TY - JOUR

T1 - pfm-cracks

T2 - A parallel-adaptive framework for phase-field fracture propagation[Formula presented]

AU - Heister, Timo

AU - Wick, Thomas

N1 - Funding Information: We thank Katrin Mang, Daniel Jodlbauer, Sanghyun Lee, Nima Noii, and Meng Fan for contributions and discussions, and also financial support through the DFG (German Research Foundation) priority program SPP 1748 as well as the FWF (Austrian Science Fund) project P29181 . Moreover, past support through a Feodor-Lynen fellowship (years 2013–2014 at UT Austin) via the Humboldt foundation and the Center for Subsurface Modeling headed by Professor Mary Wheeler initiating this code is gratefully acknowledged. Moreover, designing and analyzing the governing mathematical models would not have been possible with our friend and colleague Professor Andro Mikelić. Timo Heister was partially supported by the National Science Foundation (NSF) Award DMS-2028346 , OAC-2015848 , EAR-1925575 , by the Computational Infrastructure in Geodynamics initiative (CIG) , through the NSF under Award EAR-0949446 and EAR-1550901 and The University of California — Davis , and by Technical Data Analysis, Inc. through US Navy STTR Contract N68335-18-C-0011

PY - 2020/11

Y1 - 2020/11

N2 - This paper describes the main features of our parallel-adaptive open-source framework for solving phase-field fracture problems called pfm-cracks. Our program allows for dimension-independent programming in two- and three-dimensional settings. A quasi-monolithic formulation for the coupled two-component system of displacements and a phase-field indicator variable is used. The nonlinear problem is solved with a robust, efficient semi-smooth Newton algorithm. A highlight is adaptive predictor–corrector mesh refinement. The code is fully parallelized and scales to 1000 and more MPI ranks. Illustrative tests demonstrate the current capabilities, from which some are parts of benchmark collections.

AB - This paper describes the main features of our parallel-adaptive open-source framework for solving phase-field fracture problems called pfm-cracks. Our program allows for dimension-independent programming in two- and three-dimensional settings. A quasi-monolithic formulation for the coupled two-component system of displacements and a phase-field indicator variable is used. The nonlinear problem is solved with a robust, efficient semi-smooth Newton algorithm. A highlight is adaptive predictor–corrector mesh refinement. The code is fully parallelized and scales to 1000 and more MPI ranks. Illustrative tests demonstrate the current capabilities, from which some are parts of benchmark collections.

KW - Adaptivity

KW - Open-source software

KW - Parallel computing

KW - Phase-field fracture

KW - Primal–dual active set

KW - Semi-smooth Newton

UR - http://www.scopus.com/inward/record.url?scp=85115828279&partnerID=8YFLogxK

U2 - 10.1016/j.simpa.2020.100045

DO - 10.1016/j.simpa.2020.100045

M3 - Article

AN - SCOPUS:85115828279

VL - 6

JO - Software Impacts

JF - Software Impacts

M1 - 100045

ER -