摘要
针对车用燃料电池离心压缩机转子振动、强度和轻量化的结构设计问题,实现了多目标带约束的燃料电池空压机转子结构优化设计方法。针对某燃料电池离心空压机转子,基于不平衡力的谐响应分析获得转子样本的结构性能目标参数,运用神经网络程序建立有限元分析近似模型,以转子结构在150%设计转速范围内最大等效应力和挠度为目标、转子质量为约束,运用多目标灰狼算法程序进行了转子结构设计的多目标全局寻优。结果表明:描述性能目标与设计变量关系的神经网络近似模型准确度较高,与有限元计算所得性能目标的最大误差均小于8%,优化计算找到了压缩机转子在质量约束下的最优构形参数;优化后转子在前两阶临界转速附近的最大等效应力和最大挠度显著降低,转子的最大等效应力下降46.3%,最大挠度下降7.35%。
For the structural design problems of the vibration,strength and lightweighting of the rotor of vehicle fuel-cell centrifugal compressor,a multi-objective rotor structural optimization design method with constraints for fuel-cell compressor rotor was proposed.Taking the rotor of a two-stage fuel-cell centrifugal compressor as the optimization object,based on the analysis of harmonic response caused by the unbalance moment,the corresponding structural performance target parameters of the rotor samples were calculated.On this basis,the neural network program was used to establish the finite element analysis surrogate model.Taking the maximum equivalent stress and maximum deflection of the rotor structure within the 150%design speed range as the target and the rotor quality as the constraint,the multi-objective global optimization of the rotor structure design was carried out by using the multi-objective grey wolf algorithm.The results show that the surrogate model established by the neural network can accurately express the mapping relationships between the design variables and the performance objects,and the maximum error between the performance targets obtained by the surrogate model and finite element calculation is less than 8%.The optimal structural parameters of the compressor rotor under mass constraint are found by the optimization.After the optimization,the maximum equivalent stress and maximum deflection of the rotor near the first two critical speeds are reduced obviously.The maximum equivalent stress of the rotor decreases by 46.3%,and the maximum deflection decreases by 7.35%.
作者
肖军
黄文俊
XIAO Jun;HUANG Wenjun(State Key Laboratory of Compressor Technology,Hefei General Machinery Research Institute,Hefei 230031,China)
出处
《流体机械》
CSCD
北大核心
2023年第3期33-41,共9页
Fluid Machinery
基金
国家重点研发计划项目(2019YFB1504601)
合肥市自然科学基金项目(2021039)。
关键词
燃料电池
离心压缩机
转子
结构优化设计
多目标灰狼算法
fuel-cell
centrifugal compressor
rotor
structural optimization design
multi-objective grey wolf algorithm
