In order to reduce noise in electric energy quality disturbance signals

a wavelet de-noising algorithm based on adaptive decomposition level and threshold is proposed. The algorithm adaptively determine number of optimal wavelet decomposition levels by calculating peak-to-sum ratio of the wavelet detail coefficients and according to distribution characteristic of useful signals and the noise signals in detail coefficients of each levels and the ratio of peak value of the negative and positive of the detail coefficients

dynamically adjust upper and lower thresholds of the detail coefficients of each levels. The transient oscillation and pulse signals are de-noised by using Matlab

and compared with conventional hard

soft threshold algorithm and an improved wavelet threshold algorithm. The results show that number of adaptive decomposition level and the proposed threshold wavelet denoising algorithm is superior to the other three methods in terms of signal-to-noise ratio（ SNR） and root mean square error（ RMSE） and the reconstructed signal is closer to the original signal

and better preserves the characteristic information of the signal during the disturbance period.