基于遗传算法的动力总成悬置系统隔振优化  

Design Optimization of Power-train Mounting Systems Using the Genetic Algorithms Method

作　　者：娄小宝 陈达亮[2] 

机构地区：[1]浙江吉利汽车研究院有限公司 [2]中国汽车技术研究中心

出　　处：《汽车工程师》2018年第1期20-23,54,共5页Automotive Engineer

摘　　要：为了提高动力总成悬置系统的隔振性能,达到设计可行域内的全局最优解,以满足整车NVH性能的要求,在动力总成曲轴坐标系下建立了发动机悬置系统6自由度模型。以动力总成垂向和绕曲轴方向的振动解耦为目标函数,建立了悬置系统参数优化模型,应用序列二次规划法(SQP)和遗传算法(GA)分别对该模型进行了优化。实际悬置计算案例的结果表明,发动机悬置系统参数的优化是一个多目标非凸性优化问题,存在多极值点现象;GA法相比SQP法在解决此问题时表现出了优越性,而且不依赖于问题的初始值,能够获得设计可行域内的全局最优解,适用于汽车发动机悬置系统多目标优化设计问题。To improve powertrain mounting system vibration isolation performance, achieve optimum solution in the design feasible scope, and satisfy vehicle NVH performance requirements, a 6-degree freedom model under powertrain crankshaft coordinate system is established. Vibration decoupling is used as target function in the vertical and crankshaft axial directions of powertrain, mounting system parameter optimization model is built, using sequential quadratic programming （SQP） and genetic algorithm （GA） to optimize model is described respectively. The actual mounting calculation case results show that engine mounting system parameter optimization is a multiple objectives non-convexity optimization issue, with more extreme points, GA method has advantages in solving this case compared with SQP method, meanwhile it does not rely on case initial value, thus can obtain optimum solution in design feasible scope, and can be applicable to multiple objectives optimization design issues of vehicle engine mounting system.

关 键 词：动力总成悬置系统 隔振 优化设计 解耦 遗传算法 

分 类 号：U464.13[机械工程—车辆工程]