| The sensorless drive system based on the interior permanent magnet synchronous motor(IPMSM) has been widely used in some operation fields, such as high speed motor, aeronautics and astronautics and household appliances, for its advantages——high reliabilty and low cost. However, the sensorless control system has the problems in the wide speed range because of rotor position estimation methods restrictions. Therefore, IPMSM rotor position estimation has been studied in the wide speed range and higher rotor position estimation precision, which have great theoretical significance and appliance value in IPMSM sensorless technology.This dissertation further studies and analyzes the IPMSM rotor position estimation in the standstill, sensorless technology in the low speed and high speed region and compound sensorless thenology in the different rotor position estimation methods in the wide speed range based on the IPMSM.Firstly, based on the traditional mathematical model of IPMSM, the thesis presents the mathematical model including cross-coupling magnetic saturation. The variation of incremental self-inductance and cross-coupling inductance is evaluated by the finite element method to analyze the relation of current and rotor position with d-axis and q-axis. The variation range of inductance parameter is determined. Stator current trajectory under maximum torque per current control and the speed range of field weakening zone are obtained based on thoroughly analyzing inductance change.Secondly, a rotating high frequency voltage injection method is used to estimate the rotor position. The mathematical model of high frequency current response is established based on considering IPMSM cross-coupling effect in the range of the low speed and even standstill. Stator current is demodulated by bandpass filter and synchronous reference frame filter to extract high frequency negative sequence current component; The phase locked loop(PLL) is used to estimate rotor position and speed. According to direct axis flux saturation, the magnetic polarity is identified by using magnetic saturation model. The estimated angle errors, which result from magnetic cross-coupling and cascade filter of the current demodulation, are compensated.Thirdly, in the stationary frame, IPMSM state equation is established based on linear flux, on the basis of which rotor position is estimated by the extended state sliding mode observer. The second order non-singular terminal sliding mode control is put forward to improve response speed and suppress the chattering in combination with the advantages of the high order sliding mode and non-singular terminal sliding mode. The observer is stable with Lyapunov theory and the adaptive of sliding mode gain matrix is obtained. In constant torque and weakening field, the effect of deviation of stator resistance, quadrature inductance and estimated angle speed on sliding mode observer estimated flux amplitude and phase is analyzed quantitatively. Rotor position and speed are obtained by PLL.Next, in the wide speed range, the rotor position is estimated by blending the two methods—rotating high frequency voltage injection and extended state sliding mode observer. The rotor position is estimated by rotating high frequency voltage injection in the low speed range while it is estimated by extended state sliding mode observer in the high speed range. Thus, it is showed that IPMSM is in stable operation from standstill to field weakening zone.Finally, the experimental platform of sensorless IPMSM full-digital vector control system is designed. Hardware circuit and algorithm of the experimental platform have been implemented. IPMSM rotor position estimation, based on the rotating high frequency voltage injection, the extended state sliding observer and the transition between them, has been analyzed. And the experiment results show that this algorithm is feasible.
|
PDF Full Text Request