Adaptive Discrete Extended Kalman Filter for Parameter Estimation in IRFOC-Based IMD with Optimized Gains

In this work, an Adaptive Discrete-time model Extended Kalman Filter technique for simultaneous multi-parameters estimation of induction motor drive is proposed. An Ant-lion optimization algorithm for fine-tuning the PI error regulator gains in indirect rotor flux-oriented control (IRFOC) with space vector pulse width modulation switching technique is implemented. The measure of parameters improves the operation of the drive speed and torque performance, based on proper tuning of slip angular frequency. The effectiveness of the standard discrete Extended Kalman Filter technique is influenced by the choice of system and measurement error covariance noise matrices. These matrices are evaluated using the trial-and-error process and are intended to be consistent. However, the operating environment has an impact on these covariance matrices. To improve monitoring reliability and avoid the time consumption by trial-and-error approach for selecting optimum matrices, an adaptive estimating strategy with the capacity of online matrix updating is proposed. Further, to enhance the effectiveness by eliminating the time required for manual tuning of gains of PI in IRFOC are obtained by the Ant-Lion optimization (ALO) technique. The avoidance of local optimum is an additional aspect of ALO. The proposed technique's feasibility and performance are evaluated and verified under various operational scenarios. The simulation results demonstrate the adopted ADEKF technique has a faster response and thus can precisely estimate the parameters of stator resistance, rotor resistance, stator inductance, and rotor inductance.