采用多模型自适应的阶梯型广义预测控制

Generalized Predictive Control Based on Multiple Model Adaptive

针对模型参数随时间突变的系统,利用多个元素模型覆盖被控对象的不确定性对该多个元素模型建立相应的预测控制器。采用多个固定模型、一个常规自适应模型和一个可重新赋值的自适应模型实时在线辨识系统特性。在每个采样时刻根据性能指标函数选择最佳控制器。同时设计阶梯式广义预测控制器以实现系统的全局控制。仿真结果表明:其控制效果明显优于单一常规自适应模型和多个固定模型的控制器。

Generalized predictive control based on multiple model adaptive is proposed for systems with jumping parameters. By using multiple model adaptive control, multiple element models were utilized to cover the uncertainty of the controlled system, and multiple controllers were set up accord-ing to multiple element models. Multiple fixed models, a conventional adaptive model and an assigned initial value adaptive model are established to identify dynamic characteristic in parallel. A best con-troller is chosen by the performance function at every sampling time. And a stair like generalized pre-dictive controller is designed for this model to achieve the full-range operation. Finally, simulation shows that the control effect of proposed method is superior to a single controller and multiple models control-ler.