Infinite horizon MPC of a heat-exchanger network

Abstract

A method to control a heat-exchanger network (HEN) is presented where both, the control objective and the economic objective are taken into account. It is assumed that we have a two-level structure in which the steady state economic optimisation is performed in the upper level and the MPC controller is used to enforce the optimal operating point defined by the economic layer. Since HENs are multivariable highly interactive systems, which integrate large energy consumers of a process plant, stability of the proposed structure becomes one of the issues of the HEN control problem. In this approach, integration is achieved through the definition of an extended cost-function that provides the controller with the ability of driving the system to optimal conditions. The proposed MPC algorithm uses a linear state space model that is particularly suitable to the development proposed in this work. The method is exemplified with the simulation of a HEN system that shows the typical characteristics of an industrial system.