Hybrid multi-objective optimization with Particle Swarm Optimization and Extremal Optimization for engineering design

摘要

A new hybrid multi-objective optimization (MO) solution with the combination of Particle Swarm Optimization (PSO) and Extremal Optimization (EO), called “PSO-EO-MO”, was presented in authors' early studies. The proposed algorithm is based on the superior functionalities of PSO for searching a Pareto dominance and extremal dynamics oriented EO for fine tuning and adjustment. The concept of crowding and lattice for the external archive is also employed for diversity preservation and getting a well-distributed sets of non-dominated solutions. Based on our previous studies, in this study the proposed algorithm is applied to four MOPs in engineering design by comparison with other multi-objective evolutionary algorithms (MOEAs). The results indicate the algorithm is able to find better and much wider spread of solutions. Consequently, the proposed solution may be applied to more complex real-world MOPs.