Abstract: A new method to Power Quality Disturbance（PQD） signal de-noise is proposed.It adaptes the rapid ascent wavelet to decompose and reconstruct PQD signals，introduces adaptive algorithm to the design of update operators and predict operator，improves the swiftness of this algorithm and well satisfies the instantaneity requirement of PQD detection.Comprehensively considering the advantages of soft and hard threshold de-noising methods，it puts forward a weighted threshold de-noising algorithm，so that this algorithm can simultaneously satisfy de-noise of transient and steady-state PQD signal.Simulation results show that，this algorithm can improve the SNR of de-noised signal，reduce the MSE and the execution time，and is very suitable for the DSP realization，has certain practical value.

Key words: de-noise, power quality, ascent wavelet transform, adaptive algorithm, weighted threshold

摘要： 提出了一种新的电能质量扰动（Power Quality Disturbance，PQD）信号去噪方法。采用快速提升小波变换进行PQD信号的分解和重构，将自适应算法引入到更新算子和预测算子的设计中，提高了算法的快速性，满足了PQD检测实时性要求。综合软、硬阈值去噪的优点，提出了加权阈值去噪的概念，使得算法能同时很好地满足暂态PQD信号和稳态PQD信号的去噪。仿真结果表明，该算法能够提高去噪后信号的信噪比SNR，降低去噪后的均方误差百分值MSE，降低了算法的执行时间，且适宜于算法的DSP实现，具有一定的实用价值。

关键词: 去噪, 电能质量, 提升小波变换, 自适应算法, 加权阈值