Premium
Identification in nonlinear systems by using an automatic choosing function and a genetic algorithm
Author(s) -
Hachino Tomohiro,
Takata Hitoshi
Publication year - 1998
Publication title -
electrical engineering in japan
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.136
H-Index - 28
eISSN - 1520-6416
pISSN - 0424-7760
DOI - 10.1002/(sici)1520-6416(199812)125:4<43::aid-eej6>3.0.co;2-q
Subject(s) - nonlinear system , algorithm , identification (biology) , function (biology) , genetic algorithm , least squares function approximation , domain (mathematical analysis) , computer science , system identification , mathematics , nonlinear system identification , mathematical optimization , mathematical analysis , data modeling , statistics , physics , botany , quantum mechanics , biology , database , evolutionary biology , estimator
This paper deals with an identification method based on an automatic choosing function (ACF) for nonlinear systems. A full data region or an entire domain is partitioned into subdomains and the unknown nonlinear function to be estimated is approximately described by a linear equation on each subdomain. These linear equations are smoothly united into a single expression by the ACF, and the resulting model is linear in its parameters. Hence these parameters are easily evaluated by the linear least‐squares method. The subdomains and the ACF are properly determined by a genetic algorithm that has a high potential for global optimization. Numerical experiments demonstrate the effectiveness of the proposed method. © 1998 Scripta Technica, Electr Eng Jpn, 125(4): 43–51, 1998