Parametric models for samples of random functions

A new class of parametric models, referred to as sample parametric models, is developed for random elements that match sample rather than the first two moments and/or other global properties of these elements. The models can be used to characterize, e.g., material properties at small scale in which case their samples represent microstructures of material specimens selected at random from a population. The samples of the proposed models are elements of finite-dimensional vector spaces spanned by samples, eigenfunctions of Karhunen-Loeve (KL) representations, or modes of singular value decompositions (SVDs). The implementation of sample parametric models requires knowledge of the probability laws of target random elements. Numerical examples including stochastic processes and random fields are used to demonstrate the construction of sample parametric models, assess their accuracy, and illustrate how these models can be used to solve efficiently stochastic equations.

[1]  Raúl Tempone,et al.  Galerkin Finite Element Approximations of Stochastic Elliptic Partial Differential Equations , 2004, SIAM J. Numer. Anal..

[2]  G. Karniadakis,et al.  Multi-Element Generalized Polynomial Chaos for Arbitrary Probability Measures , 2006, SIAM J. Sci. Comput..

[3]  R. Ghanem,et al.  Stochastic Finite Elements: A Spectral Approach , 1990 .

[4]  D. O’Leary,et al.  Efficient iterative algorithms for the stochastic finite element method with application to acoustic scattering , 2005 .

[5]  BabuskaIvo,et al.  A Stochastic Collocation Method for Elliptic Partial Differential Equations with Random Input Data , 2007 .

[6]  Christoph Schwab,et al.  Karhunen-Loève approximation of random fields by generalized fast multipole methods , 2006, J. Comput. Phys..

[7]  Nikos E. Mastorakis Singular value decomposition in multidimensional arrays , 1996, Int. J. Syst. Sci..

[8]  G. Karniadakis,et al.  An adaptive multi-element generalized polynomial chaos method for stochastic differential equations , 2005 .

[9]  Singular Value Decomposition for Multidimensional Matrices , 2013 .

[10]  Mircea Grigoriu Response Statistics for Random Heterogeneous Microstructures , 2014, SIAM/ASA J. Uncertain. Quantification.

[11]  M. Talagrand The Glivenko-Cantelli Problem , 1987 .

[12]  Bert Fristedt,et al.  A modern approach to probability theory , 1996 .

[13]  M. Grigoriu Stochastic Calculus: Applications in Science and Engineering , 2002 .

[14]  E. O. Brigham,et al.  The Fast Fourier Transform , 1967, IEEE Transactions on Systems, Man, and Cybernetics.

[15]  Mircea Grigoriu A method for solving stochastic equations by reduced order models and local approximations , 2012, J. Comput. Phys..

[16]  C. Eckart,et al.  The approximation of one matrix by another of lower rank , 1936 .

[17]  Ivo Babuška,et al.  On solving stochastic initial-value differential equations , 2003 .

[18]  H. Kuo Introduction to stochastic integration , 2006 .

[19]  Vagn Lundsgaard Hansen Functional Analysis: Entering Hilbert Space , 2006 .

[20]  G. W. Stewart,et al.  On the Early History of the Singular Value Decomposition , 1993, SIAM Rev..

[21]  Mircea Grigoriu,et al.  Probabilistic models for stochastic elliptic partial differential equations , 2010, J. Comput. Phys..

[22]  Diego Bricio Hernández,et al.  Lectures on Probability and Second Order Random Fields , 1995 .

[23]  Neil Muller,et al.  Singular Value Decomposition, Eigenfaces, and 3D Reconstructions , 2004, SIAM Rev..

[24]  Mircea Grigoriu,et al.  On the accuracy of the polynomial chaos approximation for random variables and stationary stochastic processes. , 2003 .

[25]  P. Frauenfelder,et al.  Finite elements for elliptic problems with stochastic coefficients , 2005 .

[26]  R. Adler,et al.  The Geometry of Random Fields , 1982 .

[27]  Fabio Nobile,et al.  A Stochastic Collocation Method for Elliptic Partial Differential Equations with Random Input Data , 2007, SIAM Rev..

[28]  M. Powell,et al.  Approximation theory and methods , 1984 .

[29]  Christian Soize,et al.  Non-Gaussian simulation using Hermite polynomial expansion: convergences and algorithms , 2002 .

[30]  George E. Karniadakis,et al.  The multi-element probabilistic collocation method (ME-PCM): Error analysis and applications , 2008, J. Comput. Phys..

[31]  Joos Vandewalle,et al.  A Multilinear Singular Value Decomposition , 2000, SIAM J. Matrix Anal. Appl..

[32]  L. Devroye Bounds for the Uniform Deviation of Empirical Measures , 1982 .

[33]  Stanisław Mordwa Doran B.J., Burgess M.B., 2012, Putting Fear of Crime on the Map. Investigating Perceptions of Crime Using Geographic Information Systems, Springer, New York–Dordrecht–Heidelberg–London, ss. 283 , 2012 .

[34]  R. Ghanem,et al.  Polynomial chaos decomposition for the simulation of non-gaussian nonstationary stochastic processes , 2002 .

[35]  W. Root,et al.  An introduction to the theory of random signals and noise , 1958 .

[36]  Christian Soize,et al.  Uncertainty quantification in computational stochastic multiscale analysis of nonlinear elastic materials , 2013 .

[37]  V. Bryant Metric Spaces: Iteration and Application , 1985 .

[38]  D. Xiu Numerical Methods for Stochastic Computations: A Spectral Method Approach , 2010 .