A feasible-side globally convergent modifier-adaptation scheme. (June 2017)

• Record Type: Journal Article
• Authors: Marchetti, Alejandro G.; Faulwasser, Timm; Bonvin, Dominique
• Is Part Of: Journal of process control. Volume 54(2017)
• Page Start: 38
• Page End: 46
• Publication Date: 2017-06
• Subjects: Real-time optimization -- Static optimization -- Optimization under uncertainty -- Global convergence -- Feasible-side convergence; Process control -- Periodicals; Fabrication -- Contrôle -- Périodiques; Process control; Periodicals; Electronic journals; 660.281
• DOI: 10.1016/j.jprocont.2017.02.013
• Languages: English
• ISSNs: 0959-1524

Abstract

Abstract : Highlights: Real-time optimization schemes do not in general guarantee feasible-side convergence. A general RTO formulation that guarantees feasible-side global convergence is proposed. The approach is based on convex inner-approximations of the cost and constraints. A modifier-adaptation algorithm with feasible-side global convergence is proposed. Abstract: In the context of static real-time optimization (RTO) of uncertain plants, the standard modifier-adaptation scheme consists in adding first-order correction terms to the cost and constraint functions of a model-based optimization problem. If the algorithm converges, the limit is guaranteed to be a KKT point of the plant. This paper presents a general RTO formulation, wherein the cost and constraint functions belong to a certain class of convex upper-bounding functions. It is demonstrated that this RTO formulation enforces feasible-side global convergence to a KKT point of the plant. Based on this result, a novel modifier-adaptation scheme with guaranteed feasible-side global convergence is proposed. In addition to the first-order correction terms, quadratic terms are added in order to convexify and upper bound the cost and constraint functions. The applicability of the approach is demonstrated on a constrained variant of the Williams–Otto reactor for which standard modifier adaptation fails to converge in the presence of plant-model mismatch.