An extended stage-wise superstructure for heat exchanger network synthesis with intermediate placement of multiple utilities

The heat exchanger network (HEN) synthesis methodologies based on mathematical programming are recognized to be the significant methods to improve energy utilization efficiency and achieve a prominent techno-economic performance. As impacted by the combinatorial and nonconvex nature in the mathematical model, the problem of HEN synthesis may be easily trapped into suboptimal solutions. In this study, an extended stage-wise superstructure with intermediate placement of multiple utilities was presented. The heaters and coolers can be put in different stream split branches at intermediate stages, as opposed to the traditional allocation at stream ends. Such an extended superstructure yielded a more complex mathematical model compared to the stage-wise superstructure. To solve the model, the solution approach based on Genetic Algorithm (GA) was adopted. The local optimizing strategy, the adaptive scheme, the elite strategy and the structure identification and structure control strategy were introduced to enhance the efficiency and improve the search ability of the GA. Finally, the analysis of four examples verifies the validity of the presented meta-heuristics GA with optimization strategies, demonstrating that the proposed method can solve the complex model with acceptable computational effort and obtain better solutions with lower total annual costs than those reported in previous works.