Heat integration of energy system using an integrated node-wise non-structural model with uniform distribution strategy

Abstract The present paper proposes an original node-wise non-structural model (NW-NSM) with no stream splits for the heat integration of energy system. Global optimization approaches using structural models such as the stage-wise superstructure model, have limitations for heat exchanger network synthesis. Instead of complying with the typical fixed multi-stage network frameworks of stage-wise superstructures, the new model generates stream matches by randomly connecting preset nodes in the hot and cold streams. Then, the NW-NSM is optimized using the random walk algorithm with compulsive evolution. Furthermore, the effects of nodes on the solution space and performance of the NW-NSM are analyzed. Finally, a uniform distribution strategy for node locations of existing heat exchangers is used to expand the solution space and to ensure the generation of all possible stream matches. The combination of a flexible model and stochastic algorithm improves the global optimization. The model was applied to medium and large-scale case studies to verify the feasibility and high efficiency of the NW-NSM with respect to the global optimization of heat exchanger networks. The model achieved more economical results compared to other models previously presented in the literature.