基于混合试验设计的纯电动车多学科轻量化设计

Multidisciplinary Lightweight Design of Electric Vehicles based on Hybrid Experimental Design

为实现纯电动汽车结构轻量化设计,常采用基于试验设计和近似模型及优化技术的多学科优化设计。文章为了保证样本矩阵的正交性和均匀性,同时兼顾时间成本,对拉丁方试验设计进行改进,开发了混合试验设计算法。以某纯电动汽车开发为例,根据结构性能特点和工程经验对纯电动汽车白车身进行不同区域不同工况的轻量化设计,并利用混合试验设计方法生成兼顾正交性的均匀样本矩阵,建立近似模型并进行优化,整车减重8.8 kg,中间区域减重5.2 kg,减重率达6.4%。证明了采用混合试验设计方法可以生成具有较好效果的仿真方案,并减少样本数量,表明该算法具有工程实用性。结合分区域解耦优化策略,可以应用较少的资源和较低的时间成本,达到较好的减重效果。

To achieve lightweight design of electric vehicle sturcture,a multidisciplinary op⁃timization design based on experimental design,approximate models,and optimization techniques is often employed.In order to ensure the orthogonality and uniformity of the sample matrix while consid⁃ering time costs,an improved Latin square experimental design algorithm is developed.Taking the development of an electric vehicle as an example,the lightweight design of the body-in-white of the electric vehicle is carried out under different areas and load conditions based on structural perfor⁃mance characteristics and engineering experience.The hybrid experimental design method is used to generate a uniformly distributed sample matrix with orthogonality,and an approximate model is estab⁃lished and optimized.The vehicle weight is reduced by 8.8 kg,with a weight reduction of 5.2 kg in the middle area,and a weight reduction rate of 6.4%.This proves that the hybrid experimental design method can generate simulation schemes with good effects and reduce the number of samples,indicat⁃ing that the algorithm has engineering practicality.Combined with the strategy of partitioned decou⁃pling optimization,better weight reduction effects can be achieved with fewer resources and lower time costs.