Discrete-Time Recurrent Neurofuzzy Network for Identification of Nonlinear Systems

Motivated by the research works in adaptive observers, this paper presents a structure for black-box identification based on state-space recurrent neural networks for a class of dynamic nonlinear systems in discrete-time. The proposed network catches the dynamics of the unknown plant by generating state estimates of a network and jointly identifying its parameters using only output measurements. The stability of the network, the convergence of the training algorithm and the ultimate bound on the identification error as well as the parameter error are established. Numerical examples using simulated and experimental systems are included to demonstrate the effectiveness of the proposed method.