Taking the advantage of Bayesian network in dealing with the issue of uncertainty，this paper putsforward a fault diagnosis method based on Bayesian network for wind turbines according to its strong uncertainty offault feature information such as coupling and ambiguity． We focus on two key problems of this method，namelyBayesian network modeling and Bayesian network inference of fault diagnosis of Wind Turbines，building three lay-ers of CME Bayesian network model based on the fault tree analysis method and analytic the fault reasoning processof Bayesian network． Through an example of the wind turbines gearbox fault diagnosis，we verify the feasibility ofthe model and the effectiveness of Bayesian network inference． Ｒesearch on electromechanical equipment fault diag-nosis with associated uncertainty problems has referential significance．