Nonlinear System Analysis and Identification Using Volterra Series with Harmonic Excitation

Inverse problems of system identification and parameter estimation are crucial in nonlinear analysis. Response behavior of nonlinear systems, under specific excitation conditions, can be predicted accurately only when the system structure and the parameters are completely known. While extensive research has been carried out in identification and parameter estimation of linear systems, research on nonlinear system identification has been relatively less. In the present paper, a complete identification procedure is presented using single-tone and multi-tone harmonic excitations. The response characteristics are derived through Volterra series representation of input-output mapping of a general physical system. Nonlinearity structure identification procedure is developed using harmonic analysis of structured response components and system parameters are estimated by solving a set of nonlinear equations in terms of first order FRFs and the nonlinear parameters.