Robust Error Estimates for Adaptive Phase Field Simulations

Phase field equations are commonly used as a regularized model, where bulk phases are separated by interface regions that have a thickness of the order γ. Their numerical analysis is well established for a fixed parameter size γ, but conventional error estimates depend exponentially on γ⁻¹ and thus become useless in the relevant case if γ→0. Technical applications include e.g. the simulation of Sn–Cu alloys for the production of lead free solder or Ni–Al alloys used for rotor blade surfaces. (© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Some numerical techniques for solving elliptic interface problems

Many physical phenomena can be modeled by partial differential equations with singularities and interfaces. The standard finite difference and finite element methods may not be successful in giving satisfactory numerical results for such problems. Hence, many new methods have been developed. Some of them are developed with the modifications in the standard methods, so that they can deal with the discontinuities and the singularities. In this article, a survey has been done on some recent efficient techniques to solve elliptic interface problems. © 2010 Wiley Periodicals, Inc. Numer Methods Partial Differential Eq, 28: 94‐114, 2012     

Difference approximations for wave equations via finite elements. I. Construction and performance

Many practical applications of wave equations involve media in which there are interfaces, or discontinuities in material properties. The accurate numerical representation of these interfaces is important in mathematical models. One can develop generalizations of standard finite-difference methods that accommodate sharp interfaces by modifying a straightforward finite-element approach. In two space dimensions, these methods yield explicit, 5-point or 9-point difference schemes that accurately capture reflection, transmission, and refraction at interfaces. The approach also extends readily to the simulation of waves in elastic media. A companion article presents an error analysis for the approach. © 1995 John Wiley & Sons, Inc.     

曲率流的参数化有限元逼近

许多物理现象可以在数学上描述为受曲率驱动的自由界面运动,例如薄膜和泡沫的演变、晶体生长,等等.这些薄膜和界面的运动常依赖于其表面曲率,从而可以用相应的曲率流来描述,其相关自由界面问题的数值计算和误差分析一直是计算数学领域中的难点.参数化有限元法是曲率流的一类有效计算方法,已经能够成功模拟一些曲面在几类基本的曲率流下的演化过程.本文重点讨论曲率流的参数化有限元逼近,它的产生、发展和当前的一些挑战.     

Isogeometric Analysis of a Phase Field Model for Darcy Flows with Discontinuous Data

The authors consider a phase field model for Darcy flows with discontinuous data in porous media; specifically,they adopt the Hele-Shaw-Cahn-Hillard equations of[Lee,Lowengrub,Goodman,Physics of Fluids,2002] to model flows in the Hele-Shaw cell through a phase field formulation which incorporates discontinuities of physical data,namely density and viscosity,across interfaces. For the spatial approximation of the problem,the authors use NURBS—based isogeometric analysis in the framework of the Galerkin method,a computational framework which is particularly advantageous for the solution of high order partial differential equations and phase field problems which exhibit sharp but smooth interfaces. In this paper,the authors verify through numerical tests the sharp interface limit of the phase field model which in fact leads to an internal discontinuity interface problem; finally,they show the efficiency of isogeometric analysis for the numerical approximation of the model by solving a benchmark problem,the so-called"rising bubble" problem.     

Difference approximations for wave equations via finite elements II. Error analysis

Difference-like schemes for the wave equation arise naturally from a Galerkin finite-element formulation, if we adopt certain quadrature rules in evaluating the mass and stiffness matrices. One can extend these schemes to problems involving sharp interfaces by applying the quadrature on a refinement of the finite-element grid that includes the interfaces. We develop error estimates for this modified scheme that corroborate numerical results for acoustic and elastic wave equations, presented in a companion article. © 1995 John Wiley & Sons, Inc.     

Error control for the approximation of Allen–Cahn and Cahn–Hilliard equations with a logarithmic potential

A fully computable upper bound for the finite element approximation error of Allen–Cahn and Cahn–Hilliard equations with logarithmic potentials is derived. Numerical experiments show that for the sharp interface limit this bound is robust past topological changes. Modifications of the abstract results to derive quasi-optimal error estimates in different norms for lowest order finite element methods are discussed and lead to weaker conditions on the residuals under which the conditional error estimates hold.     

Gradient based enhanced finite element formulation for diffuse phase interfaces

Interfaces between adjacent phases, so-called domain walls, appear as non-linear gradients of order parameters in diffuse phase field models. Usually, the interface width is much smaller than the dimension of the simulated region. Since the position of domain walls is not known a priori the maximum size of finite elements needs to be adapted to the length scale of interfaces within the entire region. We suggested a selective finite element method to improve the numerical solution of diffuse phase field models [1, 2]. It enhances the finite element interpolation space using supplementary local degrees of freedom. However, corresponding additional nodes are strictly located in the interior of elements, thus, C⁰-continuity at element border is guaranteed. Since C⁰-continuity limits the performance of this method we propose in this paper a relaxation of C⁰-requirements perpendicular to the gradient of the order parameter. Therefore, the direction of interfaces is analyzed as additional information for further adaptive improvement of the interpolation space. A dual phase field model is used to validate the proposed method. The analytical solution of a stationary domain wall allows error analysis of regular and distorted finite element meshes. (© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)     

A Regularized Sharp Interface Model for Phase Transformation Accounting for Prescribed Sharp Interface Kinetics

The macroscopic mechanical behavior of multi-phasic materials depends on the formation and evolution of their microstructure by means of phase transformation. In case of martensitic transformations, the resulting phase boundaries are sharp interfaces. We carry out a geometrically motivated discussion of the regularization of such sharp interfaces by use of an order parameter/phase-field and exploit the results for a regularized sharp interface model for two-phase elastic materials with evolving phase boundaries. To account for the dissipative effects during phase transition, we model the material as a generalized standard medium with energy storage and a dissipation function that determines the evolution of the regularized interface. Making use of the level-set equation, we are thereby able to directly translate prescribed sharp interface kinetic relations to the constitutive model in the regularized setting. We develop a suitable incremental variational three-field framework for the dissipative phase transformation problem. Finally, the modeling capability and the associated numerical solution techniques are demonstrated by means of a representative numerical example. (© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Multi-Dimensional Quadrature of Singular and Discontinuous Functions

In many simulations of physical phenomena, discontinuous material coefficients and singular forces pose severe challenges for the numerical methods. The singularity of the problem can be reduced by using a numerical method based on a weak form of the equations. Such a method, combined with an interface tracking method to track the interfaces to which the discontinuities and singularities are confined, will require numerical quadrature with singular or discontinuous integrands. We introduce a class of numerical integration methods based on a regularization of the integrand. The methods can be of arbitrary high order of accuracy. Moment and regularity conditions control the overall accuracy.     

Contrôlabilité de l'équation des ondes à densité rapidement oscillante à une dimension d'espace

We consider the one-dimensional wave equation with periodic density of period ε → 0. By a counterexample due to Avellaneda, Bardos, and Rauch, we know that the boundary controllability property does not hold uniformly as ε → 0. We prove that the control remains uniformly bounded if we control the projection of the solution over the subspace generated by the eigenfunctions associated with the eigenvalues λ ≤ C_ε⁻², C > 0 being small enough. This result is sharp in the sense that the control diverges when the projection over the eigenfunctions such that λ ~ C_ε⁻², with C large, is controlled. We use the classical WKB asymptotic development that provides sharp results on the convergence of the spectrum and the theory of non-harmonic Fourier series.     

l 1-error estimates on the immersed interface upwind scheme for linear convection equations with piecewise constant coefficients: A simple proof

Abstract A linear convection equation with discontinuous coefcients arises in wave propagation through interfaces.An interface condition is needed at the interface to select a unique solution.An upwind scheme that builds this interface condition into its numerical flux is called the immersed interface upwind scheme.An l1-error estimate of such a scheme was frst established by Wen et al.(2008).In this paper,we provide a simple analysis on the l1-error estimate.The main idea is to formulate the solution to the underline initial-value problem into the sum of solutions to two convection equations with constant coefcients,which can then be estimated using classical methods for the initial or boundary value problems.     

On corner singularities in Reissner-Mindlin's theory of elastic plates

In the framework of linear elasticity, singularities occur in domains with non-smooth boundaries. Particularly in Fracture Mechanics, the local stress field near stress concentrations is of interest. In this work, singularities at re-entrant corners or sharp notches in Reissner-Mindlin plates are studied. Therefore, an asymptotic solution of the governing system of partial differential equations is obtained by using a complex potential approach which allows for an efficient calculation of the singularity exponent λ. The effect of the notch opening angle and the boundary conditions on the singularity exponent is discussed. The results show, that it can be distinguished between singularities for symmetric and antisymmetric loading and between singularities of the bending moments and the transverse shear forces. Also, stronger singularities than the classical crack tip singularity with free crack faces are observed. (© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)     

An accurate semi-analytic finite difference scheme for two-dimensional elliptic problems with singularities

A high-order semi-analytic finite difference scheme is presented to overcome degradation of numerical performance when applied to two-dimensional elliptic problems containing singular points. The scheme, called Least-Square Singular Finite Difference Scheme (L-S SFDS), applies an explicit functional representation of the exact solution in the vicinity of the singularities, and a conventional finite difference scheme on the remaining domain. It is shown that the L-S SFDS is “pollution” free, i.e., no degradation in the convergence rate occurs because of the singularities, and the coefficients of the asymptotic solution in the vicinity of the singularities are computed as a by-product with a very high accuracy. Numerical examples for the Laplace and Poisson equations over domains containing re-entrant corners or abrupt changes in the boundary conditions are presented. © 1998 John Wiley & Sons, Inc. Numer Methods Partial Differential Eq 14: 281–296, 1998     

On a Sharp Interface Model for Phase Transitions in Steel

Steel is one of the most widely used materials in the world with a broad spectrum of properties. The microstructure and the distribution of the different phases are of great importance, since they each possess different properties. A sharp interface model for the austenite-ferrite phase transition is presented. Mechanical effects due to eigenstrains resulting from the different densities of the phases are taken into account. The governing PDEs in each phase are a diffusion equation for the carbon concentration and the balance of momentum. Across the free interface, separating the two phases, the physical quantities may have discontinuities, which are controlled by jump conditions. Consistency of the model with the 2nd law of thermodynamics is shown. Numerical simulations for these types of free boundary problems are quite complex and involve appropriate methods to determine the interface position. One possibility to circumvent the explicit determination of the free boundary is the use of regularization techniques in form of phase field methods, where the interface is tracked implicitly. (© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Six combinations of the Ritz-Galerkin and finite element methods for elliptic boundary value problems

Overall comparisons are made for six efficient combinations of the Ritz-Galerkin and finite element methods for solving elliptic boundary value problems with singularities or interfaces. The comparisons are done by using theoretical analysis and numerical experiments. Significant relations among the six combinations are also found. A survey of the six combinations and their coupling strategies are given. These combinations are important not only for matching the Ritz-Galerkin method and the finite element method but also for matching other numerical methods such as the Ritz-Galerkin method and the finite difference method.     

Discontinuous Galerkin least-squares finite element methods for singularly perturbed reaction-diffusion problems with discontinuous coefficients and boundary singularities

In this paper, a discontinuous Galerkin least-squares finite element method is developed for singularly perturbed reaction-diffusion problems with discontinuous coefficients and boundary singularities by recasting the second-order elliptic equations as a system of first-order equations. In a companion paper (Lin in SIAM J Numer Anal 47:89–108, 2008) a similar method has been developed for problems with continuous data and shown to be well-posed, uniformly convergent, and optimal in convergence rate. In this paper the method is modified to take care of conditions that arise at interfaces and boundary singularities. Coercivity and uniform error estimates for the finite element approximation are established in an appropriately scaled norm. Numerical examples confirm the theoretical results.     

and Norms Error Estimates in Finite Element Method for Linear Parabolic Interface Problems

We derive new a priori error estimates for linear parabolic equations with discontinuous coefficients. Due to low global regularity of the solutions the error analysis of the standard finite element method for parabolic problems is difficult to adopt for parabolic interface problems. A finite element procedure is, therefore, proposed and analyzed in this paper. We are able to show that the standard energy technique of finite element method for non-interface parabolic problems can be extended to parabolic interface problems if we allow interface triangles to be curved triangles. Optimal pointwise-in-time error estimates in the L ²(Ω) and H ¹(Ω) norms are shown to hold for the semidiscrete scheme. A fully discrete scheme based on backward Euler method is analyzed and pointwise-in-time error estimates are derived. The interfaces are assumed to be arbitrary shape but smooth for our purpose.     

Robust a-posteriori error control of Cahn-Hilliard type equations with elasticity

Phase separation of an initially homogeneous mixture into two different phases can be modeled on a mesoscopic scale by the Cahn-Hilliard equation. The interface thickness between the pure phases enters as a small parameter γ into this mass conserving fourth order semilinear parabolic equation. Numerical analysis is well established for a fixed parameter size, but error estimates depend exponentially on γ^–1 and thus become useless if γ → 0. We consider the case, that elastic stresses due to a lattice misfit become important and the equation has to be coupled to a system of linear elasticity. Applications include e. g. the simulation of Sn-Cu alloys for the production of lead free solder or Ni-Al alloys used for rotor blade surfaces. (© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)