Two-Stage Intelligent Layout Design of Curved Cabin Door

Abstract

As one of the most complex and critical components of spacecraft, the structural design of the curved cabin door faces two challenges. On the one hand, it is difficult to obtain innovative configurations for the cabin door in the preliminary design stage. On the other hand, the traditional optimization design algorithm is inefficient in the detailed design stage. In this paper, a two-stage intelligent method for the layout design of the curved cabin door is proposed. In the first stage, the innovative stiffener layout of the cabin door is obtained based on the topology optimization method. Then the mesh deformation method is used for rapid modeling and geometric reconstruction. In the second stage, a recently proposed powerful evolutionary algorithm, named elite-driven surrogate-assisted Covariance Matrix Adaptation Evolution Strategy (ES-CMA-ES), is employed to optimize the parameters of the cabin door and its surrounding thin-wall structure. To verify the effectiveness of the proposed method, a curved cabin door example from the spacecraft (cargo spaceship) is carried out. Compared with the traditional orthogrid stiffener design, the mass of the optimal design is reduced by 52.21% while satisfying the constraints, which indicates the excellent optimization ability of the proposed method and demonstrates huge potential for improving the carrying capacity and efficiency of the spacecraft.