Article
Materials Science, Multidisciplinary
Minh-Quy Le
Summary: Nodal stresses were used to analyze the mode-I J-integral of single edge- and center-cracked plates with initial crack length-plate width ratios ranging from 0.1 to 0.5 in peridynamics. The computed J-integral values on six different contours differed by about 1% and were compared with analytical methods, finite element analysis, and previous peridynamic studies. Simulation results demonstrated high accuracy of the mode-I J-integral obtained via peridynamic stresses.
INTERNATIONAL JOURNAL OF FRACTURE
(2023)
Article
Engineering, Mechanical
Pengfei Jia, Kai Huang, Hongjun Yu, Takahiro Shimada, Licheng Guo, Takayuki Kitamura
Summary: This paper proposes a novel atomic J-integral calculation method, which calculates the local displacement gradient and stress field for discrete models, and verifies its effectiveness and path-independency. It is also confirmed that this method is dimension-independent even at ultra-small scales.
THEORETICAL AND APPLIED FRACTURE MECHANICS
(2022)
Article
Mechanics
Longkun Lu, Zhanli Liu, Zhuo Zhuang
Summary: This paper explores the physical meanings of two incremental J-integrals for crack growth in elastic-plastic materials, one based on free energy density and the other based on internal energy density. It indicates that energy balance criteria must be satisfied during crack growth in an elastic-plastic solid, and considers a cracked body as containing two independent thermodynamic systems.
ENGINEERING FRACTURE MECHANICS
(2022)
Article
Engineering, Mechanical
Suhib Abu-Qbeitah, Mahmood Jabareen, Konstantin Y. Volokh
Summary: In this study, a finite element formulation is developed to model quasi-static crack propagation in hyperelastic solids using the material-sink approach. Breakage of molecular bonds leads to material separation and appearance of new crack surfaces. The diffusion of bond breakage causes localized material loss. To account for this, mass density is considered as a variable that decreases in the damaged area. Mathematically, mass balance is included as an additional law to regularize the computational model. The developed finite element formulation has displacement and density degrees of freedom and a monolithic approach is applied for stable solution of the nonlinear problem. Numerical examples demonstrate the robustness of the proposed approach for modeling aneurysm material fracture.
INTERNATIONAL JOURNAL OF MECHANICAL SCIENCES
(2023)
Article
Mechanics
Otmar Kolednik, Masoud Sistaninia, Stefan Kolitsch
Summary: This paper presents a modified procedure for J-integral testing on micro-scale cantilever beam specimens. The study determines the plastic eta-factor, eta pl, for the cantilever beam specimens as a function of the relative crack length, relative cantilever length, relative base distance, and load. The results indicate that the constant value of eta pl 2 is only applicable for longer crack lengths, and for smaller crack ratios, eta pl can be much smaller and dependent on the base distance ratio. A fitting procedure is proposed to determine the correct value of eta pl and modifications to the J-integral testing procedure for cantilever beam specimens are suggested. A numerical validation of the procedure is also provided.
ENGINEERING FRACTURE MECHANICS
(2023)
Article
Engineering, Mechanical
Zihao Yu, R. K. L. Su, Hongniao Chen, Jie Shen, D. K. L. Tsang
Summary: In this study, three-point bending tests were conducted on nuclear graphite beams to investigate fracture properties. Electronic speckle pattern interferometry (ESPI) was used to measure deformation, and numerical simulations were performed to understand crack growth. The results of the experiments and simulations were compared, confirming the validity of the models. Additionally, fracture resistance was evaluated experimentally and compared with empirical formulas, and critical crack opening displacements and J-integral at the fracture initiation were determined.
THEORETICAL AND APPLIED FRACTURE MECHANICS
(2023)
Article
Mechanics
Ali Shivaie Kojouri, Haniyeh Khosravi Rikaee, Kalliopi-Artemi Kalteremidou, Danny Van Hemelrijck
Summary: This study examines the capability of the digital image correlation method to determine the Jintegral for polymers with elastic and elastic-plastic behavior. To achieve this, single-edge notch tension specimens manufactured out of PMMA and HDPE are tested under pure mode I loading conditions, and the displacement field at the surface of each specimen is obtained using DIC. The J-integral for each specimen is then computed by combining the higher-order singular and non-singular terms derived from DIC.
ENGINEERING FRACTURE MECHANICS
(2023)
Article
Materials Science, Multidisciplinary
Margi Gajjar, Himanshu Pathak
Summary: This study investigates the influence of plasticity graded property and thermal boundary conditions on the fracture parameter J-integral using the extended finite element method. Various cracked geometries subjected to thermal and thermo-mechanical loads are simulated for stress contours and J-integrals. The results show the accuracy and effectiveness of the presented computational approach.
PROCEEDINGS OF THE INSTITUTION OF MECHANICAL ENGINEERS PART L-JOURNAL OF MATERIALS-DESIGN AND APPLICATIONS
(2021)
Article
Engineering, Geological
Edoardo Pezzulli, Morteza Nejati, Saeed Salimzadeh, Stephan K. Matthai, Thomas Driesner
Summary: This article revisits the formulation of the J-integral in hydraulic fracture mechanics and presents two novel contributions. The first contribution is two variations of the J-integral that accurately predict viscosity-dominated propagation. The second contribution is a methodology to extract the propagation velocity from the energy release rate applicable throughout the toughness-viscous propagation regimes. These techniques are combined to form an algorithm capable of quickly converging on the location of the fracture front independently to the toughness-viscous regime of propagation.
INTERNATIONAL JOURNAL FOR NUMERICAL AND ANALYTICAL METHODS IN GEOMECHANICS
(2022)
Article
Materials Science, Multidisciplinary
N. Leitao, F. A. Gilabert
Summary: The proposed methodology based on J-Integral concept allows for accurate assessment of fracture toughness under different loading scenarios, with the use of Finite Element modeling and Digital Image Correlation technique. By simulating various loading conditions, it is possible to track the propagation and arrest of dynamic cracks, providing detailed correlation between fracture energy evolution and stress field at each material point.
JOURNAL OF THE MECHANICS AND PHYSICS OF SOLIDS
(2021)
Article
Engineering, Mechanical
Shuting Miao, Peng-Zhi Pan, Wenbo Hou, Mei Li, Zhenhua Wu
Summary: Static fracture testing was conducted on ISRM-suggested semi-circular bend specimens using the digital image correlation technique. Two displacement-based methods were employed to measure the fracture toughness of different rocks and their effectiveness was verified with synthetic images. The results showed that the displacement-based methods provide reliable fracture toughness results and have the ability to consider the existence of the fracture process zone above the notch tip.
THEORETICAL AND APPLIED FRACTURE MECHANICS
(2022)
Article
Mechanics
Nikolaus Bechler, Thomas Seifert
Summary: In this paper, a J-integral is derived for temperature-dependent elastic-plastic materials with incremental plasticity, and is implemented using the equivalent domain integral method for assessment of three-dimensional cracks. The derived J-integral considers contributions from inhomogeneous temperature fields, temperature-dependent elastic and plastic material properties, as well as gradients in the plastic strains and the hardening variables. Results obtained for a plate with different crack configurations and thermal loading conditions demonstrate the applicability of the derived J-integral for crack assessment in thermomechanically loaded components.
ENGINEERING FRACTURE MECHANICS
(2023)
Article
Materials Science, Multidisciplinary
Christer Stenstrom, Kjell Eriksson
Summary: The study extends the calculation of J-integral in peridynamics from the J-contour integral to the J-area integral, showing that the latter is less sensitive to local disturbances. Both formulations have straightforward implementation in peridynamics, with similar scope and applicability.
INTERNATIONAL JOURNAL OF FRACTURE
(2021)
Article
Engineering, Multidisciplinary
Mojtaba Armandei, Diego F. Sarzosa Burgos, Claudio Ruggieri, S. Luise Guilherme Tomba, Neilon de Souza da Silva
Summary: In this study, a regression machine learning approach is used to propose a new equation for estimating J-integral in circumferential surface clad pipelines. Training data is obtained from simulations using a parametric finite element model generator script. The generated models are used to develop a machine learning model for fracture assessment. The aim of this study is to provide an easy-to-use regression model that encompasses various geometrical and material parameters and to validate it against different approaches.
INTERNATIONAL JOURNAL OF PRESSURE VESSELS AND PIPING
(2022)
Article
Engineering, Mechanical
Weigang Wang, Wei Yang, Chun-Qing Li
Summary: This paper aims to determine a normalized fully plastic J-integral solution to mixed-mode fracture caused by inclined surface cracks, and develops a set of predictive formulas. The study finds that an increase in strain hardening exponent can increase the difference of normalized fully plastic J-integrals in pipes with different wall thicknesses.
ENGINEERING FAILURE ANALYSIS
(2021)
Article
Engineering, Biomedical
Fernanda F. Fontenele, Nikolaos Bouklas
Summary: In this study, a viscoelastic-plastic constitutive model was developed to analyze the effects of inelasticity and energy dissipation in collagen fibrils, specifically in the context of cyclic loading and overload. The model was validated by comparing the stress-stretch data obtained from experiments with single fibrils subjected to cyclic loading conducted by Liu et al. (2018).
ACTA BIOMATERIALIA
(2023)
Article
Water Resources
Carlos A. S. Ferreira, Teeratorn Kadeethum, Nikolaos Bouklas, Hamidreza M. Nick
Summary: This study proposes the adaptation and application of conditional generative adversarial networks (CGAN) for upscaling the permeability of single fractures. The results show that the framework employing CGAN provides accurate capture of both the permeability angle and anisotropy of discrete fractures, with a substantial reduction in computational time.
ADVANCES IN WATER RESOURCES
(2022)
Article
Computer Science, Interdisciplinary Applications
T. Kadeethum, D. O'Malley, Y. Choi, H. S. Viswanathan, N. Bouklas, H. Yoon
Summary: In this work, the previous approach of conditional generative adversarial networks (cGAN) is extended to time-dependent problems using the concept of continuous cGAN (CcGAN). The proposed framework shows sufficient accuracy and computational speed-up, enabling real-time reservoir management and robust uncertainty quantification in poroelastic problems.
COMPUTERS & GEOSCIENCES
(2022)
Article
Materials Science, Multidisciplinary
Yu-Sheng Lo, Thomas J. R. Hughes, Chad M. Landis
Summary: The phase-field modeling framework for linear elastic fracture mechanics problems is modified to analyze crack growth in large structures. The approach replaces sharp crack surfaces with a diffuse fracture zone, which is characterized by a length scale. However, extending the model to large structures presents challenges due to meshing requirements for the crack length scale. A new formulation is introduced to decouple the phase-field length scale from the physical process zone length scale.
JOURNAL OF THE MECHANICS AND PHYSICS OF SOLIDS
(2023)
Article
Materials Science, Multidisciplinary
Chad M. Landis, Rui Huang, John W. Hutchinson
Summary: This paper investigates the stability issues of a dielectric elastomeric layer bonded to a stiff substrate and subjected to a voltage difference. The study focuses on the competition between wrinkling and creasing, which is influenced by the equi-biaxial pre-stretch of the layer. Numerical simulations reveal the existence of localized crease solutions and predict the presence of a finite energy barrier.
JOURNAL OF THE MECHANICS AND PHYSICS OF SOLIDS
(2022)
Article
Engineering, Mechanical
Jaemin Kim, Erik Mailand, Mahmut Selman Sakar, Nikolaos Bouklas
Summary: Processes such as embryonic development and wound healing involve complex interactions between cellular contractility, migration, extracellular matrix (ECM) mechanics, and remodeling. However, the regulatory principles underlying the coupled cell migration and contractility in dynamically morphing tissues are not well understood. In this study, a mechanosensitive transient nonlinear theory and a finite element implementation were proposed to capture the interplay between cell migration and contractility in 3D ECM. Simulation results from microtissue experimental setups demonstrated the changes in tissue shape and cell concentration for wounded and intact microtissues. This theoretical model and computational framework provide insights into experimental data and facilitate hypothesis-driven research.
EXTREME MECHANICS LETTERS
(2023)
Article
Engineering, Multidisciplinary
Jan Niklas Fuhg, Craig M. Hamel, Kyle Johnson, Reese Jones, Nikolaos Bouklas
Summary: The development of accurate constitutive models for materials undergoing path-dependent processes remains a challenging task in computational solid mechanics. Data-driven modeling approaches have been proposed to establish stress-evolution laws using machine learning representations. However, these approaches require a large amount of data and do not fully enforce necessary thermodynamic principles. In this work, a hybrid framework is proposed that combines classical phenomenological models and data-driven models to handle different amounts of available data and response complexity.
COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING
(2023)
Article
Mechanics
Jan Niklas Fuhg, Amelie Fau, Nikolaos Bouklas, Michele Marino
Summary: By integrating machine learning techniques, a data-driven correction term is added to a phenomenological yield model, enhancing its performance in plasticity formulation. Similar results can be obtained using Support Vector Regression, Gaussian Process Regression, and Neural Networks, ensuring the convexity of the model-data yield functions. A highly anisotropic yield response with tension/compression asymmetries can be successfully reproduced by incorporating a limited number of synthetic data points.
EUROPEAN JOURNAL OF MECHANICS A-SOLIDS
(2023)
Article
Engineering, Mechanical
Solon Tsimpoukis, Stelios Kyriakides, Chad M. Landis
Summary: This article presents a new experimental setup that allows for precise control of thermal and mechanical loads, and simultaneous monitoring of the temperature and the full-field deformation of small SMA structures. The experiments reveal that the velocities of the phase transformation fronts depend on the rate at which heat is removed/added by the controlled environment. The simulation also demonstrates the need for further enhancement of the heat exchange between the structure and the environment.
INTERNATIONAL JOURNAL OF PLASTICITY
(2023)
Article
Mechanics
Mohammad A. Ansari, Rui Huang, Kenneth M. Liechti
Summary: Cohesive zone models are commonly used in interfacial fracture analyses with large damage process zones. The accurate representation of interfacial traction-separation relations is crucial for the successful implementation of cohesive zone modeling. In this study, we experimentally extract mixed-mode traction-separation relations by comparing different loading conditions using laminated beam specimens. A mixed-mode double cantilever beam model with linear normal and shear interactions is developed and validated through finite element simulations. Rotation control is found to be the optimal loading configuration for extracting the traction-separation relations.
INTERNATIONAL JOURNAL OF SOLIDS AND STRUCTURES
(2023)
Article
Biophysics
Byumsu Kim, Nikolaos Bouklas, Itai Cohen, Lawrence J. Bonassar
Summary: Tissue-engineered cartilage constructs have shown promise in treating cartilage defects, but the understanding of the local mechanical responses and cellular outcomes of collagen scaffolds is limited. This study investigated the effects of local mechanical responses on chondrocyte viability and found that local instabilities were directly correlated with cell death. It highlights the importance of understanding the architecture-dependent mechanical responses for tissue-engineered cartilage constructs.
JOURNAL OF BIOMECHANICS
(2023)
Article
Materials Science, Multidisciplinary
Sida Hao, Zhigang Suo, Rui Huang
Summary: When a thin layer of elastomer is sandwiched between two rigid blocks and pulled, it forms numerous small cracks, instead of a single large crack. This is due to the development of hydrostatic tension at a critical point, which causes a small crack to deform substantially and initiate growth. As the crack grows, the energy release rate decreases and the crack arrests, allowing other cracks to grow. The crack radius and spacing decrease with increasing normalized toughness of the elastomer.
JOURNAL OF THE MECHANICS AND PHYSICS OF SOLIDS
(2023)
Article
Mathematics, Applied
J. N. Fuhg, A. Karmarkar, T. Kadeethum, H. Yoon, N. Bouklas
Summary: This study presents a parametric PDE solver called Deep Convolutional Ritz Method (DCRM), which reduces the order of differential operators by minimizing energy functionals compared to residual-based methods. Based on studies involving the Poisson equation with spatially parameterized source term and boundary conditions, CNNs trained on labeled data outperform DCPINNs in convergence speed and generalization abilities. However, the surrogates generated from DCRM converge significantly faster than their DCPINN counterparts and prove to generalize faster and better than the surrogates obtained from both CNNs trained on labeled data and DCPINNs, suggesting the possibility of training PDE solution surrogates without labeled data using DCRM.
APPLIED MATHEMATICS AND MECHANICS-ENGLISH EDITION
(2023)
Article
Biochemistry & Molecular Biology
Tamara Rossy, Tania Distler, Lucas Meirelles, Joern Pezoldt, Jaemin Kim, Lorenzo Tala, Nikolaos Bouklas, Bart Deplancke, Alexandre Persat
Summary: The opportunistic pathogen Pseudomonas aeruginosa forms antibiotic-recalcitrant biofilms in the respiratory tract by contracting luminal mucus early during colonization. This study developed AirGels, tissue-engineered human lung models that mimic the airway mucosal environment, to investigate the process of biofilm formation in realistic conditions. The findings suggest that mucus provides a breeding ground for biofilms while protecting the epithelia.
Article
Computer Science, Information Systems
Teeratorn Kadeethum, John D. Jakeman, Youngsoo Choi, Nikolaos Bouklas, Hongkyu Yoon
Summary: This study proposes a method for constructing machine learning-based reduced order models (ROMs) that accurately simulate nonlinear contact problems while quantifying epistemic uncertainty. These ROMs significantly reduce computational costs compared to traditional full order models (FOMs). The technique combines adversarial training with an ensemble of Barlow twins reduced order models (BT-ROMs) to maximize the information content of the reduced manifolds. These ROMs, called UQ-BT-ROMs, demonstrate improved accuracy and computational efficiency compared to existing alternatives, with a relative error ranging from approximately 3% to 8% across all benchmarks. The UQ-BT-ROMs provide a cost-effective solution with significantly reduced computational times while maintaining a high level of accuracy.