Efficient Deep Learning for Gradient-Enhanced Stress Dependent Damage Model

authored by: Xiaoying Zhuang, L. C. Nguyen, Hung Nguyen-Xuan, Naif Alajlan, Timon Rabczuk
Abstract: This manuscript introduces a computational approach to micro-damage problems using deep learning for the prediction of loading deflection curves. The location of applied forces, dimensions of the specimen and material parameters are used as inputs of the process. The micro-damage is modelled with a gradient-enhanced damage model which ensures the well-posedness of the boundary value and yields mesh-independent results in computational methods such as FEM. We employ the Adam optimizer and Rectified linear unit activation function for training processes and research into the deep neural network architecture. The performance of our approach is demonstrated through some numerical examples including the three-point bending specimen, shear bending on L-shaped specimen and different failure mechanisms.
Organisation(s): Institute of Continuum Mechanics
External Organisation(s): Ton Duc Thang University
Vietnam National University Ho Chi Minh City
King Saud University
Type: Article
Journal: Applied Sciences (Switzerland)
Volume: 10
ISSN: 2076-3417
Publication date: 08.04.2020
Publication status: Published
Peer reviewed: Yes
ASJC Scopus subject areas: Materials Science(all), Instrumentation, Engineering(all), Process Chemistry and Technology, Computer Science Applications, Fluid Flow and Transfer Processes
Electronic version(s): https://doi.org/10.3390/app10072556 (Access: Open)
: Details in the research portal "Research@Leibniz University"

BibTeX

@article{2da68e15baaa4b0dbdff2e717b4e87cc,
title = "Efficient Deep Learning for Gradient-Enhanced Stress Dependent Damage Model",
abstract = "This manuscript introduces a computational approach to micro-damage problems using deep learning for the prediction of loading deflection curves. The location of applied forces, dimensions of the specimen and material parameters are used as inputs of the process. The micro-damage is modelled with a gradient-enhanced damage model which ensures the well-posedness of the boundary value and yields mesh-independent results in computational methods such as FEM. We employ the Adam optimizer and Rectified linear unit activation function for training processes and research into the deep neural network architecture. The performance of our approach is demonstrated through some numerical examples including the three-point bending specimen, shear bending on L-shaped specimen and different failure mechanisms.",
keywords = "Deep learning, Deep neural network, Gradient enhanced damage, Stress-level dependent damage model",
author = "Xiaoying Zhuang and Nguyen, {L. C.} and Hung Nguyen-Xuan and Naif Alajlan and Timon Rabczuk",
note = "Funding information: The authors extend their appreciation to the Distinguished Scientist Fellowship Program (DSFP) at King Saud University for funding this work. The authors extend their appreciation to the Distinguished Scientist Fellowship Program (DSFP) at King Saud University for funding this work.",
year = "2020",
month = apr,
day = "8",
doi = "10.3390/app10072556",
language = "English",
volume = "10",
journal = "Applied Sciences (Switzerland)",
issn = "2076-3417",
publisher = "Multidisciplinary Digital Publishing Institute",
number = "7",
}