Otimização do filtro de kalman estendido aplicado na estimação de velocidade de um motor de indução trifásico

Abstract

This work consists in the study of the velocity estimation of a three-phase induction motor, because the information of this magnitude is extremely important for both the drive and the application of control methodologies. One way to estimate this speed is use the extended Kalman filter. However, the performance of the extended Kalman filter depends directly on the tuning of its covariance and prediction matrices, and these are traditionally determined by trial and error procedures. The aim of this work is to determine the values of these matrices by means of an optimization algorithm based on natural selection and social behavior of the population. The success allows the sensorless to replace the encorder in the control methodologies, increasing the economic advantage of the three-phase induction motor. For this, the entire structure of the motor, the sensorless and the optimization algorithm, besides the computational simulations were realized in the software MATLAB®.