Structural optimization of a small earth remote sensing satellite.

This article focuses on the structural sizing optimization of a small earth remote sensing satellite with a primary structure based on aluminum honeycomb sandwich plates. It extends the optimization problem by simultaneously considering main static and dynamic loads during the launch phase, using results from static, buckling, modal, and harmonic response finite element analyses conducted via ANSYS Workbench. The optimization problem is formulated with six design variables related to panel thicknesses, aiming to minimize structural mass while satisfying constraints on stress, displacement, buckling load factor, and natural frequency. Various optimization techniques are employed, including an Adaptive Multiple-Objective (AMO) algorithm in ANSYS, response surface methods using polynomial regression and kriging meta-models, and MATLAB-based genetic and sequential quadratic programming algorithms. All approaches achieve approximately 30% reduction in structural mass (about 22 kg), with kriging providing higher regression accuracy and genetic algorithms showing better global optimization capability than gradient-based methods. The study highlights the importance of meta-models in efficiently solving complex satellite structural optimization problems under combined static and dynamic loading conditions.