Nonconforming Discretization Research Articles

Numerical analysis of a mixed-dimensional poromechanical model with frictionless contact at matrix–fracture interfaces

We present a complete numerical analysis for a general discretization of a coupled flow–mechanics model in fractured porous media, considering single-phase flows and including frictionless contact at matrix–fracture interfaces, as well as nonlinear poromechanical coupling. Fractures are described as planar surfaces, yielding the so-called mixed- or hybrid-dimensional models. Small displacements and a linear elastic behavior are considered for the matrix. The model accounts for discontinuous fluid pressures at matrix–fracture interfaces in order to cover a wide range of normal fracture conductivities. The numerical analysis is carried out in the Gradient Discretization framework (see J. Droniou, R. Eymard, T. Gallouët, C. Guichard, and R. Herbin [The gradient discretisation method, Springer, Cham, 2018]), encompassing a large family of conforming and nonconforming discretizations. The convergence result also yields, as a by-product, the existence of a weak solution to the continuous model. A numerical experiment in 2D is presented to support the obtained result, employing a Hybrid Finite Volume scheme for the flow and second-order finite elements ( P 2 \mathbb {P}_2 ) for the mechanical displacement coupled with face-wise constant ( P 0 \mathbb P_0 ) Lagrange multipliers on fractures, representing normal stresses, to discretize the contact conditions.

Assessment of New Adaptive Finite Elements Based on Carrera Unified Formulation for Meshes with Arbitrary Polygons

The meshing technique represents the capability to discretize the domain of interest, to fit the real physical continuum in the best possible way. The most used approach is the finite-element method (FEM), a numerical method to solve partial differential equations. To overcome the classical issues presented by FEM, other models are investigated. The goal is to allow the problem domain to be discretized by elements represented by arbitrary polygons, which can be concave and convex. Moreover, different polynomial consistency is sought within these methods with the possibility to handle non-conforming discretizations, mainly for local refinement and so on. This work aims to present the new adaptive elements, which are finite elements based on Carrera unified formulation, to demonstrate that all the previous capabilities can be done with these new elements, with easy implementation of the relative model. First, a classical patch test is done to investigate the mesh distortion sensitivity. Then, different study cases are presented with more complex meshes combining very distorted concave and convex elements.

Adaptive inexact smoothing Newton method for a nonconforming discretization of a variational inequality

Adaptive inexact smoothing Newton method for a nonconforming discretization of a variational inequality

A deep learning-enhanced framework for multiphysics joint inversion

Joint inversion has drawn considerable attention due to the availability of multiple geophysical data sets, ever-increasing computational resources, the development of advanced algorithms, and its ability to reduce inversion uncertainty. A key issue of joint inversion is to develop effective strategies to link different geophysical data in a unified mathematical framework, in which the information obtained from different models can complement each other. We have developed a deep learning-enhanced joint inversion framework to simultaneously reconstruct different physical models by fusing different types of geophysical data. Traditionally, structure similarity constraints are pursued by joint inversion algorithms using manually crafted formulations (e.g., cross gradient). The constraint is constructed by a deep neural network (DNN) during the learning process. The framework is designed to combine the DNN and a traditional independent inversion workflow and improve the joint inversion result iteratively. The network can be easily extended to incorporate multiphysics without structural changes. Numerical experiments on the joint inversion of 2D DC resistivity data and seismic traveltime are used to validate our method. In addition, this learning-based framework demonstrates excellent generalization abilities when tested on data sets using different geologic structures. It also can handle different sensing configurations and nonconforming discretization.

A novel SAM/X-FEM coupling approach for the simulation of 3D fatigue crack growth under rolling contact loading

The present work proposes a novel efficient numerical approach for the simulation of three-dimensional propagation of non-planar frictional crack under Rolling Contact Fatigue (RCF). The developed model relies on a global–local strategy involving the Semi-Analytical Method (SAM), dedicated to the resolution of 3D contact problems, and the eXtended-Finite Element Method (X-FEM), dedicated to the solving of 3D crack problems. Similar to a submodeling technique, it consists in performing a local X-FEM analysis of the 3D RCF crack problem by means of boundary conditions extracted from a prior global SAM analysis which solves the contact problem between two semi-infinite bodies without considering the crack. A powerful procedure for the transfer of relevant mechanical quantities between the SAM and X-FEM models is developed. It allows the use of non-matching and non-conforming discretization scheme for the global SAM model and the local X-FEM model, particularly worthwhile for the pre-processing stage. An initial arbitrary stress profile can be considered to investigate the influence of residual stresses on crack behavior. The non-intrusive SAM/X-FEM coupling is integrated into a fully automatic crack propagation algorithm. This provides a powerful and user-friendly tool suitable for industrial applications, which easily handles 3D long non-planar frictional crack growth in the region of interest. After introducing some details concerning the implementation of the model, a validation of the innovative SAM/X-FEM coupling is performed by using results from the literature. Furthermore, a practical example demonstrates the great potential of this novel numerical technique to simulate in a fast, robust and accurate way the 3D complex behavior of fatigue crack under moving contact. • Simulation of the 3D frictional crack propagation under rolling contact conditions. • Non-intrusive global–local approach based on the coupling of SAM and X-FEM. • Field transfer procedure allowing independent meshes for the SAM and X-FEM models. • Fast, efficient and robust resolution coupled with a pre-processing time saving. • Influence of residual stresses on crack behavior highlighted.

An interface-enriched generalized finite element formulation for locking-free coupling of non-conforming discretizations and contact

We propose an enriched finite element formulation to address the computational modeling of contact problems and the coupling of non-conforming discretizations in the small deformation setting. The displacement field is augmented by enriched terms that are associated with generalized degrees of freedom collocated along non-conforming interfaces or contact surfaces. The enrichment strategy effectively produces an enriched node-to-node discretization that can be used with any constraint enforcement criterion; this is demonstrated with both multi-point constraints and Lagrange multipliers, the latter in a generalized Newton implementation where both primal and Lagrange multiplier fields are updated simultaneously. We show that the node-to-node enrichment ensures continuity of the displacement field—without locking—in mesh coupling problems, and that tractions are transferred accurately at contact interfaces without the need for stabilization. We also show the formulation is stable with respect to the condition number of the stiffness matrix by using a simple Jacobi-like diagonal preconditioner.

Non-conformal field-only boundary integral method for modeling EM scattering problems

Recently, a novel boundary integral equation method, namely field-only boundary integral equation (FOBIE) method, has been proposed in analyzing the electromagnetic (EM) scattering from perfectly electrically conducting (PEC) and homogeneous dielectric objects in the frequency domain. The main feature of the FOBIE method is that the Cartesian components of its electric fields can be obtained directly by solving a set of singularity free boundary integral equations. This paper will give a brief review of the FOBIE method from the EM point of view first, then extend it to handle non-conformal discretization since the continuity conditions of electric fields have already been imposed by the boundary conditions and no additional conditions are needed. A block matrix solution scheme is proposed to accelerate the solution process of the discretized matrix equation, which makes the numerical framework much more efficient. The accuracy and efficiency of the proposed method are demonstrated by analyzing the EM scattering from several canonical objects with non-conformal discretization. In addition, the time domain transient EM characteristic can be obtained by using the inverse Fourier transform to reflect the physical process of the interaction between the EM fields and object conveniently in the time domain.

Analysis of finite element methods for surface vector-Laplace eigenproblems

In this paper we study finite element discretizations of a surface vector-Laplace eigenproblem. We consider two known classes of finite element methods, namely one based on a vector analogon of the Dziuk-Elliott surface finite element method and one based on the so-called trace finite element technique. A key ingredient in both classes of methods is a penalization method that is used to enforce tangentiality of the vector field in a weak sense. This penalization and the perturbations that arise from numerical approximation of the surface lead to essential nonconformities in the discretization of the variational formulation of the vector-Laplace eigenproblem. We present a general abstract framework applicable to such nonconforming discretizations of eigenproblems. Error bounds both for eigenvalue and eigenvector approximations are derived that depend on certain consistency and approximability parameters. Sharpness of these bounds is discussed. Results of a numerical experiment illustrate certain convergence properties of such finite element discretizations of the surface vector-Laplace eigenproblem.

Conservation of Forces and Total Work at the Interface Using the Internodes Method

The Internodes method is a general purpose method to deal with non-conforming discretizations of partial differential equations on 2D and 3D regions partitioned into disjoint subdomains. In this paper we are interested in measuring how much the Internodes method is conservative across the interface. If hp-fem discretizations are employed, we prove that both the total force and total work generated by the numerical solution at the interface of the decomposition vanish in an optimal way when the mesh size tends to zero, i.e., like mathcal {O}(h^{p}), where p is the local polynomial degree and h the mesh-size. This is the same as the error decay in the H1-broken norm. We observe that the conservation properties of a method are intrinsic to the method itself because they depend on the way the interface conditions are enforced rather then on the problem we are called to approximate. For this reason, in this paper, we focus on second-order elliptic PDEs, although we use the terminology (of forces and works) proper of linear elasticity. Two and three dimensional numerical experiments corroborate the theoretical findings, also by comparing Internodes with Mortar and WACA methods.

Energy-stable discretization of two-phase flows in deformable porous media with frictional contact at matrix–fracture interfaces

We address the discretization of two-phase Darcy flows in a fractured and deformable porous medium, including frictional contact between the matrix–fracture interfaces. Fractures are described as a network of planar surfaces leading to the so-called mixed- or hybrid-dimensional models. Small displacements and a linear elastic behavior are considered for the matrix. Phase pressures are supposed to be discontinuous at matrix–fracture interfaces, as they provide a better accuracy than continuous pressure models even for high fracture permeabilities.The general gradient discretization framework [31] is employed for the numerical analysis, allowing for a generic stability analysis and including several conforming and nonconforming discretizations. We establish energy estimates for the discretization, and prove existence of a solution. To simulate the coupled model, we employ a Two-Point Flux Approximation (TPFA) finite volume scheme for the flow and second-order (P2) finite elements for the mechanical displacement coupled with face-wise constant (P0) Lagrange multipliers on fractures, representing normal and tangential stresses, to discretize the frictional contact conditions. This choice allows to circumvent possible singularities at tips, corners, and intersections between fractures, and provides a local expression of the contact conditions. We present numerical simulations of two benchmark examples and one realistic test case based on a drying model in a radioactive waste geological storage structure.

A nonsymmetric approach and a quasi-optimal and robust discretization for the Biot’s model

We consider the system of partial differential equations stemming from the time discretization of the two-field formulation of the Biot’s model with the backward Euler scheme. A typical difficulty encountered in the space discretization of this problem is the robustness with respect to various material parameters. We deal with this issue by observing that the problem is uniformly stable, irrespective of all parameters, in a suitable nonsymmetric variational setting. Guided by this result, we design a novel nonconforming discretization, which employs Crouzeix-Raviart and discontinuous elements. We prove that the proposed discretization is quasi-optimal and robust in a parameter-dependent norm and discuss the consequences of this result.

A Flexible FEM-BEM-DDM for EM Scattering by Multiscale Anisotropic Objects

In this article, a nonconformal domain decomposition method (DDM) based on the hybrid finite element method (FEM) and boundary element method (BEM) is proposed to solve arbitrary 3-D anisotropic (including bianisotropic and single anisotropic) multiscale composite objects. The object is divided into interior and exterior regions. The interior region is solved by the bianisotropic finite element domain decomposition method (FEM-DDM), and the exterior region is solved by the discontinuous Galerkin (DG) method based on the combined field integral equation (CFIE). In order to decouple the electric and magnetic fields of the bianisotropic object, a new finite element–boundary element–domain decomposition method (FEM-BEM-DDM) formulation is derived. To ensure the continuity of electric current and electric field between anisotropic FEM subdomains, a new auxiliary current is introduced into the interface of different subdomains of FEM and the interface of FEM subdomains with BEM subdomains. The former uses second-order transmission conditions (SOTCs) to ensure continuity, whereas the latter uses Robin transmission conditions (RTCs) to ensure continuity. The BEM dense matrix–vector multiplication is accelerated by a multilevel fast multipole algorithm (MLFMA). The proposed FEM-BEM-DDM allows for nonconformal discretization between any touching subdomains and provides an effective preconditioner. In numerical examples, the accuracy and effectiveness of the method are verified.

Discretization error estimates for discontinuous Galerkin isogeometric analysis

Isogeometric analysis is a spline-based discretization method to partial differential equations which show the approximation power of a high-order method. The number of degrees of freedom, however, is as small as the number of degrees of freedom of a low-order method. This does not come for free as the original formulation of isogeometric analysis requires a global geometry function. Since this is too restrictive for many kinds of applications, the domain is usually decomposed into patches, where each patch is parameterized with its own geometry function. In simpler cases, the patches can be combined in a conforming way. However, for non-matching discretizations or for varying coefficients, a non-conforming discretization is desired. An symmetric interior penalty discontinuous Galerkin method for isogeometric analysis has been previously introduced. In the present paper, we give error estimates that are explicit in the spline degree. This opens the door towards the construction and the analysis of fast linear solvers, particularly multigrid solvers for non-conforming multipatch isogeometric analysis.

Characteristic basis function method based on mixed discretization for analyzing scattering from electrically large inhomogeneous objects

The characteristic basis function method (CBFM) based on the mixed discretization (MD) is firstly proposed to analyze the electromagnetic scattering from electrically large inhomogeneous objects. The discontinuous Galerkin volume integral equation (DGVIE) method which employs nonconformal discretization permits to independently discretize the various component regions of the inhomogeneous object depending on its local shape or dielectric permittivity. Moreover, by the means of the CBFM, the low-level basis functions employed in the DGVIE method are converted into high-level basis functions, dramatically reducing the number of basis functions for modeling. In addition, the conformal discretization which is superior in dealing with homogenous regions is mixed with the nonconformal discretization to further cut down the number of the basis functions. Numerical results show that the CBFM based on mixed discretization (MD-CBFM) generates much less unknowns in contrast to several conventional methods. Several composite inhomogeneous examples are given to demonstrate the accuracy and efficiency of the proposed method.

Blended isogeometric Kirchhoff–Love and continuum shells

The computational modeling of thin-walled structures based on isogeometric analysis (IGA), non-uniform rational B-splines (NURBS), and Kirchhoff–Love (KL) shell formulations has attracted significant research attention in recent years. While these methods offer numerous benefits over the traditional finite element approach, including exact representation of the geometry, naturally satisfied high-order continuity within each NURBS patch, and computationally efficient rotation-free formulations, they also present a number of challenges in modeling real-world engineering structures of considerable complexity. Specifically, these NURBS-based engineering models are usually comprised of numerous patches, with discontinuous derivatives, non-conforming discretizations, and non-watertight connections at their interfaces. Moreover, the analysis of such structures often requires the full stress and strain tensors (i.e., including the transverse normal and shear components) for subsequent failure analysis and remaining life prediction. Despite the efficiency provided by the KL shell, the formulation cannot accurately predict the response in the transverse directions due to its kinematic assumptions. In this work, a penalty-based formulation for the blended coupling of KL and continuum shells is presented. The proposed approach embraces both the computational efficiency of KL shells and the availability of the full-scale stress/strain tensors of continuum shells where needed by modeling critical structural components using continuum shells and other components using KL shells. The proposed method enforces the displacement and rotational continuities in a variational manner and is applicable to non-conforming and non-smooth interfaces. The efficacy of the developed method is demonstrated through a number of benchmark studies with a variety of analysis configurations, including linear and nonlinear analyses, matching and non-matching discretizations, and isotropic and composite materials. Finally, an aircraft horizontal stabilizer is considered to demonstrate the applicability of the proposed blended shells to real-world engineering structures of significant complexity.

A discontinuous Nitsche‐based finite element formulation for the imposition of the Navier‐slip condition over embedded volumeless geometries

AbstractThis work describes a novel formulation for the simulation of incompressible Navier–Stokes problems involving nonconforming discretizations of membrane‐like bodies. The new proposal relies on the use of a modified finite element space within the elements intersected by the embedded geometry, which is represented by a discontinuous (or element‐by‐element) level set function. This is combined with a Nitsche‐based imposition of the general Navier‐slip boundary condition, to be intended as a wall law model. Thanks to the use of an alternative finite element space, the formulation is capable of reproducing exactly discontinuities across the embedded interface, while preserving the structure of the graph of the discrete matrix. The performance, accuracy and convergence of the new proposal is compared with analytical solutions as well as with a body fitted reference technique. Moreover, the proposal is tested against another similar embedded approach. Finally, a realistic application showcasing the possibilities of the method is also presented.

A sharp interface Lagrangian-Eulerian method for rigid-body fluid-structure interaction

This paper introduces a sharp interface method to simulate fluid-structure interaction (FSI) involving rigid bodies immersed in viscous incompressible fluids. The capabilities of this methodology are benchmarked using a range of test cases and demonstrated using large-scale models of biomedical FSI. The numerical approach developed herein, which we refer to as an immersed Lagrangian-Eulerian (ILE) method, integrates aspects of partitioned and immersed FSI formulations by solving separate momentum equations for the fluid and solid subdomains, as in a partitioned formulation, while also using non-conforming discretizations of the dynamic fluid and structure regions, as in an immersed formulation. A simple Dirichlet-Neumann coupling scheme is used, in which the motion of the immersed solid is driven by fluid traction forces evaluated along the fluid-structure interface, and the motion of the fluid along that interface is constrained to match the solid velocity and thereby satisfy the no-slip condition. To develop a practical numerical method, we adopt a penalty approach that approximately imposes the no-slip condition along the fluid-structure interface. In the coupling strategy, a separate discretization of the fluid-structure interface is tethered to the volumetric solid mesh via stiff spring-like penalty forces. Our fluid-structure coupling scheme relies on an immersed interface method (IIM) for discrete geometries, which enables the accurate determination of both velocities and stresses along complex internal interfaces. Numerical methods for FSI can suffer from instabilities related to the added mass effect, but computational tests indicate that the methodology introduced here remains stable for selected test cases across a broad range of solid-fluid mass density ratios, including extremely small, nearly equal, equal, and large density ratios. Biomedical FSI demonstration cases include results obtained using this method to simulate the dynamics of a bileaflet mechanical heart valve in a pulse duplicator, and to model transport of blood clots in a patient-averaged anatomical model of the inferior vena cava.

Isogeometric Bézier dual mortaring: The Kirchhoff–Love shell problem

In this paper we develop an isogeometric Bézier dual mortar method for coupling multi-patch Kirchhoff–Love shell structures. The proposed approach weakly enforces the continuity of the solution at patch interfaces through a dual mortar method and can be applied to both conforming and non-conforming discretizations. As the employed dual basis functions have local supports and satisfy the biorthogonality property, the resulting stiffness matrix is sparse. In addition, the coupling accuracy is optimal because the dual basis possesses the polynomial reproduction property. We also formulate the continuity constraints through the Rodrigues’ rotation operator which gives a unified framework for coupling patches that are intersected with G1 continuity as well as patches that meet at a kink. Several linear and nonlinear examples demonstrated the performance and robustness of the proposed coupling techniques.

Nonconforming discretizations of convex minimization problems and precise relations to mixed methods

This article discusses nonconforming finite element methods for convex minimization problems and systematically derives dual mixed formulations. Duality relations lead to simple error estimates that avoid an explicit treatment of nonconformity errors. A reconstruction formula provides the discrete solution of the dual problem via a simple postprocessing procedure which implies a strong duality relation and is of interest in a posteriori error estimation. The framework applies to differentiable and nonsmooth problems, examples include p-Laplace, total-variation regularized, and obstacle problems. Numerical experiments illustrate advantages of nonconforming over standard conforming methods.

Multiscale design of nonlinear materials using a Eulerian shape optimization scheme

AbstractMotivated by recent advances in manufacturing, the design of materials is the focal point of interest in the material research community. One of the critical challenges in this field is finding optimal material microstructure for a desired macroscopic response. This work presents a computational method for the mesoscale‐level design of particulate composites for an optimal macroscale‐level response. The method relies on a custom shape optimization scheme to find the extrema of a nonlinear cost function subject to a set of constraints. Three key “modules” constitute the method: multiscale modeling, sensitivity analysis, and optimization. Multiscale modeling relies on a classical homogenization method and a nonlinear NURBS‐based generalized finite element scheme to efficiently and accurately compute the structural response of particulate composites using a nonconformal discretization. A three‐parameter isotropic damage law is used to model microstructure‐level failure. An analytical sensitivity method is developed to compute the derivatives of the cost/constraint functions with respect to the design variables that control the microstructure's geometry. The derivation uncovers subtle but essential new terms contributing to the sensitivity of finite element shape functions and their spatial derivatives. Several structural problems are solved to demonstrate the applicability, performance, and accuracy of the method for the design of particulate composites with a desired macroscopic nonlinear stress‐strain response.