Identification of Quadratic Volterra Polynomials in the “Input–Output” Models of Nonlinear Systems

In this paper, we propose a new algorithm for constructing an integral model of nonlinear dynamic system the “input–output” type in form quadratic segment Volterra integro-power series (polynomial). We consider nonparametric identification models using physically realizable piecewise linear test signals time domain. The advantage presented approach is to obtain explicit formulas calculating transient responses (Volterra kernels), which determine unique solution equations first kind with two variable integration limits. numerical method proposed paper solving corresponding includes use smoothing splines. An important result that constructed has low methodological error.