复杂高阶对象的预测PID控制

Predictive PID Control for Complex High Order

摘要:

针对复杂高阶对象提出了一种基于数值最优模型降阶方法,并基于这种降阶模型设计了预测PID控制器,将此控制器应用于原始模型能够得到很好地控制效果。数值最优模型降阶算法使高阶对象能近似为一阶加时滞对象或二阶加时滞对象,通过模型阶跃响应和Bode图对比,降阶模型曲线很好地逼近原始模型曲线。预测PID对大时滞对象有着很好地控制效果,模型降阶使得预测PID很好地控制复杂高阶对象,且其结构简单,可调参数少的特点。

Abstract:

For complex high order object is put forward based on the Numerical optimal model reduction method, and based on the model reduction method to design the predictive PID controller, the controller was applied to the original model can get a good control effect. the Numerical optimal model reduction method makes complex high order objects can approximate for a first order plus time delay or second order time delay object, through the model of step response and Bode chart compares, the reduced order model curve approximation of the original curve model. Predictive PID has a good control effect of large time delay systems; the model reduction method makes prediction PID control complex high order systems, and its simple structure, less tunable characteristics.