Self-adaptive mutation differential evolution algorithm based on particle swarm optimization

Differential evolution (DE) is an effective evolutionary algorithm for global optimization, and widely applied to solve different optimization problems. However, the convergence speed of DE will be slower in the later stage of the evolution and it is more likely to get stuck at a local optimum. More...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Applied soft computing Ročník 81; s. 105496
Hlavní autoři: Wang, Shihao, Li, Yuzhen, Yang, Hongyu
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier B.V 01.08.2019
Témata:
Arrival flights scheduling
Differential evolution
Elite archive
Particle swarm optimization
Self-adaptive mutation
Elite archive
Differential evolution
Arrival flights scheduling
Particle swarm optimization
Self-adaptive mutation
ISSN:1568-4946, 1872-9681
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Differential evolution (DE) is an effective evolutionary algorithm for global optimization, and widely applied to solve different optimization problems. However, the convergence speed of DE will be slower in the later stage of the evolution and it is more likely to get stuck at a local optimum. Moreover, the performance of DE is sensitive to its mutation strategies and control parameters. Therefore, a self-adaptive mutation differential evolution algorithm based on particle swarm optimization (DEPSO) is proposed to improve the optimization performance of DE. DEPSO can effectively utilize an improved DE/rand/1 mutation strategy with stronger global exploration ability and PSO mutation strategy with higher convergence ability. As a result, the population diversity can be maintained well in the early stage of the evolution, and the faster convergence speed can be obtained in the later stage of the evolution. The performance of the proposed DEPSO is evaluated on 30-dimensional and 100-dimensional functions. The experimental results indicate that DEPSO can significantly improve the global convergence performance of the conventional DE and thus avoid premature convergence, and its average performance is better than those of the conventional DE, PSO and the compared algorithms. Moreover, DEPSO is applied to solve arrival flights scheduling and the optimization results show that it can optimize the sequence and decrease the delay time. •A self-adaptive mutation DE based on particle swarm optimization is proposed.•An elite archive strategy is introduced to improve DE/rand/1 mutation strategy.•A self-adaptive mutation strategy is presented in DEPSO.•The effect of the parameters on DEPSO is experimentally investigated.•The DEPSO is applied to arrival flights scheduling.
ISSN:1568-4946
1872-9681
DOI:10.1016/j.asoc.2019.105496