Nonlinear predictive functional control of recursive subspace model using support vector machine

Abstract

In nonlinear predictive functional control, the speed of time varying response is slow. This problem is considered in this paper. A strategy, based on least squares support vector machine (LS-SVM) of nonlinear predictive functional control of recursive subspace model, is developed. The predictive model of the nonlinear predictive functional control is Hammerstein model. Gets output function of nonlinear static link according to principle of LS-SVM, and identifies linear dynamic link with model of recursive subspace. On the basis of distinguishing effectively to the nonlinear objects, realizes rapidly distinguish, improves time varying response speed, and has good performance of tracking ability in nonlinear predictive functional control. Simulation results show the validity and superiority of this algorithm.