基于输入输出模型的双时标分散控制方法

针对不同通道的过程动态响应速度具有显著差异的多变量系统，提出了一种基于时标分解的双时标分散控制方法.由于控制中的过程模型一般都是由实验测试方法得到的输入输出模型，该方法是以系统输入输出模型为基础，在状态空间模型的框架下对系统进行基于奇异摄动理论的时标分解，并由分解得到的快、慢模型分别设计最优控制器.此方法实现了时标模型实际上的分离，并对2个子系统进行完全独立设计，给出了在最优二次型目标下的最优控制率.最后利用该方法仿真一个实际测得的三输入三输出的精馏塔模型，给出了设定值跟踪的仿真效果，验证了该方法的有效性.

Two-time scale decentralized control method based on input and output model

A decentralized two-time scale control algorithm for multivariable systems,in which the response time contrast of different channels is notable,was presented.Since most practical models used for control was obtained in the form of transfer function matrix by plant tests,a new time scales decomposition method based on singular perturbation was proposed for separating the original transfer function matrix model to two irrelevant state-space models in different time scales.Consequently the optimal control laws were deduced for the two models which utilizing different sample intervals.The algorithm was applied in a three-in-three-out column model,and the set point tracking result was given.The simulation showed the effectiveness of the method.