Improving the Dynamic Behavior of Sensorless Control for Induction Machines Using Magnetic Anisotropies

Abstract:
The following paper proposes a concept to improve a method to estimate the anisotropy field angle of an induction machine using continuous, sinusoidal test signals. The dynamic response and robustness during load steps is increased by using a simple feedforward structure. The proposed method is based on a main-flux oriented machine model to achieve better results compared to the usual approaches in terms of dynamic behavior. Rotor speed information is estimated by using a feedback loop inside the flux-model. Simulation results are shown to prove the viability of the proposed concept. Measurements from a test stand further show the performance on a real machine for tracking and sensorless control.