On-line identification methods of parameters for permanent magnet synchronous motors based on cascade MRAS

Motor parameters should be on-line estimated to realize precise control of PMSM in sensorless vector control system. In this paper, an on-line identification method for PMSM parameters based on cascade MRAS is proposed by analyzing the conventional MRAS. By means of Popov's hyper-stability theory, the model of motor parameters identification is built in synchronous d-q coordinates, and PMSM stator voltage, stator current and their errors are used to obtain the adaptive laws of motor parameters, and it is realizable to estimate rotor speed, stator resistance and rotor flux at the same time. The simulation results demonstrate the correctness and effectiveness of the proposed method.