A New Minimum Torque-ripple and Sensorless Control Scheme of BLDC Motors Based on RBF Networks

In this paper, a new method based on adaptive radial basis function (RBF) networks is proposed to deal with the issues of rotor position requirement and high torque ripple production in a brushless DC (BLDC) motor. Two RBF networks are trained offline with a self-adjustment growing and pruning (GAP) algorithm, and all the train samples are obtained from experimental results to express the nonlinear characters of the machine more accurately. One of the trained networks is used to realize a nonlinear mapping between external voltages, phase currents and rotor position of the motor, at the same time the other is applied for estimation of phase current references with a desired torque. Actual phase currents are adjusted according to the references; therefore the torque ripples generated by non-ideal current waveforms is minimized for a BLDC motor without position sensors. Simulation results show the efficiency of this proposed method