Identification of nonlinear systems based on adaptive fuzzy systems embedding quasi-ARMAX model

This paper addresses the issues related to identification of nonlinear systems based on adaptive fuzzy systems (AFSs) embedding quasi-ARMAX model. A general nonlinear system can be represented as a quasi-ARMAX form, in which the parameters a¯_i and b¯_i contain two parts: linear part and nonlinear part. By introducing an AFS to the nonlinear part of each parameter, we propose an AFS embedding quasi-ARMAX model for identification of nonlinear systems. Since the quasi-ARMAX model consists of a combination of AFSs and ARMAX model, a priori information about system structure besides input-output data can be incorporated into the identification. Simulation results show that the proposed quasi-ARMAX model has better performances than a neural networks model