Title :
Modeling of nonlinear systems from input-output data
Author_Institution :
Princeton University, Princeton, New Jersey
Abstract :
This paper reviews a procedure for obtaining a sequence of state-affine discrete-time realizations whose input-output behavior matches a set of empirical input-output records in the least squares sense. First the input-output behavior is approximated by orthogonal polynomials. Then, state-affine partial realizations are obtained from the associated generalized Hankel matrix.
Keywords :
Analytical models; Computer science; Context modeling; Impedance matching; Mathematical model; Nonlinear systems; Piecewise linear approximation; Polynomials; State-space methods; Topology;
Conference_Titel :
Decision and Control, 1983. The 22nd IEEE Conference on
Conference_Location :
San Antonio, TX, USA
DOI :
10.1109/CDC.1983.269597
