Application of dividing wall column technology for liquid petroleum gas fractionation and optimization using whale optimization algorithm and hybrid approaches

Abstract

Dividing Wall Columns (DWC) have gained significant attention for their ability to reduce energy consumption and capital costs in chemical separation processes. However, achieving an optimal design remains a complex challenge due to the nonlinear and multivariable nature of DWC systems This study aims on the application of DWC in liquid petroleum gas (LPG) fractionation process, to reduce high energy cost of traditional two-tower process, and compares different optimization algorithms, including the Whale Optimization Algorithm (WOA) and two novel hybrid variants, GA-WOA and GA-PSO. These advanced metaheuristic solvers were benchmarked against traditional algorithms, including Genetic Algorithm (GA) and Particle Swarm Optimization (PSO), through comprehensive comparative analysis. The results demonstrate that the GA-WOA hybrid delivers the best overall performance in solution quality, achieving the lowest Total Annual Cost with reductions of 27 % in condenser duty, over 10 % savings in reboiler duty, and a 14 % decrease in TAC compared to conventional two-column configurations. Among the standalone algorithms, WOA provides the best solution quality. While PSO emerges as the fastest converging algorithm, the GA-WOA hybrid provides the most balanced performance between solution quality and computational efficiency. Among the hybrid approaches, GA-PSO ranks as the second-best performer in terms of solution optimality. These statistically validated findings confirm the economic and energy benefits of DWC technology while establishing WOA hybrids, particularly GA-WOA, as powerful, reliable tools for complex chemical process optimization.