Optymalizacja rozmytego filtru Kalmana przy wykorzystaniu algorytmów genetycznych

This paper deals with a problem of a fuzzy Kalman filter optimization for an adaptive control structure of a two-mass drive system. In the introduction section of this paper modern control structures of electrical drives with elastic joint are discussed. For their application state variables estimation of a dynamic object are required. A particular attention is given to the observers based on the Kalman filter theory. Additionally, the problematic aspects of the observer applying design by using genetic algorithm are considered. Moreover, the remainder of the paper is organized as follows. First, the object of research and its mathematical model are described. Subsequently, genetic algorithms are discussed. Next, the optimization process of the fuzzy Kalman filter using genetic algorithms is presented. The simulation results of the drive performance in a open and closed-loop control structure are shown.