基于鲁棒支持向量机的非线性系统辨识

Nonlinear systems identification based on robust support vector machine

支持向量机(SVM)是一种基于结构风险最小化原理的学习技术,该文利用鲁棒支持向量机对非线性系统进行黑箱建模,首先推导出鲁棒支持向量机的基本理论,给出了对偶优化问题,并结合一个具体的例子进行了仿真实验,结果验证了所提出的方法的正确性和有效性。

Support vector machine is a learning technique based on the structural risk minimization principle, This paper uses robust support vector machine to model nonlinear dynamical systems, firstly deduce the theory of robust support vector machine, gives its dual optimization problem. A concrete simulation example is taken to demonstrate the proposed approach's correctness and effectiveness.