Hybrid SVR-PSO for identification of nonlinear system

This paper develops a hybrid support vector regression (SVR)-particle swarm optimization (PSO) model to identify nonlinear systems. The predictive accuracy of SVR models is highly dependent on their learning parameters. Therefore, PSO is exploited to seek the optimal hyper-parameters for SVR in order to improve its generalization capability. The proposed identification procedure is successfully applied to measurements of nonlinear systems. The efficiency of the proposed algorithm was demonstrated by some simulation examples. © 2013 by CESER Publications.