Numerische Mathematik

, Volume 131, Issue 2, pp 329–368

Quasi-Monte Carlo finite element methods for elliptic PDEs with lognormal random coefficients

  • I. G. Graham
  • F. Y. Kuo
  • J. A. Nichols
  • R. Scheichl
  • Ch. Schwab
  • I. H. Sloan
Article

Abstract

In this paper we analyze the numerical approximation of diffusion problems over polyhedral domains in \(\mathbb {R}^d\) (\(d = 1, 2,3\)), with diffusion coefficient \(a({\varvec{x}},\omega )\) given as a lognormal random field, i.e., \(a({\varvec{x}},\omega ) = \exp (Z({\varvec{x}},\omega ))\) where \({\varvec{x}}\) is the spatial variable and \(Z({\varvec{x}}, \cdot )\) is a Gaussian random field. The analysis presents particular challenges since the corresponding bilinear form is not uniformly bounded away from \(0\) or \(\infty \) over all possible realizations of \(a\). Focusing on the problem of computing the expected value of linear functionals of the solution of the diffusion problem, we give a rigorous error analysis for methods constructed from (1) standard continuous and piecewise linear finite element approximation in physical space; (2) truncated Karhunen–Loève expansion for computing realizations of \(a\) (leading to a possibly high-dimensional parametrized deterministic diffusion problem); and (3) lattice-based quasi-Monte Carlo (QMC) quadrature rules for computing integrals over parameter space which define the expected values. The paper contains novel error analysis which accounts for the effect of all three types of approximation. The QMC analysis is based on a recent result on randomly shifted lattice rules for high-dimensional integrals over the unbounded domain of Euclidean space, which shows that (under suitable conditions) the quadrature error decays with \(\mathcal {O}(n^{-1+\delta })\) with respect to the number of quadrature points \(n\), where \(\delta >0\) is arbitrarily small and where the implied constant in the asymptotic error bound is independent of the dimension of the domain of integration.

Mathematics Subject Classification

65D30 65D32 65N30 

References

  1. 1.
    Adler, R.J.: The Geometry of Random Fields. Wiley, London (1981)MATHGoogle Scholar
  2. 2.
    Bogachev, V.I.: Gaussian Measures, AMS Monographs, vol. 62. American Mathematical Society, RI (1998)Google Scholar
  3. 3.
    Charrier, J.: Strong and weak error estimates for elliptic partial differential equations with random coefficients. SIAM J. Numer. Anal. 50, 216–246 (2012)MathSciNetCrossRefMATHGoogle Scholar
  4. 4.
    Charrier, J., Debussche, A.: Weak truncation error estimates for elliptic PDEs with lognormal coefficients. Stoch. PDEs: Anal. Comput. 1, 63–93 (2013)MathSciNetCrossRefMATHGoogle Scholar
  5. 5.
    Charrier, J., Scheichl, R., Teckentrup, A.L.: Finite element error analysis of elliptic PDEs with random coefficients and its application to multilevel Monte Carlo methods. SIAM J. Numer. Anal. 51, 322–352 (2013)MathSciNetCrossRefMATHGoogle Scholar
  6. 6.
    Cools, R., Kuo, F.Y., Nuyens, D.: Constructing embedded lattice rules for multivariate integration. SIAM J. Sci. Comput. 28, 2162–2188 (2006)MathSciNetCrossRefMATHGoogle Scholar
  7. 7.
    Da Prato, G., Zabczyk, J.: Stochastic Equations in Infinite Dimensions, Vol. 44 of Encyclopedia of Mathematics and its Applications. Cambridge University Press, Cambridge (1992)CrossRefGoogle Scholar
  8. 8.
    Dick, J.: On the convergence rate of the component-by-component construction of good lattice rules. J. Complexity 20, 493–522 (2004)MathSciNetCrossRefMATHGoogle Scholar
  9. 9.
    Dick, J., Kuo, F.Y., Sloan, I.H.: High dimensional integration—the quasi-Monte Carlo way. Acta Numer. 22, 133–288 (2013)MathSciNetCrossRefMATHGoogle Scholar
  10. 10.
    Dick, J., Pillichshammer, F.: Digital Nets and Sequences. Discrepancy Theory and Quasi-Monte Carlo Integration. Cambridge University Press, Cambridge (2010)MATHGoogle Scholar
  11. 11.
    Dick, J., Pillichshammer, F., Waterhouse, B.J.: The construction of good extensible rank-1 lattices. Math. Comp. 77, 2345–2374 (2008)MathSciNetCrossRefMATHGoogle Scholar
  12. 12.
    Frauenfelder, P., Schwab, Ch., Todor, R.A.: Finite elements for elliptic problems with stochastic coefficients. Compu. Methods Appl. Mech. Eng. 194, 205–228 (2005)Google Scholar
  13. 13.
    Ghanem, R.G., Spanos, P.D.: Stochastic Finite Elements. Dover, New York (1991)CrossRefMATHGoogle Scholar
  14. 14.
    Gittelson, C.J.: Stochastic Galerkin discretization of the isotropic lognormal diffusion problem. Math. Models Methods Appl. Sci. 20, 237–263 (2010)MathSciNetCrossRefMATHGoogle Scholar
  15. 15.
    Graham, I.G., Kuo, F.Y., Nuyens, D., Scheichl, R., Sloan, I.H.: Quasi-Monte Carlo methods for elliptic PDEs with random coefficients and applications. J. Comput. Phys. 230, 3668–3694 (2011)MathSciNetCrossRefMATHGoogle Scholar
  16. 16.
    Hickernell, F.J., Sloan, I.H., Wasilkowski, G.W.: On tractability of weighted integration for certain Banach spaces of functions. In: Niederreiter, H. (ed.) Monte Carlo and Quasi-Monte Carlo Methods 2002, pp. 51–71. Springer-Verlag, Berlin (2004)CrossRefGoogle Scholar
  17. 17.
    Hoang, V.-H., Schwab, Ch.: \(N\)-term Wiener chaos approximation rates for elliptic PDEs with lognormal Gaussian random inputs. Math. Models Methods Appl. Sci. 24, 797–826 (2014)Google Scholar
  18. 18.
    Kuo, F.Y.: Component-by-component constructions achieve the optimal rate of convergence. J. Complexity 19, 301–320 (2003)MathSciNetCrossRefMATHGoogle Scholar
  19. 19.
    Kuo, F.Y., Schwab, Ch., Sloan, I.H.: Quasi-Monte Carlo finite element methods for a class of elliptic partial differential equations with random coefficient. SIAM J. Numer. Anal. 50, 3351–3374 (2012)Google Scholar
  20. 20.
    Kuo, F.Y., Schwab, Ch., Sloan, I.H.: Quasi-Monte Carlo methods for very high dimensional integration: the standard weighted-space setting and beyond. ANZIAM J. 53, 1–37 (2011)Google Scholar
  21. 21.
    Kuo, F.Y., Sloan, I.H., Wasilkowski, G.W., Waterhouse, B.J.: Randomly shifted lattice rules with the optimal rate of convergence for unbounded integrands. J. Complexity 26, 135–160 (2010)MathSciNetCrossRefMATHGoogle Scholar
  22. 22.
    Kuo, F.Y., Wasilkowski, G.W., Waterhouse, B.J.: Randomly shifted lattice rules for unbounded integrals. J. Complexity 22, 630–651 (2006)MathSciNetCrossRefMATHGoogle Scholar
  23. 23.
    Lord, G.J., Powell, C.E., Shardlow, T.: An Introduction to Computational Stochastic PDEs. Cambridge University Press, Cambridge (2014)CrossRefMATHGoogle Scholar
  24. 24.
    McLean, W.: Strongly Elliptic Systems and Boundary Integral Equations. Cambridge University Press, Cambridge (2000)MATHGoogle Scholar
  25. 25.
    Naff, R.L., Haley, D.F., Sudicky, E.A.: High-resolution Monte Carlo simulation of flow and conservative transport in heterogeneous porous media 1. Methodology and flow results. Water Resour. Res. 34, 663–677 (1998)Google Scholar
  26. 26.
    Naff, R.L., Haley, D.F., Sudicky, E.A.: High-resolution Monte Carlo simulation of flow and conservative transport in heterogeneous porous media 2. Transport Results. Water Resour. Res. 34, 679–697 (1998)CrossRefGoogle Scholar
  27. 27.
    Nichols, J.A.: Quasi-Monte Carlo methods with applications to partial differential equations with random coefficients. Ph.D. Thesis, University of New South Wales (2014)Google Scholar
  28. 28.
    Nichols, J.A., Kuo, F.Y.: Fast CBC construction of randomly shifted lattice rules achieving \({{\cal O}}(N^{-1+\delta })\) convergence for unbounded integrands in \({{\mathbb{R}}}^s\) in weighted spaces with POD weights. J. Complexity 30, 444–468 (2014)MathSciNetCrossRefMATHGoogle Scholar
  29. 29.
    Nuyens, D., Cools, R.: Fast algorithms for component-by-component construction of rank-\(1\) lattice rules in shift-invariant reproducing kernel Hilbert spaces. Math. Comp. 75, 903–920 (2006)MathSciNetCrossRefMATHGoogle Scholar
  30. 30.
    Nuyens, D., Cools, R.: Fast component-by-component construction of rank-\(1\) lattice rules with a non-prime number of points. J. Complexity 22, 4–28 (2006)MathSciNetCrossRefMATHGoogle Scholar
  31. 31.
    Schwab, Ch., Gittelson, C.J.: Sparse tensor discretizations of high dimensional and stochastic PDEs. Acta Numer. 20, 291–467 (2011)Google Scholar
  32. 32.
    Schwab, Ch., Todor, R.A.: Karhunen–Loève approximation of random fields by generalized fast multipole methods. J. Comput. Phys. 217, 100–122 (2006)Google Scholar
  33. 33.
    Sloan, I.H., Kuo, F.Y., Joe, S.: Constructing randomly shifted lattice rules in weighted Sobolev spaces. SIAM J. Numer. Anal. 40, 1650–1665 (2002)MathSciNetCrossRefMATHGoogle Scholar
  34. 34.
    Sloan, I.H., Woźniakowski, H.: When are quasi-Monte Carlo algorithms efficient for high-dimensional integrals? J. Complexity 14, 1–33 (1998), 125, 569–600 (2013)Google Scholar
  35. 35.
    Sloan, I.H., Wang, X., Woźniakowski, H.: Finite-order weights imply tractability of multivariate integration. J. Complexity 20, 46–74 (2004)MathSciNetCrossRefMATHGoogle Scholar
  36. 36.
    Teckentrup, A.L., Scheichl, R., Giles, M.B., Ullmann, E.: Further analysis of multilevel Monte Carlo methods for elliptic PDEs with random coefficients. Numer. Math. 125, 569–600 (2013)MathSciNetCrossRefMATHGoogle Scholar
  37. 37.
    Wasilkowski, G.W., Woźniakowski, H.: Complexity of weighted approximation over \(\mathbb{R}^1\). J. Approx. Theory. 103, 223–251 (2000)MathSciNetCrossRefMATHGoogle Scholar
  38. 38.
    Wasilkowski, G.W., Woźniakowski, H.: Tractability of approximation and integration for weighted tensor product problems over unbounded domains. In: Fang, K.-T., Hickernell, F.J., Niederreiter, H. (eds.) Monte Carlo and Quasi-Monte Carlo Methods 2000, pp. 497–522. Springer, Berlin (2002)CrossRefGoogle Scholar
  39. 39.
    Widom, H.: Asymptotic behaviour of the eigenvalues of certain integral operators. Trans. Am. Math. Soc. 109, 278–295 (1963)MathSciNetCrossRefMATHGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2014

Authors and Affiliations

  • I. G. Graham
    • 1
  • F. Y. Kuo
    • 2
  • J. A. Nichols
    • 2
  • R. Scheichl
    • 1
  • Ch. Schwab
    • 3
  • I. H. Sloan
    • 2
  1. 1.Department of Mathematical SciencesUniversity of BathBathUK
  2. 2.School of Mathematics and StatisticsUniversity of NSWSydneyAustralia
  3. 3.Seminar für Angewandte MathematikETH ZürichZurichSwitzerland

Personalised recommendations