期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2013
卷号：48
期号：2
页码：967-972
出版社：Journal of Theoretical and Applied
摘要：Given the influence of the selection of regression parameters on the accuracy of SVR model and its ability of learning and generalization, this article adopts the particle swarm optimization algorithm to build the SVR model and applies it to the modeling of nonlinear system identification. Through the simulation experiments, it is found that this model is more accurate in identification and has a stronger ability of learning and generalization compared with GA. In addition, it demonstrates that the application in nonlinear system identification based on PSO-SVR algorithm could be considerably effective.
关键词：Particle Swarm Optimization (PSO); Support Vector Regression (SVR); Nonlinear System
