Charitopoulos, Vasileios; (2018) Uncertainty-aware integration of control with process operations and multi-parametric programming under global uncertainty. Doctoral thesis (Ph.D), UCL (University College London).

Abstract

Advanced decision making in the process industries requires efficient use of information available at the different hierarchical levels. However, given the time and decision space sparsity the consideration of such integrated problems poses a plethora of challenges. It is goal of the present thesis to present some recent algorithmic and modelling developments with special focus on the uncertainty aware integration of control with process operations. To this end, the thesis comprises of two parts that are orchestrated towards the attainment of the aforestated goal. The first part discusses theoretical and algorithmic advances in the field of multiparametric programming. Initially, the case of multi-parametric linear programs under simultaneous variations in the left-handside, right-handside of the constraints and the objective function’s coefficients is examined. For the first time theoretical characterisation of the explicit solution is proven while an algorithm for their exact computation is proposed. Later on, the aforementioned algorithm is extended to the mixed-integer case where problems of process synthesis and scheduling under global uncertainty are studied. Next, the concept of multi-setpoint explicit controllers and their potential in the context of enterprise wide optimisation problems is introduced. While a prototype implementation of the aforementioned works is also discussed. The second part is dedicated to the development of a systematic framework for the uncertainty aware integration of process planning, scheduling and control (iPSC) of continuous processes. Initially, a Traveling Salesman Problem based formulation is presented and a decomposition method for the deterministic case is proposed. Next, the multi-setpoint explicit controllers developed in the first part of the thesis, enable the development of a reactive closed-loop framework for the iPSC. Ultimately, proactive and reactive approaches are employed in order to instantiate the uncertainty aware iPSC while Monte-Carlo simulations are conducted to evaluate the robustness of the proposed framework.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item