Neural Network-Based Fuzzy Predictive Current Control for Doubly Fed Machine

Abstract

In this paper, based on the radial basis function (RBF) neural network, a fuzzy predictive current control strategy for the doubly fed machine (DFM) is presented. The dynamic model of voltage, flux linkage, electromagnetic torque and mechanical motion equation for DFM are expressed. Because the DFM structure is complex and the DFM parameters are variable according to the operating conditions and environments, in order to improve the dynamic performances of DFM, the RBF neural network and fuzzy predictive control theories are employed to design the current controller in the DFM adjustable speed system. Simulation results show effectiveness of the proposed control strategy.