一种卫星分步结构优化方法

A Satellite Multi-step Structure Optimization Method

在卫星结构设计中,为了在满足各种工况条件下有效实现减重,基于优化驱动设计的思想,文章提出了一种分步结构优化方法:第一步采用固体各项同性惩罚微结构法(SIMP)进行多工况下最小柔顺度的拓扑优化,通过将多工况下的优化目标加权,转化为单目标优化问题,优化求解得到最优构型;第二步,提取第一步优化方案的拓扑构型关键特征,以质量最小、整星结构基频最高为目标,采用带精英策略的非支配排序遗传算法(Non-dominated Sorting Genetic Algorithm-Ⅱ,NSGA-Ⅱ)对形状和尺寸参数进行多目标优化;第三步,根据得到的形状尺寸参数,考虑实际加工和设计要求,经过人工修改迭代,得到最终设计。此方法应用于天拓-3卫星的支撑腿结构设计中,实现了考虑多工况的拓扑优化和多目标的形状尺寸优化。按照工程设计制造要求进行处理,获得的优化方案在满足各项约束条件下使支撑腿结构有效减重44.57%,整星结构基频提高3.97%,说明该方法具有较强的工程应用价值。

During the design of satellite structure, in order to achieve weight reduction under vari-ous working conditions, a multi-step optimization method to optimize and design structure is pro-posed. First step, an optimal topology configuration is obtained by utilizing Solid Isotropic Mate-rial Penalty （SIMP） approach, considering minimization of compliance. The optimization objec-tive under multiple loading cases is transformed into a single objective. The second step is to ex-tract the topology configuration of the first-step, and the multi-objective optimization of shape and size parameters is achieved utilizing the Non-dominated Sorting Genetic Algorithm-II（NSGA- II） method. The third step, engineer conducts several rounds of redesign considering require-ments of manufacture and assemblage and a new design is given. This method is applied to the de-sign of the supporting leg of Tiantuo-3 satellite, which considering the topology optimization and multi-objective shape and size optimization under multi-loading cases. The mass reduction of sup-porting leg is 44. 57% and the satellite structural fundamental frequency is increased by 3. 97%, which indicates that the method has strong engineering application value.