摘要
为了降低车身质量,以朗逸230TSI款轿车白车身为研究对象,采用Radioss求解器对车身扭转刚度、弯曲刚度、模态进行分析。通过模态试验验证了车身有限元模型准确性,采用Plackett-Burman实验设计法选取18个部件厚度为设计变量,建立白车身径向基神经网络近似模型,以车身质量,一阶扭转频率为优化目标,车身扭转刚度,弯曲刚度和一阶固有频率为约束条件对白车身进行多目标优化。通过近似模型与有限元模型结果对比,验证了优化结果的准确性。结果表明,车身质量减少18.3kg,一阶扭转频率提高5.24Hz,优化效果显著。
In order to reduce the body mass,the body in white of the Langyi 230TSI car was taken as the research object,and the Radioss solver was used to analyze the torsional stiffness,bending stiffness and modality of the vehicle body.The finite element model accuracy of the vehicle body was verified by modal test.The Plackett-Burman experimental design method was used to select 18 component thicknesses as design variables,and a radial-body neural network approximation model of bodyin-white was established.The body mass,first-order torsional frequency were the optimization target,and body torsional stiffness,bending stiffness and first-order natural frequency were a constraint to multi-objective optimization of the body.The accuracy of the optimization results was verified by comparing the approximate model with the finite element model results.The results show that the body mass is reduced by 18.3 kg,and the first-order torsional frequency is increased by 5.24 Hz,and the optimization effect is remarkable.
作者
张功学
牛顾根
王德雨
郭珊珊
ZHANG Gong-xue;NIU Gu-gen;WANG De-yu;GUO Shan-shan(College of Mechanical and Electrical Engineering,Shaanxi University of Science&Technology,Shaanxi Xi’an710021,China)
出处
《机械设计与制造》
北大核心
2020年第11期60-63,68,共5页
Machinery Design & Manufacture
基金
陕西省科技厅自然科学基础研究计划(2014JM7264)。
关键词
白车身
径向基神经网络模型
多目标优化
轻量化
有限元法
Body-in-White
Radial Basis Neural Network Model
Multi-Objective Optimization
Lightweight
Fnite Element Method
