基于小波的电力系统信号预定阈值去噪技术

Wavelet-Based Predefined-Thredhold Denoising Technique to Power System Signal

目前使用小波对电力系统信号去噪时一般是在信号处理的过程中求取阈值,这样就会影响到信号的实时处理速度和阈值的准确性。本文提出的方法是在电力系统无暂态干扰的情况下预先确定阈值然后再对电力系统信号去噪。由于对电力系统信号去噪时不进行求取阈值运算,这就提高了实时处理信号的速度;同时在预先求取阈值时除了工频信号,其他信号都可看成是噪声,这就提高了求取阈值的准确性,而且由于是预先非实时求取阈值,就可以对信号进行较复杂的运算求取更符合要求的阈值。此方法经过仿真和实际应用,在去噪效果和运算速度上都有较大提高。

When wavelet transform is used to denoise power system signal, the thresholds are usually calculated with processing signal, which will affect the speed of processing and the accuracy of thresholds. The technique is to denoise power system signal after predefining the threshold while no transients in power system. As thresholds need not to be calculated in denoising, the speed to process signal real-time is improved. At the same time when threshold is calculated, other signal can be regarded as noise except for 50Hz signal, which improve the accuracy of thresholds and since calculating thresholds is not gained atreal-time, better threshold can be calculated using more efficient methods. Simulating and application in practice prove this technique improves denoising effect and have more quickly calculating speed than others.