摘要
实际工业生产中存在着大量多变量时滞过程,由于此类系统输入输出变量间的耦合及时滞因子的影响,使得针对这类系统的控制效果往往达不到令人满意的效果。本文利用解耦控制器,首先对多变量时滞过程的输入输出进行解耦;由于解耦得到的子系统模型非常复杂,因此对解耦后的复杂系统模型提取公共时滞因子,然后用观测器/卡尔曼滤波器辨识(OKID)算法对解耦后的系统进行模型精简,基于该简化模型使用极点配置方法设计离散时间PID控制律,实现解耦系统对定值信号的跟踪控制。通过对一个化工过程系统进行仿真研究,证明了所提方法的有效性。
There are many multivariable system s with time delay i n the industrial process . Due to the coupling and time delay in these processes, the control performance s are usually not satisf ying . In this article, a model simplification method together with a digital PID controller design ing method is employ ed to the decoupled subsystems . Firstly, the multivariable process with time delay is decoupled into subsystems which are quite complicated. Then, the decoupled subsystems are simplified by a model-reduction method , using observer/Kalman filter identification （OKID） method, to obtain the discrete-time state-space models without involving the common time delay term . Based on the obtained simplified model, a pole assignment method is utilized to design discrete-time PID controller s which can guarantee the set-point tra cking performance of the decoupled subsystems. At last, an illustrative example is given to demonstrate the effectiveness of the proposed method.
出处
《控制工程》
CSCD
北大核心
2015年第5期820-825,共6页
Control Engineering of China
基金
国家自然科学基金(61374047)
