基于差分元胞多目标遗传算法的动压滑动轴承优化设计  

Optimization Design of Hydrodynamic Sliding Bearing based on Differential Cellular Genetic Algorithm

作　　者：张屹[1] 郑小东[1] 万兴余 张灯皇 

机构地区：[1]三峡大学机械与动力学院,湖北宜昌443002

出　　处：《机械传动》2014年第9期64-68,共5页Journal of Mechanical Transmission

基　　金：国家自然科学基金(51275274)

摘　　要：为研究液体动压滑动轴承的多目标优化设计问题,提出了一种改进的多目标差分元胞遗传算法。改进的差分算子利用个体的Pareto支配关系进行排序,在排序的基础上进行差分操作,将该差分操作融入到元胞遗传算法中,以加快种群的收敛速度。同时采用了一种带扰动的多项式变异以提高种群的多样性。测试结果表明,该算法在收敛性和多样性方面优于其它优异算法。将该算法应用于液体动压滑动轴承的多目标优化设计,优化结果表明该算法具有较高的工程实用价值。In order to solve multi--objective optimization design problem of the hydrodynamic sliding bearing, an improved cellular genetic algorithm with differential operator is proposed. The se- lected individuals are ranked according to their Pareto dominance relationship and the improved differ- ential operator is applied based on the ranking. Then the improved differential operator is integrated into the cellular genetic algorithm to speed up convergence. And a polynomial mutation operator with disturbance is adopted to improve diversity. The test results reveal that the proposed algorithm outper- forms some state--of--the--art algorithms in terms of convergence and diversity. The engineering ex- ample demonstrates the algorithm is of high practical value.

关 键 词：液体动压滑动轴承 差分进化 元胞遗传算法 随机扰动 多目标优化 

分 类 号：TH133.31[机械工程—机械制造及自动化]