Nonlinear Flux Observer Based on Extended Flux Model for Sensorless Control of IPMSM in Electric Vehicles

Sensorless control of interior permanent magnet synchronous motors (IPMSM) is indispensable for electric vehicles to guarantee safety and reliability. However, the performance and stability will be severely deteriorated caused by signal noise and time-varying model, especially in the low-speed region. Therefore, this paper proposes a nonlinear flux observer based on extended flux model, utilizing gradient descent method with cascaded second-order generalized integrators (SOGI), which is less sensitive to inaccuracy of initial value and eliminates the negative effect of DC disturbance and harmonic fluctuation. Validity of the proposed sensorless control system is demonstrated through a set of simulations under various conditions.