Explore cutting-edge applications of machine learning in structural mechanics through this one-hour lecture on reduced order modeling and inverse design of flexible structures. Delve into two representative problems showcasing the power of neural networks in capturing nonlinearity and solving complex inverse problems. Learn how neural ordinary differential equations can dramatically speed up simulations of highly deformable structures, using a slinky as a practical example. Discover a novel approach to designing shape-morphing structures that transition from 2D to 3D forms, employing symmetry-constrained active learning to navigate vast design spaces efficiently. Gain insights from Dr. M. Khalid Jawed, an Associate Professor at UCLA, on the intersection of structural mechanics, robotics, and artificial intelligence in creating programmable smart structures.
Reduced Order Modeling and Inverse Design of Flexible Structures by Machine Learning
Inside Livermore Lab via YouTube
Overview
Syllabus
DDPS | Reduced Order Modeling and Inverse Design of Flexible Structures by Machine Learning
Taught by
Inside Livermore Lab