Surrogate-based Shape Optimization Design for the Stable Descent of Mars Parachutes

Abstract

The crucial role that the supersonic parachute plays in space exploration missions has been widely recognized, as it directly impacts the safe landing of probes. However, parachute models with optimization on different aerodynamic performances often involve design conflicts with each other. Additionally, the parachute design focusing on a single point cannot fully adapt to different speed ranges during stable descent. This complexity makes it challenging to use traditional shape design methods, which rely on empirical knowledge, to address these coupled design issues. Faced with the design challenges of Mars parachutes, this study, inspired by aircraft aerodynamic optimization principles, establishes a shape design method specifically for the stable descent phase of Mars parachutes. The method combines numerical simulation and surrogate-based optimization strategies, aiming to enhance the overall performance during stable descent and meet various demands of different exploration missions. Meanwhile, by providing a rapid estimate of the shape during the design phase, the method significantly improves computational efficiency. The optimal models effectively balance comprehensive performance in the supersonic-transonic-subsonic speed domain by conducting shape optimization research on the disk-gap-band parachute using the surrogate-based optimization strategy. Also, it exhibits better deceleration and stability across the entire speed range compared to the base model, even when deviating from the design Mach number. Importantly, the advantages of canopy-only optimization for drag performance extend to the capsule-canopy two-body system, enhancing the drag performance of the canopy in the two-body system. This strategic approach reduces the transient calculation time for the two-body system, further improving computational efficiency. The method provides a practical and forward-thinking solution for the design of Mars parachutes.