The rectangular fluctuation control of improved direct torque controlled permanent magnet synchronous machines based on RBF network

In order to solve to the problem of serious torque and flux ripple in motor direct torque control, an amendatory direct torque control method based on the radial basis function neural network technology was presented in this paper. This method can reduce the torque and flux ripple in static run and enhance the performance of low speed. The simulation experimental results indicate that this method may be feasible alternative and high robust capabilities.