Recursive identification of nonlinear two-dimensional (2-D) systems

In this paper a recursive identification method is proposed for two-dimensional nonlinear time-varying systems for the first time in the literature. No structural information, mathematical model, external testing signals and training process are needed. It is designed only by using I/O data of the systems. The algorithm is based on a dynamic linearization technique and is proposed for a class of two-dimensional discrete-time single-input single-output nonlinear systems. The algorithm uses the pseudo-partial-derivatives derived on-line from the input and output information of the system using a recursive projection type of identification algorithm. Numerical simulations are presented to show the effectiveness of the proposed algorithm for the estimation of temperature of catalytic rod during process in a reactor.