On identification and stabilization of nonlinear plants using fuzzy neural network

This paper is concerned with the stable tracking of a nonlinear plant using fuzzy clustering of ARMA models. The nonlinear input-output model is identified as fuzzy clusters of local ARMA models using a fuzzy neural net (FNN). It is thoroughly investigated whether the overall system can be made stable by stabilizing the individual linear subsystems. Two nonlinear systems have been taken and the controller is designed based on stabilizing each fuzzy region. The overall controller is determined by fuzzy blending of the individual controllers and it is shown that the systems are stable. Also, the overall system tracks a desired trajectory if all the individual subsystems track the same.