Hybrid Optimization Method for Structural Configuration and Size of CubeSat Deployer

In order to achieve the standardization and interchangeability of the CubeSat deployer minimum system units and meet the design requirements of high stiffness to mass ratio, a multi type variable hybrid optimization method based on orthogonal experimental design method, neural network proxy model and intelligent optimization algorithm is proposed to perform structural hybrid optimization on the CubeSat deployer. Firstly, the deployer components are topologically discretized with the presence or absence of discrete parts represented by introducing 0-1 topological variables, thus the topology optimization, shape optimization and size optimization variables of the components are mixed to achieve a unified solution. Secondly, based on the orthogonal experimental design method, the design points are selected from the hybrid variables space of the components and substituted into the finite element simulation software to obtain corresponding structural response values such as mass, stress and deformation as training samples. Thirdly, the neural network proxy model technology is used to establish a proxy model of the functional mapping relationship between design variables and structural response values based on training samples, and the prediction accuracy of the proxy model is tested using test samples, with an error of less than 5%. Finally, with the neural network proxy model as the fitness calculation function, minimizing structural quality as the goal and stiffness as the constraint condition, an intelligent optimization algorithm is used to solve the optimal solution of the hybrid design variables. This paper uses a hybrid optimization strategy to simultaneously solve topology, shape and size variables, thereby obtaining a global optimal solution. The optimization results show that the mass of deployer only accounts for 17% of the total mass of CubeSat and deployer, and the deformation of the deployer is only 0.468mm, meeting the design requirements of high stiffness to mass ratio of the deployer. The optimized 6U CubeSat deployer has been successfully applied in orbit after ground overload, sine vibration, random vibration and impact tests.