Evolutionary-based multi-objective optimization of multi-stream plate-fin heat exchangers

Plate-fin heat exchangers (PFHEs) are widely used in industrial applications because of their compactness and high thermal efficiency. To achieve optimal PFHE performance, it is necessary to reduce the friction factor while increasing the Nusselt number. In this work, more sophisticated multi-objective evolutionary algorithms, such as the Multi-Objective Adaptive Genetic Algorithm (MOAGA) and the Multi-Objective Differential Evolution (MODE), are used to identify the best design configurations. An experimental design based on Taguchi is employed to identify the key design parameters, and then Response Surface Methodology (RSM) is applied to develop predictive models for the friction factor and Nusselt number. The major optimisation factors are the Reynolds number, core length, fin length, and core area. The optimisation results indicate that both MOAGA and MODE can be useful in determining Pareto-optimal solutions that enhance heat transfer performance while minimising pressure drop. A comparative analysis of the two algorithms highlights their ability to address complex multi-objective optimisation challenges within the PFHE design space.