Multi-material Multi-physics Topology Optimization with Physics-informed Gaussian Process Priors скачать в хорошем качестве

Multi-material Multi-physics Topology Optimization with Physics-informed Gaussian Process Priors

Join our Discord channel or email studentchapter@usacm.org to receive updates on upcoming seminars and events. Speaker Details:Alex (Xiangyu) Sun Date: December 22th, 2025 Title: Multi-material Multi-physics Topology Optimization with Physics-informed Gaussian Process Priors Abstract: Traditional ML-based topology optimization (TO) methods suffer from high computational cost, spectral bias, and limited applicability to multi-material, multi-physics problems with non–self-adjoint objectives. To overcome these limitations, a mesh-free physics-informed Gaussian process (GP) framework for TO is introduced with the following advantages: 1. Represents the primary, adjoint, and design variables using independent GP priors, whose mean functions are parameterized by neural networks (NNs). 2. Estimates all parameters simultaneously by minimizing a unified loss constructed from adjoint objectives, potential energy functionals, and design constraints. 3. Effectively solves multi-physics, multi-material TO problems, producing super-resolution topologies with sharp interfaces and physically interpretable material distributions. Bio: Dr. Alex (Xiangyu) Sun is a postdoctoral scholar in Mechanical and Aerospace Engineering at the University of California, Irvine, supervised by Prof. Ramin Bostanabad. His current research integrates physics-informed machine learning, computational mechanics, and design optimization to enable efficient and interpretable topology optimization. Prior to UCI, Dr. Sun worked as a research scientist at the Corning Corp, and as a postdoctoral fellow at the University of Wisconsin–Madison. He earned his Ph.D. in Mechanical Engineering from Johns Hopkins University. Dr. Sun’s work spans scientific machine learning, multi-physics design, and high-rate experimental mechanics under extreme loading. His research vision is to establish an AI-enabled framework that unifies experimental methodologies and computational tools for designing engineered material–structure systems that perform reliably under extreme environments for aerospace and defense applications.