摘要
基于弹性流体动力润滑理论,以体积最小和膜厚比最大为目标函数,建立了齿轮传动多目标优化数学模型,并将问题转化为无约束单目标优化问题。针对遗传算法的局限性,采用整数编码和实数编码相结合的混合编码,调整了适应函数,设计了随机多父辈交叉方法,结合模拟退火算法,形成了混合遗传算法。在操作过程中,部分约束条件自动得到满足,大大减少了不可行解的产生。算例说明了该优化方法的有效性。
Based on the theory of elastohydrodynamic lubrication, multi-objective optimization model of gear transmission was established, in which the minimum volume and the maximum film thickness ratio were taken as objective functions, and the problem was transformed to unconstrained single-objective optimization problem. Aimed at the limitation of genetic al- gorithm, the hybrid coding combining integer coding with real coding was adopted, the fitness function was adjusted, the random multi-parent crossover was designed, and simulated annealing algorithm was combined with, then the hybrid genet- ic algorithm was formed. In the operation process, some restrictions were satisfied automatically, the number of infeasible solutions was reduced. Example shows that the algorithm is efficient.
出处
《润滑与密封》
CAS
CSCD
北大核心
2009年第12期69-71,59,共4页
Lubrication Engineering
关键词
齿轮传动
弹流润滑
多目标优化
混合遗传算法
gear transmission
elastohydrodynamie lubrication
multi-objective optimization
hybrid genetic algorithm
