Adjustable Robust Optimal Control for Nonlinear Chemical Process under Uncertainty

Abstract

Abstract This paper has proposed an adjustable robust dynamic optimization (ARDO) scheme for nonlinear chemical process under time-dependent uncertainties. A novel continuous affine decision rule extended from multistage affine decision rule is developed to approximate the causal dependence of wait-and-see decision variables on the infinite dimensional uncertainty factors. Through applying the state-of-the-art first-order Taylor expansion method, the adjustable robust counterpart can be formulated. The effectiveness and applicability of the proposed ARDO scheme is demonstrated by a classic Williams-Otto semi-batch problem. Compared with traditional robust optimization for uncertain dynamic systems, the proposed framework can adjust the decision variables based on true values of the revealed uncertainties, leading to more flexible control profiles and less conservative solutions.