Dear colleagues,
We are happy to share a recent study on Large-deformation reduced order homogenization of polycrystalline materials. In this work, we ectended our previous work on eigenstrain-based reduced order homogenization method to account for larger degoramtion, and demonstrated it capability of capture both stress-strain response and texture evolution with much higher efficientcy compared to reference crystal plasticity finite element modling. For more details, see full paper at: https://doi.org/10.1016/j.cma.2021.114119, and an authors copy of the paper is avaiable here: https://www.researchgate.net/publication/354532622_Large-deformation_red....
Abstrace: In this manuscript, we present a finite strain formulation of a reduced order computational homogenization model for crystal plasticity. The proposed formulation leverages and generalizes the principles of the Eigenstrain-based reduced order homogenization (EHM) approach. Asymptotic analysis with multiple scales is employed to describe the microscale problem in the deformed configuration. A two-term Taylor series approximation of the constitutive behavior along with a geometry-based basis reduction is employed to arrive at the reduced order model. An efficient implementation scheme is proposed to evaluate the multiscale system without the need to recompute the reduced basis as a function of evolving deformation. The ability of the proposed modeling approach in capturing homogenized and localized behavior as well as texture evolution is demonstrated in the context of single crystal and polycrystal microstructures.
Thank you for your interest.
Best,
Xiang
Two Postdoctoral Positions – Computational Multiscale Research
What Your Job Will Be Like
For a collaborative project funded by the German Research Foundation (DFG) we are seeking two Postdoctoral Appointees in the areas of multiscale materials modeling and high-performance scientific computing. You will address challenges in the development of coupling a molecular simulation code (LAMMPS) with a finite element simulation code with the aim of pushing multiscale simulation capabilities to length scales never reached before.
Qualifications We Require
• You have a PhD in Physics, Mathematics, Computer Science, or related Science or Engineering fields (Mechanical, Materials Science)
• Experience in research, development, and programming of atomistic simulation methods (MD) and/or finite element methods. Experience with the LAMMPS software package is a plus
• Excellent programming skills and proven experience in C++ as well as in a script language (python)
• Experience in high-performance scientific computer programming and parallelization
• Record of research as demonstrated by publications in peer-reviewed journals
Our Offer
• Working in creative, open-minded, ambitious teams with the goal of making a difference
• Exciting working environment on attractive research campus with excellent infrastructure, located in the cities of Darmstadt and Siegen.
• Full-time position. Limited for 2 years with possible longer-term prospects.
Applications
Get in touch with us. Applications (Letter of interest, CV, transcripts – one, single pdf-file please) to be sent via Email to:
albe@mm.tu-darmstadt.de, bernhard.eidel@uni-siegen.de, stukowski@mm.tu-darmstadt.de
Two fully-funded postdoctoral positions in computational mechanics of materials and structures are available in the groups of Milan Jirásek (https://mech.fsv.cvut.cz/~milan) and Jan Zeman (https://openmechanics.fsv.cvut.cz/people/jan-zeman). Full description of the openings is available at https://euraxess.ec.europa.eu/jobs/573988, the call closes on 5 December 2020.
Developing an accurate nonlinear reduced order model from simulation data has been an outstanding research topic for many years. For many physical systems, data collection is very expensive and the optimal data distribution is not known in advance. Thus, maximizing the information gain remains a grand challenge. In a recent paper, Bhattacharjee and Matous (2016) proposed a manifold-based nonlinear reduced order model for multiscale problems in mechanics of materials. Expanding this work here, we develop a novel sampling strategy based on the physics/pattern-guided data distribution. Our adaptive sampling strategy relies on enrichment of sub-manifolds based on the principal stretches and rotational sensitivity analysis. This novel sampling strategy substantially decreases the number of snapshots needed for accurate reduced order model construction (i.e., ~5x reduction of snapshots over Bhattacharjee and Matous (2016)). Moreover, we build the nonlinear manifold using the displacement rather than deformation gradient data. We provide rigorous verification and error assessment. Finally, we demonstrate both localization and homogenization of the multiscale solution on a large particulate composite unit cell.
https://www.sciencedirect.com/science/article/pii/S0045782519305420
Context
As part of a collaborative project between different Belgian industrial partners and Universities related to the study of composite laminate under impacts, the main objective of the doctoral position will be to develop a multi-scale numerical framework to study failure of the synthesized materials.
Post-Doc opportunity
The doctoral project will be supervised by Prof. L. Noels of ULg (http://www.ltas-cm3.ulg.ac.be/), in close collaboration with the partners of the project. The position is that of a research engineer starting in January 2020.
Profile
The candidate should have a PhD degree in mechanical engineering or applied mathematics with solid knowledge of continuous mechanics and numerical methods. Good programming skills are required.
Application
Interested candidates are encouraged to send a
• a CV with a list of up to 3 references;
• a short statement (maximum of one page) describing past experience and research interests;
• a transcript of the school grades.
The file must be sent to Prof. L. Noels (L.Noels@ulg.ac.be) by e-mail.
| Attachment | Size |
|---|---|
 TheseRuptureComposites.pdf | 420.96 KB |
Professor Milan Jirásek at Faculty of Civil Engineering, Czech Technical University in Prague, is searching for a motivated Ph.D. student, who will work on our new ambitious project on non-periodic mechanical metamaterials. Candidates should be interested in and have some previous experience with mathematical modeling and numerical simulation of the mechanical response of materials and structures.
Interested candidates are invited to proceed to https://euraxess.ec.europa.eu/jobs/370091 for additional details on the opening.
Since the beginning of the industrial age, material performance and design have been in the midst of innovation of many disruptive technologies. Today’s electronics, space, medical, transportation, and other industries are enriched by development, design and deployment of composite, heterogeneous and multifunctional materials. As a result, materials innovation is now considerably outpaced by other aspects from component design to product cycle. In this article, we review predictive nonlinear theories for multiscale modeling of heterogeneous materials. Deeper attention is given to multiscale modeling in space and to computational homogenization in addressing challenging materials science questions. Moreover, we discuss a state-of-the-art platform in predictive image-based, multiscale modeling with co-designed simulations and experiments that executes on the world’s largest supercomputers. Such a modeling framework consists of experimental tools, computational methods, and digital data strategies. Once fully completed, this collaborative and interdisciplinary framework can be the basis of Virtual Materials Testing standards and aids in the development of new material formulations. Moreover, it will decrease the time to market of innovative products.
A new perspective on model reduction for nonlinear multi-scale analysis of heterogeneous materials. In this work, we seek meaningful low-dimensional structures hidden in high-dimensional multi-scale data. The model relies on a global geometric framework for nonlinear dimensionality reduction (Isomap), and machine learning algorithms. The proposed model provides both homogenization and localization of the multiscale solution in the context of computational homogenization. The manifold-based reduced order model is verified using common principles from the machine-learning community. Both homogenization and localization of the multiscale solution are demonstrated on a large three-dimensional example. This reduced order model can also be used to accelerate fully coupled multiscale computational homogenization simulations.
S. Bhattacharjee and K. Matous, "A Nonlinear Manifold-based Reduced Order Model for Multiscale Analysis of Heterogeneous Hyperelastic Materials", Journal of Computational Physics, 31, 635--653 (2016).
In our recent Extreme Mechanics Letter, we present a simulation consisting of 53.8 Billion finite elements with 28.1 Billion nonlinear equations that is solved on 393,216 computing cores (786,432 threads). The excellent parallel performance of the computational homogenization solver is demonstrated by a strong scaling test from 4,096 to 262,144 cores. A fully coupled multi-scale damage simulation predicts a complex crack profile at the micro-scale, the macroscopic crack tunneling phenomenon as well as the nonuniform fracture toughness and strength of the interface. Such large and predictive simulations are an important step towards Virtual Materials Testing and can aid in development of new material formulations with extreme properties.
Matthew Mosby, Karel Matouš, Computational homogenization at extreme scales, Extreme Mechanics Letters, Volume 6, March 2016, Pages 68-74.
Dear Fellow Members,
I am currently pursuing my PhD research under the supervision of Prof. Ralf Müller at the Institute for Applied Mechanics, TU Kaiserslautern, Germany. I have submitted my dissertation on October 2014. And expecting to defend my PhD work on March, 2015.
At this moment I am looking for a research/postdoc position in the fields of computational mechanics, multi-scale modeling and simulation of composites, multi-scale modeling and simulation of coupled (electro-mechanical, thermo-mechanical, or magneto-mechanical) problem, or any other relevant field.
During the PhD research, an in-depth investigation is carried out of FE^2-based multi-scale simulation of piezoelectric materials. To capture the effect of the micro inhomogeneities on macro level, one needs to perform homogenization of piezoelectric material. The focus of this work is to study the effects of micro inhomogeneities of piezoelectric materials on macroscopic configurational forces. Special efforts has been given on the configurational forces at certain defect situations. The configurational force at a sharp crack tip can be considered as one of these defect situations. This material is electro-mechanically coupled. A new set of boundary conditions are developed for the micro boundary value problems (BVP) of this coupled problem. By using these new boundary conditions, consistent numerical results of the configurational forces have been generated as part of PhD thesis. The detailed numerical analysis has been implemented in Matlab. The developed numerical technique has been applied to homogenize the piezoelectric material using non-evolving micro-structures. Afterwards, this method is modified and upgraded for the homogenization of piezoelectric electric material by using evolving micro-structures.
Before pursuing my PhD research, I completed my master study in “computational sciences in engineering” at TU Braunschweig, Germany. During my PhD and master studies, I acquired extensive knowledge of numerical methods for PDEs and other scientific computing methods. Additionally, I have a long term experiences in numerical programming using C/C++, Fortran, Matlab.
I will be appreciating any kind of suggestion or help to obtain a research or postdoc position. Interested people can mail me to this ID: m.khalaquzzaman10@gmail.com
Sincerely,
Md Khalaquzzaman
Note: I am uploading my CV as separate file which is mostly describing my credentials and main idea of my PhD thesis.
| Attachment | Size |
|---|---|
| curriculum vitae | 92.24 KB |
See attachement.
| Attachment | Size |
|---|---|
| postdoc_ad_mod.pdf | 211.5 KB |
We've recently published an open access journal paper that looks at the mechanics of sutures used to repair severed tendons. A homogenization strategy is used to derive effective elastic properties for tendon fibrils and intracellular matrix. We have found that regions of high stress correlate with the regions of cell death (necrosis) that are sometimes observed in patients.
If this is of interest, please feel free to view the paper here.
Hello,
I read that "In general, even if the materials on the micro-level are isotropic, the effective
material can show anisotropic behavior. A general anisotropic linear elastic material
may have twenty one independent material parameters.''
If I understand my results correctly then simple structures like ''ball in the unit cell'' result in orthotropic material.
I am a bit puzzled - what would be the simplest structure that would result in anisotropic material behaviour?
Our Mechanics of Materials Group at Eindhoven University of Technology, the Netherlands has two openings for talented PhD students in the field of multiscale mechanics of materials. They are part of a European Union funded project on multiscale methods for advanced materials. One opening is on the development of a fundamentally new multiscale approach towards material modelling and the other aims to integrate this approach with experimental methods.
More information can be found via http://jobs.tue.nl/en/job/phd-student-fundamentals-of-a-fluctuationenric... and/or http://jobs.tue.nl/en/job/phd-student-coupled-microfluctuationbased-expe..., through which candidates may also directly apply. Remaining questions may be directed to me (J.P.M.Hoefnagels@tue.nl).
You are welcome to submit abstracts to the MS 4.5 "Multiscale Computational homogenization for bridging scales in the mechanics and physics of complex materials", organized by P. Wriggers, K. Terada, V. Kouznetsova, M. Cho and myself at the 12th US National Congress on Computation Mechanics (USNCCM), July 22-25, 2013, in Raleigh, USA.
The deadline fo abstract submission is February 15, 2013.
The topics of the MS can be found here:
Looking forward to meet you in Raleigh.
Several methods have been proposed for numerical homogeniation of linear viscoelastic materials, mainly based on Laplace transform or on multilevel (FE^2) approaches. In this paper, we introduce a much simpler technique based on a discrete representation of the effective relaxation tensor related to the homogeneous medium, which can then be used to evaluate the constitutive law in the form of a convolution product. In practice, calculations on the RVE reduce to 3 transient simulations in 2D and 6 in 3D. More details in
A.B. Tran, J. Yvonnet, Q.-C. He, C. Toulemonde, J. Sanahuja, A simple computational homogenization method for structures made of linear heterogeneous viscoelastic material. Comput. Methods Appl. Mech. Eng. 200 (45-46):2956-2970 (2011).
| Attachment | Size |
|---|---|
 [32] Draft.pdf | 1.14 MB |