Aiming at solve the problem of the power quality influence by wind power integration, the integration characteristics and fault characteristics of wind power station is the thorough analysis, a novel method is proposed in this paper, the method makes combination with harmonic wavelet noise reduction and time-frequency joint analysis, and it is used for voltage fault signal feature extraction of wind power integration, this method has solved the problems that the traditional Fourier transform can′t process the fault testing signal with non-stationary character. The method makes full use of harmonic wavelet noise reduction technology for purification of non-stationary voltage fault test signal in strong noise, and then the time-frequency combined analysis is used to accurately locate the voltage fault of wind power integration, the method is able to provide necessary reference decision for fault crossing. Finally, the effectiveness of the method is verified by simulation and experiment.