parviz moradi - sahand university of technology, Tabriz, Iran
mohammad reza azizian - sahand university of technology, Tabriz, Iran
mohammad javad khosrowjerdi - sahand university of technology, Tabriz, Iran,

In this paper, a MODIFIED EKF(MEKF) method is proposed that improves the EKF’s previous methods. In the EKF previous methods usually, the measurement noise covariance matrix (R) is assumed to be persistent and calculated by trial-and-error methods, but various environmental factors such as sensor faults are efficient on the value of R, so we have to update R. In this method we use a genetic algorithm to update R to get accurate estimates. Simulation results show the effectiveness of the proposed method

induction motor - state estimation - control method without speed sensor - parameter estimation -Extended Kalman filter-genetic algorithm