A Compensation Strategy of Flux Linkage Observer in SPMSM Sensorless Drives Based on Linear Extended State Observer

Flux linkage estimation based on the voltage model has been widely employed in surface-mounted permanent magnet synchronous machine (SPMSM) sensorless drives due to its simple implementation. To deal with dc-offset and dc-drift problems, the pure integrator is modified as a programmable low pass filter (LPF)in practice. However, the estimation accuracy of the rotor position is affected by the adopted filter. Although the position deviation is compensated in steady-state based on the frequency characteristics of the filter, the compensation effect deteriorates during a dynamic process which severely degrades the sensorless control performance. In order to effectively tackle this problem, a novel compensation strategy based on the linear extended state observer (LESO) is proposed in this paper. With the constructed LESO in the estimated synchronous rotating reference frame, the position estimation error can be detected and compensated both in the steady and dynamic state. Moreover, the observer is enhanced to improve the estimation accuracy in linear acceleration and deceleration. Finally, the feasibility of the proposed method is verified under various operating conditions with a 5.5 kW SPMSM drive system.