一种新的小波阈值函数改进方法及实现

局部放电信号是一种弱信号,受周围复杂电磁环境的干扰较大,利用小波分析对它进行去噪处理是当前研究的一个热点.传统的小波阈值去噪方法主要包括硬阈值方法和软阈值方法,它们在实际应用中取得了很大的成功,但同时也存在一些需要改进的地方.文章在传统软、硬阈值函数的基础上,提出了一种新的双变量阈值函数,并分别利用软、硬阈值函数和新阈值函数对局部放电仿真信号进行了小波去噪分析,仿真结果表明,通过调节两个可变参数,新阈值函数可以获得具有最佳信噪比和最小均方误差的去噪信号,其去噪效果明显优于传统的软、硬阈值函数,而且更具灵活性.

A New Modified Method Based on Wavelet Threshold De-Noising Functions

Partial discharge signal is a kind of weak-signal, and it is easily to interference from the noise around. Wavelet threshold de-noising is an important application of wavelet analysis in partial discharge signal processing. The traditional threshold de-noising methods are hard threshold function and soft threshold function, the application of them in signals processing were successfully, but there were also some places need to be improved, such as soft threshold function losing some useful information of the signal, and hard threshold function is not continuous, and so on. So a new modified method based on wavelet de-noising functions is proposed. In this paper, three different threshold functions, soft threshold, hard threshold and the new threshold functions, were used in partial discharge signal de-noising. From the results of time-domain, frequency-domain, SNR and MSE, the simulation experimental indicates that the new threshold function can offer the best de-noising signal only by changing the variable parameters. More important, the new threshold function is more adaptability than the traditional threshold functions, and these advantages make it possible to construct an adaptive wavelet threshold de-noising algorithm.