This study presents an approach for optimal design and control of large-scale applications under uncertainty. The trust region method is employed in this work to perform optimal process design. To alleviate the complexity and nonlinearity of the problem, the method attempts to solve a series of bounded optimization problems around the worst-case scenario using surrogate models such as Power Series Expansions (PSE). The idea is to search for the optimal search direction in the optimization variables by solving a series of PSE-based optimization problems. The competence of the surrogate models is certified in trust regions where the representative optimization problems remain accurate within a certain threshold. The approach was tested for simultaneous design and control of the Tennessee Eastman (TE) process. The results show that the proposed method converges to dynamically feasible, computationally affordable, and potentially local optimum design and control configurations that can accommodate multiple disturbances and uncertainties.

这项研究提出了一种在不确定条件下优化设计和控制大型应用程序的方法。在这项工作中采用信任域方法来执行最佳过程设计。为了减轻问题的复杂性和非线性，该方法尝试使用诸如Power Series Expanions（PSE）等替代模型来解决最坏情况下的一系列有界优化问题。这个想法是通过解决一系列基于PSE的优化问题来在优化变量中搜索最优搜索方向。代理模型的能力在信任区域得到认证，在该区域中，代表性的优化问题在一定阈值内保持准确。该方法已经过测试，可以同时设计和控制田纳西州伊士曼（TE）流程。