论文部分内容阅读
针对非线性被控对象,该文提出了基于支持向量机的逆系统控制方法。对于最小相位的非线性离散系统,该方法根据系统的输入输出数据,使用支持向量机回归的方法来辨识构造原系统的α-阶逆系统。将辨识构造出的逆系统与原系统相联结,就能形成α-阶纯延时伪线性系统。这样,使用线性系统的成熟控制方法(如极点配置等等),就能有效地对非线性系统进行控制。仿真实验显示,即使对于非仿射并且非线性很强的系统,在没有系统模型的先验知识的情况下,利用该方法都能准确地建立逆系统的模型,从而获得良好的控制效果。
For non-linear controlled objects, this paper proposes an inverse system control method based on support vector machines. For the nonlinear discrete system with minimum phase, this method uses support vector machine regression to identify the α-order inverse system of the original system based on the input and output data of the system. When the constructed inverse system is connected with the original system, the pseudo-linear system with α-order pure delay can be formed. In this way, the nonlinear system can be effectively controlled using the proven control of linear systems (eg pole placement, etc.). Simulation results show that, even for a non-affine and non-linear system, the inverse system model can be accurately established without any prior knowledge of the system model, so as to obtain a good control effect.