Retrofit of heat exchanger networks with topology changes under designer control

This paper addresses the problem of determining the optimal retrofit of an existing heat exchanger network (HEN). The proposed algorithmic approach not only takes into account the selection of heat matches and their heat loads but also the proper unit-match reassignments, the need of new units and the requirement of additional area in existing exchangers. Moreover, it provides the retrofitted network at the level of structure. The model formulation was developed by incorporating new constraints and further variables for controlling the structural changes in the neighbor-based network model recently proposed by Galli and Cerda (1998a,b,c). The problem goal is to minimize the total network retrofit cost including those related to manpower for moving exchangers to other sites, re-piping, purchasing of new units and/or additional area for existing exchangers. Non-linearity of the exchanger design equation has been overcome by fixing the temperature driving force in each exchanger at some specified minimum value and developing a computationally efficient iterative solution procedure. At each iteration, temperature driving forces are properly updated and the resulting MILP problem is subsequently solved until convergence is achieved. In this way, a MILP-based methodology has been derived and applied to the retrofit of existing networks under different user-specified topology restrictions.