【题目】A Novel Method to Improve Numerical Convergence in Computational Damage and Failure Analyses of Heterogeneous Materials and Structures
【时间】2015年4月29日(周三)14:30-16:00
【地点】安中大楼A322
【摘要】
Heterogeneous materials such as composites and concrete structures typically exhibit complex, multiple damage events including multiple crack initiation at various scales from pristine materials, through their coupled growth, involving multiple bifurcation, coalescence, to eventual failure. Numerical simulations with explicit consideration of arbitrary crack-like damage events are challenging because in heterogeneous materials/structures it is often impossible to know the cracking locations a priori, yet correct and efficient treatment of crack coalescence and bifurcation is critical. There are several FE methods that can deal with arbitrary cracking in solids, e.g. the PoU-FEM, X-FEM, PNM, and A-FEM. However, numerical simulations of multiple crack initiation and propagation with these methods often suffer from numerical divergence, often without identification of the leading causes.
In this presentation, I will, through a simple 1D problem, to illustrate one of the leading causes to a class of unexplained numerical divergence in continuum damage and failure simulations with quasi-static formulation. With this revelation, a novel and very fast method of overcoming the loss of convergence will be presented. It will be shown that, when this method is incorporated into the A-FEM formulation, it can greatly improve the numerical convergence when unstable multiple local damage events occur. In particular, the ability of this method in overcoming interruption of simulations when short intervals of unstable crack growth arise in multiple crack growth simulations has significant practical importance – it has the potential of making high-fidelity material/structural failure analyses as routine exercises of engineering design or material/ structural optimization.
【报告人简介】
Qingda Yang, Ph.D.
Associate Professor
Dept. of Mechanical and Aerospace Engineering
University of Miami, Coral Gables, FL 33124-0630
Tel: (305) 284-3221; Fax: (305) 284-2580; Email: qdyang@miami.edu
Dr. Qingda Yang is an associate professor of Mechanical and Aerospace Engineering at the University of Miami (Coral Gables, FL). He obtained his B.S. (Engineering Mechanics, 1991) and M. S. (solid mechanics, 1994) from Zhejiang University, China, and a PhD (Mechanical Engineering, 2000) from University of Michigan at Ann Arbor. Prior to joining University of Miami, Dr. Yang worked for Rockwell Scientific Company (formerly known as Rockwell Science Center) as a solid mechanics scientist from 2000 to 2006.
Dr. Yang joined the faculty of the Department of Mechanical and Aerospace Engineering at the University of Miami in 2006 as an assistant professor. Dr. Yang’s recent research has focused mainly on developing multi-scale methodologies that can lead to realistic virtual testing and designing of complex heterogeneous materials and structures under general and/or extreme thermal-mechanical loading environments. His research has attracted funding from many federal agencies (NSF, NASA, AFOSR, ARO, NAVY, and DARPA) and industrial companies (Teledyne, Boeing, and Vextec). Dr. Yang is an author/coauthor of 70 peer-reviewed journal publications, 4 book chapters, and more than 30 refereed conference proceedings. He is an editorial board member for the Journal of Applied Composite Materials and the journal of Multifunctional Composites. He was a past Chair of the Composite and Heterogeneous Materials Committee in ASME (2010-2012), and is currently serving as an executive board member for the Florida Space Grant Consortium (FSGC). Dr. Yang is an award recipient of several professional and academic societies, including the Eliahu I. Jury Award for excellence in research from the College of Engineering of the University of Miami (2007), University of Miami Prevost’s General Research Award (2008), Rockwell Scientific’s Technical Excellence Awards (2001; 2004; and 2005); the Adhesion Society’s the Alan Gent best student paper award (1999); and the Ivor K. McIvor Award and the Doris Caddell Award from University of Michigan (1999, 2000).