Comparative Study Of Phase-field Regularized Cohesive Zone Model And Extended Finite Element Method In Concrete

Concrete is a typical quasi-brittle material that tends to crack.These cracks often degenerate the mechanical properties of concrete and may even make the entire structure lose its bearing capacity.Therefore,it is necessary to be able to describe the degeneration rules of the mechanical properties of concrete caused by the occurrence and expansion of cracks accurately.Numerical simulation methods for describing damage and failure in concrete can complement the theoretical model and the relative experiments,to provide theoretical basis for the analysis and design of concrete structures.Moreover,numerical simulation methods have the properties of repeatability,low cost and are easy to investigate.They play a more and more important role in theoretical research and applications in engineering.The phase-field regularized cohesive zone model(PF-CZM)for concrete and the extended finite element method(XFEM)are two of the most representative methods among various numerical simulation methods.The XFEM which is broadly adopted by many scholars enriches new degrees of freedom on the element nodes,reflecting the discontinuity properties of cracks without mesh reconstruction.The PF-CZM describes material properties by introducing the overall continuous phase-field variables,and replaces the sharp crack with a diffused local damage band,with no need of the cumbersome work of tracking the crack path.This model is more suitable for dealing with quasi-brittle fracture than the classical phase-field model and has a very good prospect in application.In order to study the advantages and disadvantages of the PF-CZM and the XFEM in the modeling of damage and failure in concrete,these two methods are adopted to simulate the type I or mixed type verification experiments in concrete in this work.Path-following techniques,which can deal with material softening induced snap-backs,are implemented in both of the two methods.Different mesh size and length scale parameter is used in numerical simulation,aiming to study the effects of the numerical parameter on numerical simulation results.It is found that both methods are independent of the discretized mesh and have no stress-locking results.On one hand,the XFEM is advantageous due to its high coarse mesh resolution,but it heavily depends on the adopted crack propagation criterion so that it is not always effective.Without extra criteria and multiple enrichments,crack branching cannot be captured by the XFEM.On the other hand,the PF-CZM can intrinsically deal with crack nucleation,propagation,merging and branching in an elegant manner via evolution of the damage field,with no need of any extra strategy.The internal length scale has no effect to the simulation results.As there is no need to explicitly track the crack path at all,the PF-CZM is very appealing in the modeling of arbitrary crack propagation.