An adaptive generalized sliding mode observer is proposed for the state estimation and actuator fault reconstruction of a wind power system. By constructing an augmentation system, the actuator fault becomes a part of the system state. A generalized sliding mode observer is designed to estimate the state of the augmentation system to obtain the required fault information. Given that the upper bound of fault is unknown in the actual wind power generation system, an adaptive algorithm is designed to estimate it. Taking a 5 MW wind turbine as the research object, a simulation is carried out on the Matlab/Simulink platform, and the results show that the proposed method can accurately estimate the system state and fault. This work is supported by the National Natural Science Foundation of China (No. 51477109 and No. 61703296) and the Scientific Research and Practical Innovation for Postgraduates in Jiangsu Province (No. KYCX19_2016).