TOWARD THE SYNTHESIS OF GLOBAL OPTIMUM HEAT EXCHANGER NETWORKS: THE UNPINCHED CASE

Publisher Website
Google Scholar
Abstract
An algorithmic-evolutionary procedure is studied for developing improved maximum energy recovery (MER) and minimum number of units (MNU) networks with the goal of achieving the global optimum network. The procedure is initiated with a starting MER/MNU network and uses path tracing/list processing constructions that allow development of successive MER/MNU networks which are in some sense adjacent to the initial network. Heuristic rules are used to generate choices which lead to an improving sequence of network cost function values. The properties and limitations of the constructions and procedures are established and the effectiveness of the composite algorithm is demonstrated via literature test problems. Attention is restricted to unpinched problems: extensions to the pinched case are given in a companion paper.