基于支持向量机的参数自整定PID非线性系统控制

Self-tuning PID controller for a nonlinear system based on support vector machines

对非线性系统提出了一种基于支持向量机的自整定PID控制新方法.用支持向量机辨识系统的非线性关系,并对之进行线性化,提取出瞬时线性模型,采用最小方差的准则获取PID控制器的最优参数.为改善控制器的性能,提出了一些改进措施,包括使用一阶滤波器、控制器参数更新标准及惩罚系数的调整等.通过对典型非线性系统的仿真.验证了该方法的有效性和可行性.

For a nonlinear system, we propose a new self-tuning PID controller based on support vector machines （SVM）. The SVM is applied to identify the non-linear system, which is then linearized to extract the instantaneous linear model. The criterion of minimum variance is used to obtain the optimal tuning PID controller. Some additional measures, including the adoption of first order filter, parameter updating criterion and the adjustment of penalty factor, are taken to improve the performance of PID controller. Simulation results of typical non-linear systems show the effectiveness and feasibility of the self-tuning PID controller based on SVM.