Support vector machines (SVM) have been widely used in pattern recognition and have also drawn considerable interest in control areas. Based on rolling optimization method and on-line learning strategies, a novel approach based on weighted least squares support vector machines (WLS-SVM) is proposed for nonlinear dynamic modeling. The good robust property of the novel approach enhances the generalization ability of kernel method-based modeling and some experimental results are presented to illustrate the feasibility of the proposed method.
展开▼
