The wear of bearing and gear surface is the most common cumulative failure of wind gearbox,which will seriously threaten the normal operation of the whole wind turbine,and even greatly reduce its service life when the cumulative wear of wind gearbox reaches a certain amount.The maintenance strategy of wind turbine was studied based on the spectrum analysis technology.The maintenance strategies were established according to the different cumulative wear of each device and based on the operation and maintenance criteria of wind turbine gearbox system.The time cost model of gearbox systems was established based on probability model.With the minimum maintenance time cost as the optimization goal,the Monte Carlo method was used to simulate the random degradation process of the equipment.The simulation shows that method is feasible and effective.