Abstract: A new method of nonlinear controller is proposed based on support vector machine and the concept of nonlinear predictive controller.This controller can optimize these parameters in LS-SVM using chaos optimization according to the feature of chaotic mapping and COA can also be utilized as online optimization.It can avoid calculating the complicated gradient and the inverse matrix in the nonlinear predictive control.The simulation studies show the effective performance of the proposed controller.

Key words: LS-SVM, chaos optimization, predictive controller, rolling optimization

摘要： 针对非线性系统的控制问题，提出了一种基于LS-SVM的预测控制技术。该方法将LS-SVM和非线性预测控制思想有机结合，利用混沌映射的特性，通过引入混沌优化技术对LS-SVM参数进行优化，同时又将其作为滚动优化策略，避免了非线性预测控制中复杂的梯度计算和矩阵求逆问题，仿真研究说明了该非线性预测控制器的有效性及实时性。

关键词: LS-SVM, 混沌优化, 预测控制, 滚动优化