On-line Identification Methods for the Parameters of PMSM Based on Cascade MRAS

Motor parameters should be on-line estimated to realize precise control of PMSM in sensorless vectorcontrol system. With the deep analysis of MRAS,an on-line identification method for motor parameters based oncascade MRAS was proposed. By means of Popov′s hyper-stability theory,the model of motor parameters identificationwas built in synchronous dq coordinates,and PMSM d,q-axis voltage,current and their errors were used to obtain theadaptive laws of motor parameters,and it was realizable to estimate rotor speed,stator resistance and rotor flux at thesame time. The simulation results demonstrate the correctness and efficiency of the proposed method.