The Düsseldorf Advanced Material Simulation Kit:
 DAMASK

Crystal plasticity modelling has gained considerable momentum in the past 20 years [1]. Developing this field from its original mean-field homogenization approach using viscoplastic constitutive hardening rules into an advanced multi-physics continuum field solution strategy requires a long-term initiative. The group “Theory and Simulation” of Franz Roters is working in this field since 2000. Code development during the last years was coordinated by the group “Integrated Computational Materials Engineering” headed by Martin Diehl. 

DAMASK has developed from a plain crystal plasticity user subroutine into a modular multi-physics crystal plasticity simulation package. The capabilities of DAMASK have been described in an overview paper [2] together with numerous usage examples including contributions from many international users of DAMASK. The paper was published in 2019 and was by now cited almost 150 times, indicating the popularity of DAMASK in the computational materials science community.

The work after the publication of the overview paper focused on the development of DAMASK 3. This new version features three major improvements:

  1. a modularization of the code
  2. new self-explanatory formats for input/output
  3. a python library for pre- and post-processing

The main aim of the code refactoring was a consistent and decoupled implementation of the different physics (crystal plasticity, damage, temperature, etc.). The new structure of the code strictly separates the different field problems and their solvers but at the same time implements the functionality for interaction among the fields necessary for the fully coupled treatment of multi-field problems. As a side effect, we were able to decrease the memory footprint of DAMASK and increase the performance of the code, which is now fully parallelized.

The use of standardised formats for in- and output necessitated from the fact, that with the growing capabilities and speed of DAMASK, postprocessing became a bottleneck in the overall simulation process. For this reason, HDF5 [3] was chosen as the base for a flexible output format. HDF5 is a widely used structured binary format that also allows storing any kind of meta data together with the actual simulation results. All input files have been changed to the YAML [4] format. YAML is a human readable format used for structuring configuration data. In addition, both changes strengthen the use of DAMASK simulation data following the FAIR (Findable, Accessible, Interoperable, and Re-usable) principle.

The tools for pre- and post-processing originally provided in the form of shell scripts are now cast into a Python library. This change allows to store complete simulation workflows in the form of Jupiter notebooks for easy documentation and reuse. Usage examples as well as the full documentation of the library can be found on the newly released DAMASK website.

DAMASK is developed as free and open source software and contributions in the form of additional features or bug fixes are welcome. The full sources are available at our GitLab repository.

Ultra-high-resolution high deformation simulations enabled by the recent DAMASK improvements give unprecedented insights into the deformation behaviour of crystalline materials.

Franz Roters, Philip Eisenlohr, Luc Hantcherli, Denny Dharmawan Tjahjanto, Thomas R. Bieler, and Dierk Raabe, "Overview of constitutive laws, kinematics, homogenization and multiscale methods in crystal plasticity finite-element modeling: Theory, experiments, applications," Acta Materialia 58 (4), 1152-1211 (2010).
Franz Roters, Martin Diehl, Pratheek Shanthraj, Philip Eisenlohr, Jan Christoph Reuber, Su Leen Wong, Tias Maiti, Alireza Ebrahimi, Thomas Hochrainer, Helge-Otto Fabritius, Svetoslav D. Nikolov, Martin Friák, Noriki Fujita, Nicolò Grilli, Koenraad G.F. Janssens, Nan Jia, Piet Kok, Duancheng Ma, Felix Meier, Ewald Werner, Markus Stricker, Daniel M. Weygand, and Dierk Raabe, "DAMASK – The Düsseldorf Advanced Material Simulation Kit for modeling multi-physics crystal plasticity, thermal, and damage phenomena from the single crystal up to the component scale," Computational Materials Science 158, 420-478 (2019).

3. The HDF Group. Hierarchical Data Format, version 5, 1997-2021. https://www.hdfgroup.org/solutions/hdf5/

4. YAML.org Website, https://yaml.org/

Other Interesting Articles

Go to Editor View