摘要
为提高多目标粒子群优化(MOPSO)算法处理多目标优化问题的性能,降低计算复杂度,改善算法的收敛性,提出了一种改进的多目标粒子群优化算法。通过运用比例分布及跳数改进机制策略的方法,使该算法不仅继承了MOPSO算法的优点,而且具有很强的局部搜索能力和较好的鲁棒性能,使非劣解集均匀分布,尽可能逼近真实的非劣前沿。通过对多连杆悬架空间结构硬点的多目标优化,进一步验证了该算法的实用性及其优越性。
In order to enhance the multi-objective particle swarm optimization (MOPSO) algorithm processing performance for multi-objective optimization, reduce the computational complexity and improve the convergence of algorithm, this paper put for- ward an improved multi-objective particle swarm optimization algorithm, which used proportional distribution and jump improved mechanism, not only inherited the advantages of MOPSO algorithm, but had a strong local searching ability, good robust performance and uniform non-inferior solution set, as far as possible approximation real non-inferior front. The pract!cability and superiority of the proposed algorithm is verified by applying it into multi-objective optimization of the spatial structure geometry pa- rameters of a multi-link suspension.
出处
《计算机应用研究》
CSCD
北大核心
2014年第3期675-678,683,共5页
Application Research of Computers
基金
国家"十二五""863"计划重大项目(2011AA11A265
2012AA110701)
国家自然科学基金资助项目(50875173)
上海市科委科研计划资助项目(11140502000)
上海汽车工业科技发展基金资助项目(1104)
关键词
多目标粒子群优化
比例分布
跳数改进机制
多连杆悬架
muhi-objective particle swarm optimization
proportional distribution
jump improved mechanism
multi-link sus-pension