基于支持向量机的一类非线性系统预测控制

针对一类具有输出反馈耦合的离散非线性系统,将过程的非线性部分通过支持向量机转化为全局线性状态空间模型,并在目标函数中引入系统状态的变化,给出一种类似于离散PI最优调节器的新型预测控制器.该方法不需要在线辨识系统参数,因为系统的内模已转换成全局离线模型.由于引入了新的优化目标函数,该控制器的控制效果和鲁棒性优于仅考虑预测输出误差的传统预测控制器.仿真结果表明,它也优于经典离散PI最优调节器.

Predictive control of a class of nonlinear systems based on support vector machine

A predictive control method for a class of discrete nonlinear systems is presented by using support vector machine to change the nonlinear model into a global off-line linear model. The method does not need on-line parameters estimation because the internal model of the system has been transformed into an off-line global model. The structure of the controller is similar to classic PI optimal controller. Its control performance and robustness are more advantageous than traditional predictive control which only considers output errors. Simulation results show it is better than classic PI optimal controller.