The National Renewable Energy Laboratory has published a new study examining the use of advanced machine learning and 3D printing techniques to save costs and size in wind turbine generators.
Wind Energy Technologies Office
July 14, 2021
minute read time
The U.S. Department of Energy’s (DOE’s) National Renewable Energy Laboratory (NREL) published a study that examines using advanced machine learning and 3D printing technologies to design lightweight wind turbine generators with higher power density. Taller, more efficient and reliable wind turbines are expanding in the wind market and accessing new, harsher environments—especially those offshore. These turbines currently rely on direct-drive generators that have historically faced challenges in minimizing weight, size, and use of critical rare-earth materials.
NREL’s study assesses weight- and cost-saving opportunities for the International Energy Agency 15-MW direct-drive generator by using novel single- and multi-material designs. The team used new software, Manufacturing and Additive Design of Electric Machines enabled by 3-Dimensional printing and Advanced Machine Learning—or MADE3D-AML—to quickly explore and identify a range of designs that optimize magnet distribution and materials to maximize power per unit weight. The findings provide new opportunities to accelerate developing low-cost, lightweight, multi-material designs for wind turbine generators. The authors presented findings virtually at the Conference for Wind Power Drives and at The Minerals, Metals & Materials Society Annual Meeting.