基于近似模型的在轨加注站贮箱多目标优化设计

Multi-objective optimization design of on-orbit refuel storage tank based on the approximate model

为了提高在轨加注站贮箱的结构稳定性并减小贮箱质量,提出了以推进剂装载质量为约束条件,贮箱质量、1阶固有频率、最大形变量为目标函数的多目标优化设计方法。通过参数化建模、模态分析、拉丁超立方试验设计、4阶响应面近似模型及NSGA-Ⅱ算法对贮箱进行多目标优化设计。与有限元计算结果对比表明:在满足推进剂装载质量的条件下,贮箱质量减少32.95%,1阶固有频率提高36.30%,最大形变量减少17.03%,说明所提出的优化设计方法能够精确并高效实现对贮箱多目标优化设计。

In order to improve the structural stability of on-orbit refuel storage tank and reduce its mass,the multi-objective optimization design is conducted,through which the propellant loading mass serves as the constraint,and the tank mass,the firstorder natural frequency and the maximum deformation constitute the objective functions. The design is carried out by means of parametric modeling,modal analysis,Latin hypercube experiment design,approximation model of the fourth-order response surface,and NSGA-II algorithm. The results,compared with those based on the finite-element calculation,show that when the requirements of propellant loading mass are met,the tank mass decreases by 32.95%,the first-order natural frequency increases by36.30%,and the maximum deformation reduces by 17.03%. Thereby,with the aid of this method,the multi-objective optimization design of storage tank is completed accurately and efficiently.