Thermodynamics-Based Machine Learning and Data-Driven Computing for Inelastic and Fracture Modeling
Jiun-Shyan (JS) Chen
Department of Structural Engineering
Department of Mechanical & Aerospace Engineering
University of California, San Diego
Faculty host: Ming-Chen Hsu
Seminar on April 22, 2025, at 11:00 AM in 2004 Black Engineering
Abstract
While many machine learning algorithms have gained popularity in various real-world applications, pure black-box data-driven models require enormous datasets, limiting their applicability to problems with scarce measurable data. This presentation introduces recent advances in thermodynamics-based machine learning approaches based on universal thermodynamics principles, where the internal state variables essential to the physics are inferred automatically from the hidden state of the deep neural network. An extension of this approach is using the thermodynamics-based machine learning algorithms to enhance the effectiveness in inelastic and fracture modeling. In this approach, standard approximation spaces, such as those formed by the finite element or meshfree basis functions, are enriched by the neural network constructed basis functions under a Partition of Unity framework. The proposed neural network enhanced Partition of Unity and the feature-encoded transfer learning form a neural net adaptive approximation framework for solving general PDEs and its particular applications to inelastic and fracture modeling. These unique combinations of machine learning techniques and advanced computational methods have expanded the horizon of scientific computing beyond what the conventional computational methods can offer.
J. S. Chen is the William Prager Chair Professor and Distinguished Professor of Structural Engineering Department, Mechanical & Aerospace Engineering Department, and the Founding Director of Center for Extreme Events Research at University of California San Diego (UCSD). He is also the Yushan Fellow and Visiting Chair Professor of National Cheng-Kung University in Taiwan. Before joining UCSD in 2013, he was the Chancellor’s Professor of UCLA Civil & Environmental Engineering Department, Mechanical & Aerospace Engineering Department, and Mathematics Department, where he served as the Department Chair of Civil & Environmental Engineering during 2007-2012. J. S. Chen’s research is in computational mechanics, meshfree methods, multiscale materials modeling, machine-learning-enhanced computational mechanics, and physics-informed data-driven computing. He is the Past President of US Association for Computational Mechanics (USACM) and the Past President of ASCE Engineering Mechanics Institute (EMI). He has received numerous awards, including the John von Neumann Medal from the US Association for Computational Mechanics (USACM), the Belytschko Medal from USACM, the Raymond D. Mindlin Medal from ASCE Engineering Mechanics Institute (EMI), the Computational Mechanics Award from the International Association for Computational Mechanics (IACM), the Grand Prize from Japan Society for Computational Engineering and Science (JSCES), the Ted Belytschko Applied Mechanics Award from ASME Applied Mechanics Division, the Computational Mechanics Award from Japan Association for Computational Mechanics (JACM), the ICACM Award from International Chinese Association for Computational Mechanics (ICACM), among others. He is the Fellow of USACM, IACM, ASME, EMI, SES, ICACM, and ICCEES. He received PhD in Theoretical & Applied Mechanics from Northwestern University.
This seminar counts towards the ME 6000 seminar requirement for Mechanical Engineering graduate students.