Applied Machine Learning-Based Metal Additive Manufacturing for Lightweight Aerospace Components

Metal additive manufacturing (AM) is an emerging paradigm in aerospace design; AM enables fabrication of structures that are impractical through traditional methods. Additionally, these additive designs can utilize self-supporting Topology Optimization (TO) to drastically increase performance. The resulting mass-efficient designs reduce failure stress by up to three times compared to their subtractive counterparts. Created with Kriging-based engineering surrogates in a low-dimensional parameter space, traditional AM and AM+TO designs are simulated, manufactured, tested, and analyzed. This paper validates the compressive strength and vibrational first modes of these parts. Furthermore, they will be flown on a high-powered rocket for IREC 2025, pioneering the technology in applied student aerospace engineering.