On the determination of an optimal state observer gain for multivariable systems : application to induction motors

Abstract EN

This paper deal with the determination of the optimal state observer gain matrix for multivariable dynamical system.

In this way, it has been shown that an improved optimal gain can be reached by minimizing a quadratic criterion formulated as a quadratic output feedback control of the observing error system.

The proposed method or direct observer gain optimization can be applied either for the full order observation error system or for the reduced order one for the synthesis of the a partial state is non measureable.

The importance of proposed criterion for the synthesis of the full and the reduced state observers has been tested by a simulation study of the state observation of an induction motor (IM) which has highlighted the effectiveness of the developed method in relation to that optimizing the dual system.