A non-convex optimization method for identification of cascaded nonlinear systems

Identification of cascaded dynamic nonlinear systems has received more and more attention since it can be applied to characterize a wide class of engineering problems. However, this problem is inherently non-convex, and few efforts have been devoted to identifying these systems. In this paper, we propose a novel non-convex optimization method for identifying the cascaded nonlinear systems. The identification problem is modeled as a multi-objective consensus problem. Numerical simulation shows that the proposed algorithm is able to alleviate the non-convex characteristic and identifying the nonlinear dynamic cascade systems.