In order to obtain the functions of monitoring wind turbine’s working state and the task of intelligent diagnosis of these state data, an intelligent monitoring system for wind turbines is developed. Firstly, based on the vibration theory, the dynamic model of gearbox is established, and the data characteristics in condition of various fault types are analyzed, which provides identification basis for intelligent diagnosis. Then, the full probability formula of Bayesian network is analyzed by using Bayesian network theory, and on this basis, the research is carried out and an intelligent diagnosis way is proposed which consists of simplifying generalized inverse matrix. Finally, the intelligent monitoring system based on generalized inverse matrix is constructed. The system is tested on the spot. The results show that the output of the intelligent monitoring system is consistent with the results of artificial diagnosis. The system has been preliminarily applied successfully in the wind field. This work is supported by National Key Research and Development Program of China Grid-friendly Control Technology for Large-Capacity Wind Turbines Project (No. 2018YFB0904000).