Fast evaluation of singular BEM integrals based on tensor approximations (Q443849): Difference between revisions

From MaRDI portal
Importer (talk | contribs)
Created a new Item
 
ReferenceBot (talk | contribs)
Changed an Item
 
(4 intermediate revisions by 4 users not shown)
Property / review text
 
In boundary element methods (BEMs) one often is required to evaluate integrals of the form \[ \displaystyle\int_{\sigma}\int_{\tau}\varphi(x)k(x,y)\psi(y)dydx, \] where \(\sigma,\tau\subset \Gamma\) are boundary elements on the boundary \(\Gamma:=\partial\Omega\) of a domain \(\Omega\subset R^l\) \((l=2,3)\) and \(k:\mathbb R^l\times \mathbb R^l\to \mathbb R\) is a kernel function that has a singularity at \(x=y\) and is analytic elsewhere. In this paper, the author proposes a new method for the fast evaluation of integrals stemming from boundary element methods including discretisations of the classical single and double layer potential operators. The key idea behind this new approach is the investigation of functions of the type \[ f(a):=\displaystyle\int_{\sigma(a)}\int_{\tau(a)}\varphi(x)k(x,y)\psi(y)dydx, \] where \(a\in \mathbb R^d\) is a parameter tuple describing the domains \(\sigma(a)\) and \(\tau(a)\). The function \(f\) is smooth in a \(d\)-dimensional box. A standard interpolation of \(f\) by polynomials leads to a \(d\)-dimensional tensor which is given by the values of \(f\) at the interpolation points. This tensor may be approximated in a low rank tensor format like the (CP) format or the \({\mathcal H}\)-Tucker format. The tensor approximation has to be done only once and allows us to evaluate interpolants in \({\mathcal O}(dr(m+1))\) operations in the (CP) format, or \({\mathcal O}(dk^3+dk(m+1))\) operations in the \({\mathcal H}\)-Tucker format, where \(m\) denotes the interpolation order and the ranks \(r,k\) are small integers. The author demonstrates that highly accurate integral values can be obtained at very moderate costs. Finally, the potential of the new approach is illustrated by a number of numerical examples.
Property / review text: In boundary element methods (BEMs) one often is required to evaluate integrals of the form \[ \displaystyle\int_{\sigma}\int_{\tau}\varphi(x)k(x,y)\psi(y)dydx, \] where \(\sigma,\tau\subset \Gamma\) are boundary elements on the boundary \(\Gamma:=\partial\Omega\) of a domain \(\Omega\subset R^l\) \((l=2,3)\) and \(k:\mathbb R^l\times \mathbb R^l\to \mathbb R\) is a kernel function that has a singularity at \(x=y\) and is analytic elsewhere. In this paper, the author proposes a new method for the fast evaluation of integrals stemming from boundary element methods including discretisations of the classical single and double layer potential operators. The key idea behind this new approach is the investigation of functions of the type \[ f(a):=\displaystyle\int_{\sigma(a)}\int_{\tau(a)}\varphi(x)k(x,y)\psi(y)dydx, \] where \(a\in \mathbb R^d\) is a parameter tuple describing the domains \(\sigma(a)\) and \(\tau(a)\). The function \(f\) is smooth in a \(d\)-dimensional box. A standard interpolation of \(f\) by polynomials leads to a \(d\)-dimensional tensor which is given by the values of \(f\) at the interpolation points. This tensor may be approximated in a low rank tensor format like the (CP) format or the \({\mathcal H}\)-Tucker format. The tensor approximation has to be done only once and allows us to evaluate interpolants in \({\mathcal O}(dr(m+1))\) operations in the (CP) format, or \({\mathcal O}(dk^3+dk(m+1))\) operations in the \({\mathcal H}\)-Tucker format, where \(m\) denotes the interpolation order and the ranks \(r,k\) are small integers. The author demonstrates that highly accurate integral values can be obtained at very moderate costs. Finally, the potential of the new approach is illustrated by a number of numerical examples. / rank
 
Normal rank
Property / reviewed by
 
Property / reviewed by: Ana-Maria Acu / rank
 
Normal rank
Property / Mathematics Subject Classification ID
 
Property / Mathematics Subject Classification ID: 65D32 / rank
 
Normal rank
Property / Mathematics Subject Classification ID
 
Property / Mathematics Subject Classification ID: 65N38 / rank
 
Normal rank
Property / Mathematics Subject Classification ID
 
Property / Mathematics Subject Classification ID: 15A69 / rank
 
Normal rank
Property / Mathematics Subject Classification ID
 
Property / Mathematics Subject Classification ID: 41A55 / rank
 
Normal rank
Property / Mathematics Subject Classification ID
 
Property / Mathematics Subject Classification ID: 41A63 / rank
 
Normal rank
Property / zbMATH DE Number
 
Property / zbMATH DE Number: 6065173 / rank
 
Normal rank
Property / zbMATH Keywords
 
boundary element methods
Property / zbMATH Keywords: boundary element methods / rank
 
Normal rank
Property / zbMATH Keywords
 
interpolation
Property / zbMATH Keywords: interpolation / rank
 
Normal rank
Property / zbMATH Keywords
 
tensor
Property / zbMATH Keywords: tensor / rank
 
Normal rank
Property / zbMATH Keywords
 
evaluation of integral
Property / zbMATH Keywords: evaluation of integral / rank
 
Normal rank
Property / zbMATH Keywords
 
numerical examples
Property / zbMATH Keywords: numerical examples / rank
 
Normal rank
Property / MaRDI profile type
 
Property / MaRDI profile type: MaRDI publication profile / rank
 
Normal rank
Property / full work available at URL
 
Property / full work available at URL: https://doi.org/10.1007/s00211-011-0436-6 / rank
 
Normal rank
Property / OpenAlex ID
 
Property / OpenAlex ID: W2041973919 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Black box approximation of tensors in hierarchical Tucker format / rank
 
Normal rank
Property / cites work
 
Property / cites work: Differentiability of strongly singular and hypersingular boundary integral formulations with respect to boundary perturbations / rank
 
Normal rank
Property / cites work
 
Property / cites work: Efficient automatic quadrature in 3-D Galerkin BEM / rank
 
Normal rank
Property / cites work
 
Property / cites work: Black box low tensor-rank approximation using fiber-crosses / rank
 
Normal rank
Property / cites work
 
Property / cites work: Hierarchical Singular Value Decomposition of Tensors / rank
 
Normal rank
Property / cites work
 
Property / cites work: A new scheme for the tensor representation / rank
 
Normal rank
Property / cites work
 
Property / cites work: Tensor Decompositions and Applications / rank
 
Normal rank
Property / cites work
 
Property / cites work: On a Class of Winkler‐Type Contact Problems / rank
 
Normal rank
Property / cites work
 
Property / cites work: Breaking the Curse of Dimensionality, Or How to Use SVD in Many Dimensions / rank
 
Normal rank
Property / cites work
 
Property / cites work: TT-cross approximation for multidimensional arrays / rank
 
Normal rank
Property / cites work
 
Property / cites work: The fast solution of boundary integral equations. / rank
 
Normal rank
Property / cites work
 
Property / cites work: Kernel Properties and Representations of Boundary Integral Operators / rank
 
Normal rank
links / mardi / namelinks / mardi / name
 

Latest revision as of 13:55, 5 July 2024

scientific article
Language Label Description Also known as
English
Fast evaluation of singular BEM integrals based on tensor approximations
scientific article

    Statements

    Fast evaluation of singular BEM integrals based on tensor approximations (English)
    0 references
    0 references
    13 August 2012
    0 references
    In boundary element methods (BEMs) one often is required to evaluate integrals of the form \[ \displaystyle\int_{\sigma}\int_{\tau}\varphi(x)k(x,y)\psi(y)dydx, \] where \(\sigma,\tau\subset \Gamma\) are boundary elements on the boundary \(\Gamma:=\partial\Omega\) of a domain \(\Omega\subset R^l\) \((l=2,3)\) and \(k:\mathbb R^l\times \mathbb R^l\to \mathbb R\) is a kernel function that has a singularity at \(x=y\) and is analytic elsewhere. In this paper, the author proposes a new method for the fast evaluation of integrals stemming from boundary element methods including discretisations of the classical single and double layer potential operators. The key idea behind this new approach is the investigation of functions of the type \[ f(a):=\displaystyle\int_{\sigma(a)}\int_{\tau(a)}\varphi(x)k(x,y)\psi(y)dydx, \] where \(a\in \mathbb R^d\) is a parameter tuple describing the domains \(\sigma(a)\) and \(\tau(a)\). The function \(f\) is smooth in a \(d\)-dimensional box. A standard interpolation of \(f\) by polynomials leads to a \(d\)-dimensional tensor which is given by the values of \(f\) at the interpolation points. This tensor may be approximated in a low rank tensor format like the (CP) format or the \({\mathcal H}\)-Tucker format. The tensor approximation has to be done only once and allows us to evaluate interpolants in \({\mathcal O}(dr(m+1))\) operations in the (CP) format, or \({\mathcal O}(dk^3+dk(m+1))\) operations in the \({\mathcal H}\)-Tucker format, where \(m\) denotes the interpolation order and the ranks \(r,k\) are small integers. The author demonstrates that highly accurate integral values can be obtained at very moderate costs. Finally, the potential of the new approach is illustrated by a number of numerical examples.
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    boundary element methods
    0 references
    interpolation
    0 references
    tensor
    0 references
    evaluation of integral
    0 references
    numerical examples
    0 references
    0 references