Phase field modeling of brittle fracture based on the cell-based smooth FEM by considering spectral decomposition

Fan Peng, Wei Huang, Y. E. Ma, Z. Q. Zhang, Yao Zhang

Research output: Contribution to journalArticlepeer-review

13 Scopus citations

Abstract

The spectral decomposition of the strain tensor is an essential technique to deal with the fracture problems via phase field method, and some incorrect results may be obtained without it. A novel phase field model for brittle fracture is developed based on cell-based smooth finite element (CS-FEM) and the spectral decomposition is taken into account. In order to describe the nonlinearity behaviors which contain the varied stress and elastic constitutive response caused by spectral decomposition. A second-order stress tensor and a fourth-order constitutive tensor based on decomposition of strain tensor are derived. A fundamental framework of CS-FEM is established to solve the phase field fracture problems, implemented by user-defined element (UEL) subroutine of ABAQUS software. The proposed model is validated by a typical Mode II crack, and the results show that the derived tensors are effective. Phase field parameter, CS-FEM parameter and mesh inhomogeneity are investigated to provide some useful suggestion for further development. Some classical numerical examples are solved by using the present model. The studies demonstrate that the proposed method can successfully overcome mesh distortion; the number of smooth cell does not show influences on the accuracy. Moreover, some results show that this method has the advantage over the standard FEM in convergence and computing efficiency.

Original languageEnglish
Article number2050016
JournalInternational Journal of Computational Methods
Volume18
Issue number2
DOIs
StatePublished - Mar 2021

Keywords

  • brittle fracture
  • cell-based smooth finite element
  • crack path
  • Phase field
  • spectral decomposition

Fingerprint

Dive into the research topics of 'Phase field modeling of brittle fracture based on the cell-based smooth FEM by considering spectral decomposition'. Together they form a unique fingerprint.

Cite this