基于RSM与RBF的车身多目标轻量化应用研究

Applied Research of Multi-objective Lightweighting Based on RSM and RBF

文章以某自主品牌SUV白车身为研究对象,通过Isight为优化平台,搭建了白车身模态、弯曲刚度与扭转刚度的多目标轻量化优化的仿真分析模型。通过优化超拉丁方法设计试验,响应面法(RSM)和径向基神经网络(RBF)分别进行近似模型的计算,最后以Pointer方法全局优化。在保证精度的条件下,降低计算成本。在保证性能的前提下,减重6.11kg。

This paper takes the BIW of a self-owned brand SUV as the research object,and uses Isight as the optimization platform to build a multi-objective lightweight optimization simulation analysis model of modal,bending stiffness and torsional stiffness.Through the optimization of the Optimal Latin hypercube design experiments,the response surface methodology(RSM)and radial basis functions(RBF)were respectively used to calculate the approximate model.Finally,the Pointer method was used for global optimization.Under the condition of ensuring accuracy,reduce the calculation cost.On the premise of guaranteeing performance,weight loss 6.11kg.