Abstract: In consideration of insufficient stiffness, weld cracking, weight redundancy, and possibility of resonance for frame's natural frequency closing to non-sprung mass's excitation frequency by traditionally experiential design, multi-objective optimization design for frame structure performances of electric dump trucks was studied. The multi-objective topology optimization design was conducted on initial frames based on variable density method and SIMP interpolation model, considering fatigue life as constraints during optimization processes. A new frame topology structure was obtained, which might meet minimum compliance under static multi-conditions and maximum multi-low-order natural frequencies. The anti-fatigue and lightweight co-optimization designs were conducted on the new frames based on Kriging approximation model. Utilizing the non-dominated sorting genetic algorithm to find the optimal solution in global variables, the weight of optimized frame structures were reduced by 4.3%, compared with the masses of original frame structures. The final frames have more uniform stress distributions under the same loading, and better strength, stiffness, dynamic characteristics and longer fatigue life.

Key words: dumping truck, multi-objective, topology structure, fatigue life, lightweight

摘要： 针对传统经验式设计的电动轮自卸车车架结构存在刚度不足、焊缝开裂、重量冗余，以及自身固有频率与非簧载质量激励频率接近而可能发生共振等问题，开展车架结构性能多目标优化设计的研究。首先，基于变密度法和SIMP插值模型对原始车架结构进行了考虑疲劳寿命约束的多目标拓扑优化设计，得到了满足静态多工况下柔度最小和多阶低阶固有频率最大的拓扑优化结构。再次，基于Kriging近似模型的抗疲劳轻量化设计，利用非支配排序多目标遗传算法进行全局寻优，优化后的车架结构质量比原始车架结构质量减少4.24%，且车架结构在同样载荷作用下的应力分布更加均匀，车架强度、刚度、动态特性和疲劳寿命等性能均得到改善。

关键词: 自卸车, 多目标, 拓扑结构, 疲劳寿命, 轻量化