且能够更加完整地保留原始信号波形。此外,在实验室环境下搭建了10 kV电压等级的局部放电模拟系统,采用Sagnac光纤传感检测局部放电超声信号。实验结果表明,所提方法具有良好的去噪性能,对实测信号的噪声抑制比达到了7.504。

In order to solve the problem of noise interference in the process of partial discharge ultrasonic signal detection,a denoising method based on variational mode decomposition and improved wavelet denoising is proposed.Firstly,the PD signal is decomposed by u-sing VMD into intrinsic mode components with different frequency band distributions,and the components dominated by narrow-band inter-ference are removed according to kurtosis.Then the residual white noise in the PD-dominant components is further denoised by using im-proved wavelet denoising algorithm.Finally,the components are reconstructed into a denoised signal.The results of denoising the simulated partial discharge signal show that the proposed method can improve the signal-to-noise ratio of the noise signal by about 12.1 dB,and can preserve the original signal waveform more completely.Besides,a partial discharge simulation system with 10 kV voltage level is built in the laboratory environment,and Sagnac optical fiber sensing is used to detect the partial discharge ultrasonic signal.The experimental results show that the proposed method has good denoising performanceꎬand the noise rejection ratio of the measured signal reaches 7.504.