非线性过程自优化控制的快速算法

针对非线性过程的自优化控制问题,提出了一种求取被控变量的快速算法.不同于线性过程的自优化控制,本文方法基于系统的非线性模型,并且最小化全局平均损失.为快速求解得到的非凸非线性规划问题,作者对其进行了简化.讨论了被控变量解空间的相关特性,阐述了引入的正交酉约束的合理性,并进一步提出了求解次优被控变量的解析法.对一个数值算例和蒸发过程的研究结果表明,提出的快速算法是便捷的、有效的.

A fast algorithm for self-optimizing control of nonlinear plants

We investigate the self-optimizing control (SOC) problem for nonlinear processes and propose a fast algorithm for solving controlled variables (CVs). Being different from those linear SOC methods, the proposed one minimizes the global average loss based on the rigorous nonlinear model. To quickly solve the derived non-convex linear programming (LP) problem, simplifications are introduced to the problem. Properties of the CV solution space are discussed to illustrate the rationality of introducing the orthogonal unitary constraint. On this basis, the analytical solution of suboptimal CVs is obtained in a further step. The developed methodology is applied to a numerical example and an evaporator process, where the simplicity and effectiveness of the proposed algorithm are demonstrated.