基于误差和能量值的瞬态非平稳信号分析方法

摘要

Wavelet analysis method has been widely studied and applied because it can effectively deal with transient nonstationary signals.However, for the original signal's own characteristics, how to choose the best wavelet basis function to improve its feature extraction and diagnostic performance is still rare.Based on the inherent characteristics of the original signal and wavelet basis features,combined with wavelet reconstruction error and wavelet frequency domain energy value method,a wavelet basis selection and analysis method was proposed.The problem of transient nonstationary signal analysis and processing accuracy caused by improper selection of wavelet bases was solved.In the experiment,the facial electromyography signal was used for wavelet analysis and the wavelet basis Db 4 was finally selected as the optimal wavelet basis for the analysis of the signal.The results showed the effectiveness of the method.%小波分析方法能够有效处理瞬态非平稳信号而被广泛研究与应用.然而,针对原始信号自身特性,怎样选择最佳小波基函数,从而提升其特征提取与诊断性能的研究尚属少见.从原始信号的固有特性和小波基特征出发,融合小波重构误差与小波频域能量值方法,提出一种小波基选择与分析方法,解决了因小波基选择不当而引发的瞬态非平稳信号分析与处理精度低的问题.实验采用面部肌电信号进行小波分析测试,最终选择小波基Db4作为分析该信号的最优小波基,结果表明了该方法的有效性.