Rotor Resistance Estimation of Induction Motors with A Novel Innovation-Based Adaptive Extended Kalman Filter for Self-Tuning

In this study a novel estimator is developed to identify the rotor resistance of the induction motor (IM) at standstill for self-tuning. For this purpose, an innovation-based adaptive extended Kalman (IAEKF) filter estimator is designed. IAEKF provides a more dynamic estimation compared to the conventional extended Kalman filter (EKF), as they have a mechanism where the system noise covariance matrix can be updated continuously, unlike conventional EKF. To increase estimation stability and also for position and amplitude information of the motor flux required for the dynamic control methods, stator stationary axis (-$\alpha\beta$) components of stator current and $-\alpha\beta$ components of stator flux are estimated with rotor resistance by using the correlation between states and parameters defined as nonlinear inputs. The estimation performance of the proposed IAEKF algorithm is tested both in the simulation and on the real-time IM experimental setup at standstill. Simulation and real-time results show that the estimation achievement of the proposed IAEKF algorithm is quite impressive.