Nonlinear system identification by evolutionary computation and recursive estimation method

Jih-Gau Juang, Bo-Shian Lin
Proceedings of the 2005, American Control Conference, 2005.  
Nonlinear system identification using evolutionary computation and recursive estimation method is presented. Four different recursive estimation methods, recursive least-squares, recursive least-squares with exponential forgetting, stochastic algorithm, and projection algorithm, combined with evolution algorithm are used in this study. Conventional system identification using recursive estimation methods are also given for comparison. After test, the proposed scheme has better convergence and
more » ... curacy on parameter estimation than the conventional estimation method.
doi:10.1109/acc.2005.1470820 fatcat:f6m4f2pzjbg6djpimbimcmygl4