Optimizing Extended Kalman Filter for Speed Sensorless Control of Induction Motors Using Artificial Bee Colony Algorithm

Abstract

Implementing sensor less control in induction motor drives has significantly reduced expenses and hardware weight while enhancing reliability. Among the highly effective speed sensor less control techniques, the Extended Kalman Filter (EKF) is distinguished by its superior estimation accuracy. However, the efficiency of the EKF relies on the accurate determination of noise covariance matrices. Recently, many studies have aimed to optimize these matrices to enhance performance. In this research, we introduce a unique method for speed estimation in an induction motor drive using the Artificial Bee Colony (ABC) algorithm to maximize the performance of EKF. The effectiveness of the proposed method is validated by the Matlab/Simulink simulation of a constant Voltage/Hertz controller-based drive system at different operating conditions.