Title :
The selection of neural models of nonlinear dynamical systems by statistical tests
Author :
Urbani, D. ; Roussel-Ragot, P. ; Person, L. ; Dreyfus, G.
Author_Institution :
Ecole Superieure de Phys. et de Chimie Ind., Paris, France
Abstract :
A procedure for the selection of neural models of dynamical processes is presented. It uses statistical tests at various levels of model reduction, in order to provide optimal tradeoffs between accuracy and parsimony. The efficiency of the method is illustrated by the modeling of a highly nonlinear NARX process
Keywords :
neural nets; nonlinear dynamical systems; reduced order systems; statistical analysis; efficiency; model reduction; neural models selection; nonlinear NARX process; nonlinear dynamical systems; statistical tests; Context modeling; Multi-layer neural network; Neural networks; Nonlinear dynamical systems; Polynomials; Predictive models; Recurrent neural networks; Reduced order systems; Structural engineering; System testing;
Conference_Titel :
Neural Networks for Signal Processing [1994] IV. Proceedings of the 1994 IEEE Workshop
Conference_Location :
Ermioni
Print_ISBN :
0-7803-2026-3
DOI :
10.1109/NNSP.1994.366044
