Minimizing water and energy consumptions in water and heat exchange networks.

Abstract

This study presents a mathematical programming formulation for the design of water and heat exchangers networks based on a two-step methodology. First, an MILP (mixed integer linear programming) procedure is used to solve the water and energy allocation problem regarding several objectives. The first step of the design method involves four criteria to be taken into account., ie, fresh water consumption (F1), energy consumption (F2), interconnection number (F3) and number of heat exchangers (F4). The multiobjective optimization Min [F1, F2] is solved by the so-called ɛ-constraint method and leads to several Pareto fronts for fixed numbers of connections and heat exchangers. The second step consists in improving the best results of the first phase with energy integration into the water network. This stage is solved by an MINLP procedure in order to minimize an objective cost function. Two examples reported in the dedicated literature serve as test bench cases to apply the proposed two-step approach. The results show that the simultaneous consideration of the abovementioned objectives is more realistic than the only minimization of fresh water consumption. Indeed, the optimal network does not necessarily correspond to the structure that reaches the fresh water target. For a real paper mill plant, energy consumption decreases of almost 20% as compared with previous studies.