传统油底壳设计往往采用金属材料，为了充分发挥油底壳的性能，提出基于NSGA-Ⅱ算法的油底壳轻量化设计方法。首先，建立油底壳的有限元模型，并进行自由模态分析，对原钢制油底壳进行模态测试，对比分析前六阶模态频率及振型，验证有限元模型的准确性。其次，基于等刚度近似理论，对铝合金材料油底壳选取7个典型壁厚作为设计变量，以油底壳的一阶固有频率和质量为优化目标，应用最优拉丁超立方试验设计方法对设计变量进行贡献度分析。最后，基于Isight参数优化软件分别构建优化目标与影响因素之间的RBF近似模型和RSM近似模型，经分析对比之后选择RSM近似模型进行后续优化，基于NSGA-Ⅱ算法在代理模型内进行全局寻优，获取了一组使油底壳一阶模态频率和质量最优的预测值。研究结果表明，RSM近似模型的预测值与仿真试验结果基本吻合，优化后一阶模态频率提升22.9%，质量降低37.6%。

Traditional oil sump designs usually adopt metallic materials. To give full play to the performance of oil sump, a lightweight design of oil sump was proposed based on NSGA-Ⅱ algorithm. A finite element model of oil sump was first established, and a free modal analysis was carried out. Modal tests were conducted on the original steel oil sump, and the first six modal frequencies and mode shapes were compared and analyzed to verify the accuracy of finite element model. Based on the equal stiffness approximation theory, seven typical wall thicknesses of the aluminum alloy oil sump were selected as design variables, and the first-order natural frequency and mass of the oil sump were set as the optimization objectives. The optimal Latin hypercube experimental design method was applied to analyze the contribution of design variables. Finally, based on the Isight parameter optimization software, RBF and RSM approximation models between the optimization objectives and influencing factors were constructed respectively. After analysis and comparison, RSM approximation model was selected for subsequent optimization. A global optimization was performed within the surrogate model based on NSGA-Ⅱ algorithm, and a set of predicted values for the optimal combination of first-order modal frequency and mass of the oil sump were obtained. The research results show that the predicted values of RSM approximation model are basically consistent with the simulation test results. After optimization, the first-order modal frequency increases by 22.9%, and the mass reduces by 37.6%.