增量式多变量预测函数控制及其在气体分馏装置中的应用  

Incremental multivariable predictive functional control and its application in gas fractionation unit

作　　者：苏成利[1] 施惠元[1] 李平[1] 王倩[1] 许娜[2] 

机构地区：[1]辽宁石油化工大学信息与控制工程学院,辽宁抚顺113001 [2]济南大学自动化与电气工程学院,山东济南250022

出　　处：《南京理工大学学报》2015年第5期625-631,共7页Journal of Nanjing University of Science and Technology

基　　金：国家自然科学基金(61203021);辽宁省科技攻关项目(2011216011);辽宁省自然科学基金(2013020024)

摘　　要：针对传统的比例-积分-微分(PID)控制应用于多变量、大滞后的复杂过程时无法获得满意控制效果的问题,该文提出一种增量式多变量预测函数控制(IMPFC)算法。该算法利用过程阶跃响应数据建立增量式传递函数矩阵模型,采用单点线性化方法直接递推出过程预测输出,并对控制增量进行优化求解,解决了位置式预测函数控制算法对控制量约束的不足,使控制量变化更加平缓。通过对预测输出误差和未来的设定值进行多项式拟合,克服了模型失配问题,使预测输出更接近于参考轨迹。在气体分馏装置中的应用表明,该文方法具有良好的跟踪性、鲁棒性和抑制干扰的能力,其控制品质优于传统PID控制。In view of that the effect of the traditional proportion-integration-differentiation（ PID） control is not satisfying when it is used for the complicated process with multivariable characteristics and large time delay,an incremental multivariable predictive functional control（ IMPFC） algorithm is proposed here. An incremental transfer function matrix model is set up through the step-response datesand predictive outputs are deduced with the theory of single point linear method. The incremental control variable is optimized,the constraint of the incremental control variable with the positional predictive functional control algorithm is rejected,and the control variable is made smoother. The predictive output error and future set-point are approximated by a polynomial to overcome the model mismatch and make the predictive outputs tracking the reference trajectory. The application in the gas fractionation unit shows that the proposed control strategy has good effect in tracking,robustness and disturbance rejection,and its control performance is superior to that of the PID control.

关 键 词：多变量预测函数控制 增量式模型 气体分馏装置 

分 类 号：TP13[自动化与计算机技术—控制理论与控制工程]