ABSTRACT
For structured system identification, a large number of algorithms have been developed over the last few decades. The related studies can be found in references[1,2]. However, a generic solution to this problem has not been found yet for the system where /(•,•: •) and g(y : •) are nonlinear and 0 has a higher order. In this work, a multi-model hybrid neural-genetic algorithm is employed to find unique solutions satisfying with the condition for nonlinear systems identification.
