Reduced Sum Implementation of the BURA Method for Spectral Fractional Diffusion Problems

05/19/2021 โˆ™ by Stanislav Harizanov, et al. โˆ™ 0 โˆ™

The numerical solution of spectral fractional diffusion problems in the form ๐’œ^ฮฑ u = f is studied, where ๐’œ is a selfadjoint elliptic operator in a bounded domain ฮฉโŠ‚โ„^d, and ฮฑโˆˆ (0,1]. The finite difference approximation of the problem leads to the system ๐”ธ^ฮฑ๐ฎ = ๐Ÿ, where ๐”ธ is a sparse, symmetric and positive definite (SPD) matrix, and ๐”ธ^ฮฑ is defined by its spectral decomposition. In the case of finite element approximation, ๐”ธ is SPD with respect to the dot product associated with the mass matrix. The BURA method is introduced by the best uniform rational approximation of degree k of t^ฮฑ in [0,1], denoted by r_ฮฑ,k. Then the approximation u_kโ‰ˆ u has the form u_k = c_0 ๐Ÿ +โˆ‘_i=1^k c_i(๐”ธ - d_i ๐•€)^-1๐Ÿ, d_i<0, thus requiring the solving of k auxiliary linear systems with sparse SPD matrices. The BURA method has almost optimal computational complexity, assuming that an optimal PCG iterative solution method is applied to the involved auxiliary linear systems. The presented analysis shows that the absolute values of first extremely large. In such a case the condition number of ๐”ธ - d_i ๐•€ is practically equal to one. Obviously, such systems do not need preconditioning. The next question is if we can replace their solution by directly multiplying ๐Ÿ with -c_i/d_i. Comparative analysis of numerical results is presented as a proof-of-concept for the proposed RS-BURA method.



There are no comments yet.


page 1

page 2

page 3

page 4

This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.