Initial rotor position estimation of salient-pole brushless DC motors by artificial neural network

This paper presents an initial rotor position estimation method for salient-pole brushless DC motors (BLDCMs). An artificial neural network (ANN) is employed as an estimator of BLDCM at standstill. The ANN can express the relationship between the voltage and current of BLDCMs. The initial rotor position can be estimated by using only the current-signals from the ANN. The proposed method has the advantages of measurement noise immunity as well as no sensitivity to machine parameters. Computer simulation results verify the usefulness of the proposed method.