Design of heat exchangers using a novel multiobjective free search differential evolution paradigm

•A novel approach combining differential evolution and free search (MOFSDE) is proposed.•Simulations showed that MOFSDE produces competitive results to heat exchanger design.•Results illustrate that MOFSDE efficiently achieves goals of multiobjective optimization. Free search (FS) is a recently prop...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Applied thermal engineering 2016-02, Vol.94, p.170-177
Hauptverfasser: Hultmann Ayala, Helon Vicente, Keller, Patrick, Morais, Márcia de Fátima, Mariani, Viviana Cocco, Coelho, Leandro dos Santos, Rao, Ravipudi Venkata
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•A novel approach combining differential evolution and free search (MOFSDE) is proposed.•Simulations showed that MOFSDE produces competitive results to heat exchanger design.•Results illustrate that MOFSDE efficiently achieves goals of multiobjective optimization. Free search (FS) is a recently proposed population-based metaheuristic algorithm, inspired from the animals' behavior. FS can be applied to real value numerical optimization problems, as well as evolutionary algorithms and swarm intelligence techniques. In this paper, a novel multiobjective FS approach combined with differential evolution (MOFSDE) to heat exchanger optimization is presented. Two case studies of heat exchanger design are carried out to illustrate the efficiency of the MOFSDE. Simulation results for the two multiobjective case studies using the proposed MOFSDE are compared with those obtained by using the nondominated sorting genetic algorithm, version II (NSGA-II). The results from this comparison indicate that the MOFSDE performs better than the NSGA-II. The results illustrate that MOFSDE efficiently achieves two goals of multiobjective optimization problems: to find the solutions that converge to an approximated Pareto-front which is well spread, having the advantage of no parameter tuning apart from the population size and the number of generations.
ISSN:1359-4311
DOI:10.1016/j.applthermaleng.2015.10.066