AI-Driven Fuzzy Analogical Strategy for Pinch-Based Heat Exchanger Network Optimization

Abstract

This paper presents a novel AI-assisted approach to Heat Exchanger Network (HEN) synthesis using fuzzy analogical gates. The proposed methodology involves three key steps: (1) normalization of critical design variables, (2) evaluation using a fuzzy analogical gates network comprising symmetric (AND) and asymmetric (Invoke) gates, and (3) selection of the optimal minimum approach temperature based on a computed weight index. The symmetric gate integrates the hot utility requirements along with ΔTmin, while the second gate combines this output with cold utility demand. The method was validated using real aromatic plant case study. Results demonstrate that this technique reliably identifies optimal design parameters, reduces total annual cost, and is simple enough for manual implementation—offering a competitive alternative to more complex optimization models.

Article Content

Keywords

Article Info