Combining Multi-Parametric Programming and NMPC for the Efficient Operation of a PEM Fuel Cell

• Main Authors: C. Ziogou, M.C. Georgiadis, E.N. Pistikopoulos, S. Papadopoulou, S. Voutetakis
• Format: Article
• Language: English
• Published: AIDIC Servizi S.r.l. 2013-09-01
• Series: Chemical Engineering Transactions
• Online Access: https://www.cetjournal.it/index.php/cet/article/view/6101

This work presents an integrated advanced control framework for a small-scale automated Polymer Electrolyte Membrane (PEM) Fuel Cell system. At the core of the nonlinear model predictive control (NMPC) formulation a nonlinear programming (NLP) problem is solved utilising a dynamic model which is discretized based on a direct transcription method. Prior to the online solution of the NLP problem a pre- processing search space reduction (SSR) algorithm is applied which is guided by the offline solution of a multi-parametric Quadratic Programming (mpQP) problem. This synergy augments the typical NMPC approach and aims at improving both the computational requirements of the multivariable nonlinear controller and the quality of the control action. The proposed synergetic framework is deployed to the automation system of the unit and its online response is explored by a comparative experimental case study that reveals its applicability and efficiency with respect to the fulfilment of multiple desired objectives under physical and operational constraints.