As global warming continues to impact every aspect of our lives, reducing carbon emissions in the construction and built environment sectors has become critically important. In this talk, Dr. Wang Hanmo will explore how innovative 3D-printed structural designs can significantly lower carbon emissions by reducing material usage, enhancing insulation, and improving shading performance during both the construction and operation phases. However, simulating the performance of these complex structures using traditional finite element methods (FEM) can be extremely time-consuming.
Dr. Wang will introduce the application of Graph Neural Networks (GNNs), a new class of machine learning models, to predict the performance of 3D-printed building structures more efficiently, supporting rapid and sustainable design iterations. Additionally, he will share his experience in integrating GNNs into large-scale applications, such as city-level sustainability projects, through startup initiatives. This talk will offer insights into the intersection of advanced manufacturing, AI, and sustainable urban development.