Variable pitch control of wind turbine based on RBF neural network

Traditional variable pitch controller is difficult to achieve good results in rapidly variable wind, in order to improve the dynamic performance of the system and achieve constant power output, this paper designed a variable pitch backstepping sliding-mode controller(BSMC) of wind turbine based on radial basic function neural network(RBFNN). The scheme conducted the original nonlinear system model of the global linearization first, then on the basis of radial basic function neural network to approach uncertainties, combined sliding mode control with backstepping method to designed backstepping sliding-mode controller. The designed controller guaranteed the stabilization of wind turbine in high speed, restrained the influence of uncertainties to the system, avoided the explosion of complexity in traditional backstepping method. Compared with traditional sliding mode controller, the results of the simulation indicated that the designed controller can stabilize the output power of wind turbines and improved the performance of wind turbines variable pitch system effectively.