Abstract: The study of gearbox health assessment can predict the failure time of the wind turbine in a timely and accurate manner, and provide guidance for subsequent maintenance and repair. In this paper, we propose a gearbox health assessment method based on adversarial network. Through the change of normal condition and fault condition, the fault is detected early, the residual between the original signal and the predicted signal is used to indicate whether the gearbox operating state is normal, the deviation from the normal data to the abnormal data is quantified through the deviation index and the technique of change, and it is converted into a probability value to obtain the health index of a single variable, and the overall health index(HI) of the gearbox is composed of a single variable index. Experiments show that this method can realize the health assessment of wind turbine gearbox and has a broad application prospect.