强背景噪声下电晕放电辐射信号小波阈值去噪和分析

分析电晕放电信号小波阈值去噪中小波基函数的最优选取、分解层数及阈值的确定,并对实测电晕放电信号进行小波阈值去噪。结果表明,选取db8小波作为最优小波基函数,选取双变量阈值函数作为小波阈值函数对电晕放电信号进行小波阈值去噪,放电辐射信号与背景噪声可以得到有效分离,具有很好的去噪效果。

Wavelet Threshold De-noising of Signal Radiated from Corona Discharge under Complicated Background

The concept of information loss evaluation factor is first introduced on the foundation of some contents of wavelet threshold de-noising, such as the best selection of wavelet bases, decomposition scales, threshold values and threshold functions. It offered the basis for further quantitative analysis of wavelet bases selection. According to the method of best wavelet bases selection and the new modified threshold function, the signal gathered from the mock test system is de-noised with wavelet. From the comparison of spectrum and time-frequency spectrum between the signals before and after de-noising, you can see that the signal has successfully been extracted from the background noise.