A Nonparametric Identification Technique for Nonlinear Dynamic Problems

Abstract

A nonparametric identification technique is presented that uses information about the state variables of nonlinear systems to express the system characteristics in terms of orthogonal functions. The method can be used with deterministic or random excitation (stationary or otherwise) to identify dynamic systems with arbitrary nonlinearities, including those with hysteretic characteristics. The method is shown to be more efficient than the Weiner-kernel approach in identifying nonlinear dynamic systems of the type considered.