Abstract
We present an algorithm for computing the nonlinear interface dynamics of the Mullins-Sekerka model for interfaces that are defined implicitly (e.g. by a level set function) using integral equations. The computation of the dynamics involves solving Laplace’s equation with Dirichlet boundary conditions on multiply connected and unbounded domains and propagating the interface using a normal velocity obtained from the solution of the PDE at each time step. Our method is based on a simple formulation for implicit interfaces, which rewrites boundary integrals as volume integrals over the entire space. The resulting algorithm thus inherits the benefits of both level set methods and boundary integral methods to simulate the nonlocal front propagation problem with possible topological changes. We present numerical results in both two and three dimensions to demonstrate the effectiveness of the algorithm.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
D. Adalsteinsson, J.A. Sethian, A fast level set method for propagating interfaces. J. Comput. Phys. 118(2), 269–277 (1995)
K.E. Atkinson, in The Numerical Solution of Integral Equations of the Second Kind (Cambridge University Press, Cambridge 1997)
K.E. Atkinson, G. Chandler, Boundary integral equation methods for solving Laplace’s equation with nonlinear boundary conditions: the smooth boundary case. Math. Comput. 55(191), 451–472 (1990)
I. Babus̆ka, The finite element method for elliptic equations with discontinuous coefficients. Computing 5, 207–213 (1970)
J.W. Barrett, H. Garcke, R. Nürnberg, On stable parametric finite element methods for the Stefan problem and the Mullins-Sekerka problem with applications to dendritic growth. J. Comput. Phys. 229(18), 6270–6299 (2010)
P.W. Bates, X. Chen, X. Deng, A numerical scheme for the two phase Mullins-Sekerka problem. Electron. J. Differ. Equs. (1995)
J. Bedrossian, J.J. von Brecht, S. Zhu, E. Sifakis, J. Teran, A second order virtual node method for elliptic problems with interfaces and irregular domains. J. Comput. Phys. 229, 6405–6426 (2010)
S. Börn, L. Grasedyck, W. Hackbusch, Hierarchical Matrices (Technical report, Max-Planck Institut fur Mathematik in den Naturwissenschaften, Leipzig, 2003)
S. Chen, B. Merriman, S. Osher, P. Smereka, A simple level set method for solving Stefan problem. J. Comput. Phys. 135(1), 8–29 (1997)
L.-T. Cheng, Y.-H. Tsai, Redistancing by flow time dependent Eikonal equation. J. Comput. Phys. 227(2), 4002–4017 (2008)
I.-L. Chern, Y.-C. Shu, A coupling interface method for elliptic interface problems. J. Comput. Phys. 225, 2138–2174 (2007)
P.G. Ciarlet, in The Finite Element Method for Elliptic Problems. SIAM. North-Holland, Amsterdam (1978)
S. Conti, B. Niethammer, F. Otto, Coarsening rates in off-critical mixtures. SIAM J. Math. Anal. 37(6), 1732–1741 (2006)
M. Dillencourt, H. Samet, M. Tamminen, A general approach to connected-component labeling for arbitrary image representations. J. ACM 39(2), 253–280 (1992)
J. Dolbow, I. Harari, An efficient finite element method for embedded interface problems. J. Numer. Methods Eng. 78, 229–252 (2009)
H. Federer, in Geometric Measure Theory (Springer, Berlin, 1969)
G.B. Folland, in Introduction to Partial Differential Equations (Princeton University Press, Princeton, 1976)
F. Gibou, R. Fedkiw, A fourth order accurate discretization for the Laplace and heat equations on arbitrary domains, with applications to the Stefan problem. J. Comput. Phys. 202, 577–601 (2005)
F. Gibou, R. Fedkiw, L. Cheng, M. Kang, A second order accurate symmetric discretization of the Poisson equation on irregular domains. J. Comput. Phys. 176, 1–23 (2002)
Y. Giga, in Surface Evolution Equations: A Level Set Approach. Monographs in Mathematics, vol. 99 (Birkhäuser Verlag, Basel, 2006)
A. Greenbaum, L. Greengard, G.B. McFadden, Laplace’s equation and the Dirichlet-Neumann map in multiply connected domains. J. Comput. Phys. 105, 267–278 (1993)
L. Greengard, V. Rokhlin, A fast algorithm for particle simulations. J. Comput. Phys. 73(2), 325–348 (1987)
M. Gurtin, On the two-phase Stefan problem with interfacial energy and entropy. Arch. Rational Mech. Anal. 96, 199–241 (1986)
A. Hansbo, P. Hansbo, An unfitted element method, based on Nitsche’s method for elliptic interface problems. Comput. Methods Appl. Mech. Eng. 191, 5537–5552 (2002)
A. Hansbo, P. Hansbo, A finite element method for the simulation of strong and weak discontinuities in solid mechanics. Comput. Methods Appl. Mech. Eng. 193, 3523–3540 (2004)
J. Huang, J. Zou, A mortar element method for elliptic problems with discontinuous coefficients. IMA J. Numer. Anal. 22, 549–576 (2002)
H. Johansen, Cartesian grid embedded boundary finite difference methods for elliptic and parabolic differential equations on irregular domains. Ph.D. thesis, University of California, Berkeley, 1997
H. Johansen, P. Colella, A cartesian grid embedded boundary method for Poisson’s equation on irregular domains. J. Comput. Phys. 147, 60–85 (1998)
G. Karali, P. Kevrekidis, Bubble interactions for the Mullins-Sekerka problem: some case examples. Math. Comput. Simul. 80(4), 707–720 (2009)
R. Kress, Linear Integral Equations, 2nd edn. (Springer, New York, 1999)
C. Kublik, R. Tsai, Integration over curves and surfaces defined by the closest point mapping. Res. Math.Sci. 3, 1–17 (2016)
C. Kublik, N.M. Tanushev, R. Tsai, An implicit interface boundary integral method for Poisson’s equation on arbitrary domains. J. Comput. Phys. 247, 269–311 (2013)
R. Leveque, Z. Li, The immersed interface method for elliptic equations with discontinuous coefficients and singular sources. SIAM J. Numer. Anal. 31, 1019–1044 (1994)
Z. Li , K. Ito, The immersed interface method: numerical solutions of PDEs involving interfaces and irregular domains (frontiers in applied mathematics), Society for Industrial and Applied Mathematics (2006)
X. Liu, R. Fedkiw, M. Kang, A boundary condition capturing method for Poisson’s equation on irregular domains. J. Comput. Phys. 160(1), 151–178 (2000)
S.G. Mikhlin, in Integral Equations (Pergamon, London, 1957)
T.-H. Miura, Zero width limit of the heat equation on moving thin domains. UTMS Preprint Series, the University of Tokyo, 2015
W.W. Mullins, R.F. Sekerka, Morphological stability of a particle growing by diffusion or heat flow. J. Appl. Phys. 34, 323–329 (1963)
E. Nyström, Über die praktische Auflösung von Integralgleichungen mit Anwendungen auf Randwertaufgaben. Acta Math. 54, 185–204 (1930)
S. Osher, J.A. Sethian, Fronts propagating with curvature dependent speed: algorithms based on Hamilton-Jacobi formulations. J. Comput. Phys. 79, 12–49 (1988)
D. Peng, B. Merriman, S. Osher, H.-K. Zhao, M. Kang, A PDE-based fast local level set method. J. Comput. Phys. 155(2), 410–438 (1999)
G. Russo, P. Smereka, A remark on computing distance functions. J. Comput. Phys. 163, 51–67 (2000)
J. Sethian, A fast marching level set method for monotonically advancing fronts. Proc. Natl. Acad. Sci. 93(4), 1591–1595 (1996)
Y.-H. Tsai, L. Cheng, S. Osher, H.-K. Zhao, Fast sweeping methods for a class of Hamilton-Jacobi equations. SIAM J. Numer. Anal. 41(2), 673–694 (2003)
J. Tsitsiklis, Efficient algorithms for globally optimal trajectories. IEEE Trans. Autom. Control 40, 1528–1538 (1995)
J. Zhu, X. Chen, T.Y. Hou, An efficient boundary integral method for the Mullins-Sekerka problem. J. Comput. Phys. 127, 246–267 (1996)
Acknowledgements
Kublik’s research was partially funded by a University of Dayton Research Council Seed Grant and a Dr. Schraut Faculty Research Award. Chen’s and Tsai’s research is partially supported by Simons Foundation, NSF Grants DMS-1318975, DMS-1217203, and ARO Grant No. W911NF-12-1-0519.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendix
Appendix
1.1 Connected Component Labeling
As explained in Section 2, the system of equations to solve depends on many properties including the number of connected components of the region, whether the region is bounded or not (exterior vs interior), and the orientation of the region. In our numerical simulations, we adapted a technique called connected component labeling (CCL), see e.g. [14], to find necessary topological information needed to obtain the correct set of equations. These include:
-
1.
The boundedness of the connected component \(C_{i}\). This decides which formulation (interior vs exterior) to use.
-
2.
The orientation of \(C_{i}\) which determines the normal direction.
-
3.
The total number of connected boundary components of the boundary of \(C_{i}\), denoted by \(\varGamma ^{i} = \partial C_{i}\).
-
4.
The selection of \(\mathbf {z}_{i}\) in (9) and (11) for each component \(\varGamma ^j\). This involves
-
the separation of \(\varGamma ^{i}\) into boundary components \(\varGamma ^{i}_{j}\), each bounding a hole in the region except for the most exterior one. We denote the exterior boundary as \(\varGamma ^{i}_{0}\).
-
For each hole delimited by \(\varGamma ^{i}_{j}~,j \ne 0\), find a point \(\mathbf {z}_{i}\) inside the hole that gives the least singular value of \(\Phi (\mathbf {x}-\mathbf {z}_{i})\). This means that \(\mathbf {z_{i}}\) should be as far from the boundary as possible.
-
Since we deal with closed interfaces, every grid point belongs to a unique connected component and has a well-defined component label. Note that a connected component \(C_{i}\) may belong to \(\varOmega \) or \(\mathbf R^m \setminus \varOmega \).
We use the algorithm as follows:
-
We label the unbounded component as \(C_{0}\). This is the only true exterior component for the boundary integral formulations.
-
Each interior component (\(d_{\varGamma }>0\), i.e. the solid phase) will have positive label i and each exterior component (\(d_{\varGamma }<0\) i.e. the liquid phase) will take negative label \(-i\) (except for the unbounded one).
-
Each boundary piece of each component will have label j.
-
The point \(\mathbf {z_{i}}\) in \(C_{i}\) is chosen to be the point with largest signed distance function in absolute value.
We adopt the two-pass CCL algorithm with 2m-connectivity in \(\mathbb R^m\). This algorithm uses equivalence classes for labels: after the first pass, points in the same connected component may not have the same label, but the labels of points in the same connected component will be assigned to the same equivalent class in the second pass. The root of the equivalence class denotes the smallest label (in absolute value) in the equivalence class. The scanning process works as follows:
-
1.
First Pass
-
a.
Begins with label 0 which denotes the unbounded component \(C_{0}\).
-
b.
At the current point scanned, we look at the sign of the signed distance function of its 2m neighbors already visited (in 2D, west and north of the current point)
-
If none of the neighbors have the same sign as the current point, we create a new label (positive or negative, based on the sign of the distance function at the current point) and a new equivalence class.
-
If only one neighbor has the same sign, we pick the label for the current point to be the root of that neighboring point’s equivalence class.
-
If more than one neighbor has the same sign as the current point, we pick the label for the current point to be the smallest root of the neighbors that have the same sign’s equivalence classes. Furthermore, we combine the equivalence classes of the neighboring points that have the same sign since they are connected through the current point.
-
-
c.
The largest root of all equivalence classes with a given sign (interior or exterior) will give the total number of connected components that have that sign. Thus, the sum is the total number of connected components.
-
a.
-
2.
Second Pass
-
a.
At each point, we assign its label to be the root of its equivalence class.
-
b.
We update and store the points with largest absolute distance within each equivalence class. These are the points \(\mathbf {z_{i}}\).
-
c.
For each point \(\mathbf {x}\) within the \(\varepsilon \) tubular neighborhood of the boundary \((|d_{\varGamma }|< \varepsilon )\), we look for the root of its equivalence class (say i) and the root of its projection point’s (\(P_{\varGamma }(\mathbf {x})\)) equivalence class (say j) that takes the opposite sign. To obtain j, we look at the vertices of the cell \(P_{\varGamma }(\mathbf {x})\) falls in and scan for the label with opposite sign. This step identifies the boundary piece \(\varGamma ^{i}_{j}\) and collects points within the \(\varepsilon \) neighborhood of the boundary \(\varGamma ^{i}_{j}\).
-
d.
We store the total number of connected boundary pieces.
-
a.
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Chen, C., Kublik, C., Tsai, R. (2017). An Implicit Boundary Integral Method for Interfaces Evolving by Mullins-Sekerka Dynamics. In: Maekawa, Y., Jimbo, S. (eds) Mathematics for Nonlinear Phenomena — Analysis and Computation. MNP2015 2015. Springer Proceedings in Mathematics & Statistics, vol 215. Springer, Cham. https://doi.org/10.1007/978-3-319-66764-5_1
Download citation
DOI: https://doi.org/10.1007/978-3-319-66764-5_1
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-66762-1
Online ISBN: 978-3-319-66764-5
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)