A Comparison of Four Memetic Particle Swarm Optimization Algorithms for Continuous Optimization

摘要

Particle swarm optimization (PSO) belongs to swarm intelligence category. It is a famous prototype for dealing with continuous optimization problems, and its efficiency can be enhanced by hybrid with local search methods. Based on recently designed four memetic PSO algorithms, this paper investigates the effectiveness and running time of these algorithms. Experiments are conducted on a set of mathematical test functions. The effectiveness of algorithms are compared based on the quality of solutions found in repeated runs. Their running times are compared based on clock time metric. It is found that PSO hybrid with crossover operator is much more effective than the other memetic PSO algorithms.