《计算机应用研究》|Application Research of Computers

可行性规则动态调整的多目标粒子群算法

Multi-objective particle swarm algorithm based on dynamic adjustment of feasibility rule

免费全文下载 (已被下载 次)  
获取PDF全文
作者 赵乃刚,李勇,王振荣
机构 1.山西大同大学 数学与计算机科学学院,山西 大同 037009;2.山西省大同市人民政府信息化中心,山西 大同 037009
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2018)05-1304-03
DOI 10.3969/j.issn.1001-3695.2018.05.005
摘要 针对多目标粒子群算法存在的问题,提出了一种可行性规则动态调整的多目标粒子群算法。在算法中,根据粒子之间的相似度值动态非线性地更新算法的惯性权重,使得算法可以高效地平衡全局和局部搜索之间的矛盾;采用动态加权法解决随机性抽取群体最优粒子的缺陷,保证了种群的多样性;并且动态改变可行性规则的阈值,使得算法可以有效地利用某些不可行解包含的有效信息,提高了算法收敛到Pareto前沿的能力。最后,与其他四种多目标算法的实验比较验证了新算法的性能更好。
关键词 粒子群算法;多目标优化;惯性权重;动态加权法;可行性规则
基金项目 国家自然科学基金资助项目(61672331)
山西省高等学校教学改革项目(2015090)
山西大同大学科学研究项目(2016K1)
本文URL http://www.arocmag.com/article/01-2018-05-005.html
英文标题 Multi-objective particle swarm algorithm based on dynamic adjustment of feasibility rule
作者英文名 Zhao Naigang, Li Yong, Wang Zhenrong
机构英文名 1.CollegeofMathematics&ComputerScience,ShanxiDatongUniversity,DatongShanxi037009,China;2.InformationCenterofDatongPeople'sGovernmentofShanxiProvince,DatongShanxi037009,China
英文摘要 Aiming at the problems of multi-objective particle swarm algorithm, this paper developed a multi-objective particle swarm algorithm based on dynamic adjustment of feasibility rule. This algorithm dynamically non-linear updated the inertia weight according to the similarity value between particle and best particle. In order to solve the defects of random selection of the best particle, this paper proposed dynamic update method of inertia weight and it could ensure the diversity of the population. It changed the threshold of feasibility rules dynamically so that new algorithm could take advantage of some valid information of some infeasible solutions and it could improve the ability of converging to the Pareto frontier. Experiment comparison with other four algorithms shows better performance of the new algorithm.
英文关键词 particle swarm algorithm; multi-objective optimization; inertia weight; dynamic weighting method; feasibility rule
参考文献 查看稿件参考文献
  [1] Konak A, Coit D W, Smith A E. Multi-objective optimization using genetic algorithms:a tutorial[J] . Reliability Engineering & System Safety, 2006, 91(9):992-1007.
[2] Meo S, Zohoori A, Vahedi A. Optimal design of permanent magnet flux switching generator for wind applications via artificial neural network & multi-objective particle swarm optimization hybrid approach[J] . Energy Conversion and Management, 2016, 110(2):230-239.
[3] Li Xiaodong. A non-dominated sorting particle swarm optimizer for multi-objective optimization[C] //Genetic and Evolutionary Computation. Berlin:Springer, 2003:37- 48.
[4] Lei Deming. Pareto archive particle swarm optimization for multi-objective fuzzy job scheduling problem[J] . Computers & Industrial Engineering, 2008, 54(4):960-971.
[5] Tavakkoli-Moghaddam R, Azarkish M, Sadeghnejad A. A new hybrid multi-objective Pareto archive PSO algorithm for a bio-objective job shop scheduling problem[J] . Expert Systems with Application, 2010, 38(9):10812-10821.
[6] Liu Dasheng, Tan K C, Goh C K, et al. A multiobjective memetic algorithm based on particle swarm optimization[J] . IEEE Trans on Systems, Man and Cybernctics, 2007, 37(1):42-50.
[7] 陈民铀, 张聪誉, 罗辞勇. 自适应进化多目标粒子群优化算法[J] . 控制与决策, 2009, 24(12):1851-1855.
[8] 焦李成. 多目标优化免疫算法、理论和应用[M] . 北京:科学出版社, 2009:57-60.
[9] Deb K, Pratap A. A fast and elitist multi-objective genetic algorithm:NSGA-Ⅱ[J] . IEEE Trans on Evolutionary Computation, 2002, 6(2):182-197.
[10] 高圣国, 吴忠, 李旭芳, 等. 带两类正态变异的多目标粒子群算法[J] . 控制与决策, 2015, 30(5):939-942.
[11] 胡旺, Yen G G, 张鑫. 基于Pareto熵的多目标粒子群优化算法[J] . 软件学报, 2014, 25(5):1025-1050.
[12] 雷瑞龙, 候立刚, 曹江涛. 基于多策略的多目标粒子群优化算法[J] . 计算机工程与应用, 2016, 52(8):19-24.
收稿日期 2017/1/2
修回日期 2017/3/6
页码 1304-1306,1336
中图分类号 TP301.6
文献标志码 A