基于多目标遗传算法的燃油泵叶轮优化设计  

Fuel pump impeller optimization design based on multi-object genetic algorithm

作　　者：李奇敏[1] 蒋建新 李阳 Li Qimin;Jiang Jianxin;Li Yang(State Key Laboratory of Mechanical Transmission, Chongqing University, Chongqing 400044, China;Chongqing Wanlilianxing Industrial （Group） Co. Ltd. , Chongqing 409100, China)

机构地区：[1]重庆大学机械传动国家重点实验室,重庆400044 [2]重庆万力联兴实业(集团)有限公司,重庆409100

出　　处：《现代制造工程》2018年第6期121-126,共6页Modern Manufacturing Engineering

摘　　要：针对某型号燃油泵,采用CFD方法对其内部流场进行研究。运用SST k-ω计算模型和多重参考系技术,引入多目标遗传算法作为优化算法,并结合自适应权重方法,以出口流量和容积效率作为优化目标,对燃油泵叶轮的结构参数进行优化,解决了在优化迭代过程中单个优化目标出现负优化的现象。在设定的工况下,对比改进的优化方法的出口流量和容积效率,得知优化后的出口流量提升了3.90%,容积效率提高了0.53%。此种研究方法为汽车燃油泵的结构优化设计提供了较为合适的参考解决方案。The method of CFD numerical calculation was conducted for carrying out the study of a fuel pump inner flow field. As SST κ-ε calculation model and multiple reference frame technology applied, while outflow and volumetric efficiency were chosen as the optimization goal, the structural parameters of fuel pump impeller was optimized with multi-objective genetic algorithm and adaptive weight method. The negative optimization phenomenon of the individual target in the process is solved. Comparing the re- sults of outflow and volumetric efficiency data before and after improved optimization, the outflow is increased by 3.90 %, the vol- umetric efficiency is increased by 0.53 % ,respectively. This kind of method can provide a suitable referenced solution for struc- ture optimization design of pump.

关 键 词：多目标遗传算法 自适应权重 叶轮 结构优化 轴流泵 

分 类 号：TH312[机械工程—机械制造及自动化]