Vehicle System Dynamics ( IF 3.6 ) Pub Date : 2021-03-08 , DOI: 10.1080/00423114.2021.1896011 Hanwei Gao 1, 2 , Louis Jézéquel 1 , Eric Cabrol 2 , Bernard Vitry 2
The simulation of vehicle chassis system is widely deployed at the early stage of a new car development project in order to verify if the technical specifications such as durability and comfort of current design are fulfilled. However, the numerical validations of specifications are usually carried out independently by project teams, which may fail to converge the final design at downstream phase of projects. This paper proposes an early stage multi-objective optimisation method which is able to search a set of best trade-offs between durability and comfort before the deployment of the chassis parameters. Virtual proving ground simulation is applied as a uniformed numerical validation platform, following the same process as in physical experiments, including global and local damage calculation, transfer function estimation and other comfort criteria. By realising the design of experiments on front and rear axles' global curves and key components in a deployable design space, the Pareto front can be obtained in several optimisation iterations. An application case on an example project shows the method succeeds in revealing the general trends between these two chassis specifications with acceptable simulation costs. The data-mining methods further demonstrate the design preferences and orientations of each group in the simulation database.
中文翻译:
通过虚拟试验场模拟在项目早期的底盘耐用性和舒适性权衡
车辆底盘系统的仿真在新车开发项目的早期广泛部署,以验证是否满足当前设计的耐用性和舒适性等技术规格。然而,规范的数值验证通常由项目团队独立进行,这可能无法在项目的下游阶段收敛最终设计。本文提出了一种早期多目标优化方法,该方法能够在部署底盘参数之前搜索一组耐用性和舒适性之间的最佳权衡。虚拟试验场模拟被用作统一的数值验证平台,遵循与物理实验相同的过程,包括全局和局部损伤计算,传递函数估计和其他舒适标准。通过在可展开的设计空间内实现对前后桥全局曲线和关键部件的实验设计,经过多次优化迭代即可得到帕累托前沿。一个示例项目的应用案例表明,该方法成功地揭示了这两个底盘规格之间的总体趋势,且仿真成本可接受。数据挖掘方法进一步展示了模拟数据库中每个组的设计偏好和方向。一个示例项目的应用案例表明,该方法成功地揭示了这两个底盘规格之间的总体趋势,且仿真成本可接受。数据挖掘方法进一步展示了模拟数据库中每个组的设计偏好和方向。一个示例项目的应用案例表明,该方法成功地揭示了这两个底盘规格之间的总体趋势,且仿真成本可接受。数据挖掘方法进一步展示了模拟数据库中每个组的设计偏好和方向。