【作者】 杨荣山；

【导师】 黄向东；

【作者基本信息】 华南理工大学 ， 车辆工程， 2009， 博士

【摘要】 底盘是轿车开发的平台,是保证良好的操纵稳定性、行驶平顺性和乘坐舒适性的基础。在轿车底盘系统开发中需要考虑上述多方面性能,需要将上述多个性能目标进行权衡比较和优化。目前多目标优化理论在汽车开发中的实际应用还存在诸多困难,因此本文侧重于研究如何将多目标优化理论和方法应用在轿车底盘平台开发的工程实际,具体内容涉及:1、对目前底盘平台开发中典型的系统动力学问题及其分析方法和评价目标进行分析和阐述,并提出基于操纵稳定性和行驶平顺性及路面通过悬架系统引起车内噪声的多目标协同优化的思想。2、分析多目标优化问题的定义及其解的特点,重点研究和阐述Pareto最优解的概念及其在多目标优化中的重要性;同时对多目标优化问题求解的理论方法进行了研究,并分析了常用的多目标优化算法:如线性加权法和多目标遗传算法。3、结合某企业自主品牌研发项目对某轿车底盘平台进行建模,针对操纵稳定性和行驶平顺性及车内噪声分别建立多体动力学模型、有限元声构耦合模型,并应用多目标优化工具集成多体动力学、有限元声构耦合仿真软件开展协同优化,以底盘系统的弹性元件(弹簧、减振器和横向稳定杆)特性为设计变量,以提高双移线操纵的性能、随机路面下的行驶平顺性和降低路面通过底盘悬架引起的车内噪声为目标对该轿车底盘系统进行多目标优化,取得了不错的效果。同时还运用线性加权和法的多目标优化处理方法对悬架运动特性进行优化,提高了底盘悬架系统的性能。4、在求解大型、复杂的多目标优化的工程问题时,由于大量耗时的高精度分析计算,导致算法的优化效率很低。为此,可以在整个优化寻优区域内建立比较精确的目标及约束的近似模型来代替仿真计算模型,以加快优化进程。文中运用基于响应面的近似模型对基于操稳和行驶平顺性的多目标问题进行了优化研究,研究结果表明:近似模型不仅能得到比较满意的优化结果,而且避免了大量耗时的高精度分析计算,显著提高了多目标优化的效率。更多还原

【Abstract】 Chassis is the platform of a car development, and it is the foundation of the car to guarantee a good handling, ride comfort and NVH performance. Both handling and ride comfort play an important role in the performance of a vehicle, usually influencing the customers’to purchase a car. It’s necessary to optimize the above multi-performances of a car chassis system, but there are difficulties in the application of multi-objective optimization theory and method to the car chassis design. This thesis focuses on the study & application of multi-objective optimization to the engineering practice, which mainly deals with:1. The typical problems and analysis methods of vehicle system dynamics which related with car chassis are introduced, and the idea of cooperative optimization among handling, ride and cabin noise level is presented.2. The definition of multi-objective optimization and the character of its solution are explained and the concept of Pareto solution set of the multi-objective optimization problem is the emphasized. Meanwhile the theory of multi-objective optimization problem is researched, and the algorithm of multi-objective optimization problem is analyzed, such as the linear weighting factor method and genetic algorithm of multi-objective problem.3. Based on the owned-brand passenger car development project of an enterprise, modeling of multi-body system is used to simulate the handling and ride performance, and the FEM model of acoustic-structure is built to calculate the cabin noise of the car. Further more, the multi-objective optimization tool is utilized to integrate the above multi-body software and FEM solver for cooperative optimization of chassis multi-objective optimization problem, for improving handling, ride comfort and minimize the cabin noise, according to the double lane change maneuver and riding on a random road. In the optimization process the chassis elastic element characteristics such as spring, shock absorber and anti-roll bar are set as the design variables. The optimization results indicate that the chassis performance has an obvious improvement in handling and ride comfort, even the cabin noise level.4. For the complex optimization problem, the conventional optimization method is not feasible due to huge number of iterations and time consumption. Accordingly, in this thesis, a methodology is investigated to build an approximation model based on a vehicle multi-body model by response surface method (RSM) for optimizing the vehicle’s handling and ride comfort. The results indicate that the approximation methodology is efficient to the multi-objective trade-offs in vehicle development.更多还原