Answers are grounded in this paper and cite their sources.
This article presents an evolutionary algorithm optimization framework designed to assist aerospace system-engineers in identifying optimal candidate launchers during the conceptual design stage. Employing a multi-disciplinary coupling between aerodynamics, propulsion and the trajectory of the system, the performance is comprehensively assessed to find the optimal design. A machine learning driven regression analysis is applied to an archive created during the optimization process by the Genetic Algorithm, aiming to identify and rank the most influential control variables for the performance parameter selected – and the approach is compared to a one-at-a-time (ONAT) sensitivity analysis. Aerodynamic evaluations are conducted utilizing a semi-analytical method, while propulsion coefficients are determined through CEA software, with simulations performed in a 3-degree-of-freedom equation of motion model. This approach enhances the understanding of design variable contributions, providing valuable insights for refining and optimizing aerospace conceptual designs
No takes yet. Share an insight, caveat, or question.
Silaidis et al. (2025) studied this question.