2016
DOI: 10.1134/s0965542516080030
|View full text |Cite
|
Sign up to set email alerts
|

Multiple solution of systems of linear algebraic equations by an iterative method with the adaptive recalculation of the preconditioner

Abstract: The mean time needed to solve a series of systems of linear algebraic equations (SLAEs) as a function of the number of SLAEs is investigated. It is proved that this function has an extremum point. An algorithm for adaptively determining the time when the preconditioner matrix should be recalculated when a series of SLAEs is solved is developed. A numerical experiment with multiply solving a series of SLAEs using the proposed algorithm for computing 100 capacitance matrices with two different structures-microst… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

0
4
0

Year Published

2016
2016
2019
2019

Publication Types

Select...
2
2

Relationship

1
3

Authors

Journals

citations
Cited by 4 publications
(4 citation statements)
references
References 3 publications
0
4
0
Order By: Relevance
“…We use the second approach. For this purpose we developed methods of adaptive recomputing preconditioner based on the threshold of the number of iterations [33], the average arithmetic complexity [34], and the average solution time [35]. Recomputation of the preconditioner is not necessary if the seed matrix for calculating the preconditioner has been selected properly.…”
Section: Structures and Approach For Investigationmentioning
confidence: 99%
“…We use the second approach. For this purpose we developed methods of adaptive recomputing preconditioner based on the threshold of the number of iterations [33], the average arithmetic complexity [34], and the average solution time [35]. Recomputation of the preconditioner is not necessary if the seed matrix for calculating the preconditioner has been selected properly.…”
Section: Structures and Approach For Investigationmentioning
confidence: 99%
“…For long 2D-structures a quasi-static approach is often relevant as it allows to obtain fast and accurate results. However, even for this approach the decrease of simulation time is important and can be achieved by various methods, for example by implementation of new algorithms [11]. The calculation of matrices of per unit length parameters of the structure is the essential stage of the analysis.…”
Section: Introductionmentioning
confidence: 99%
“…However, it is a priori impossible to determine when to recompute matrix M. Thus, the search for an a priori condition of the recomputation is relevant. Paper [13] is devoted to one of the conditions. Other conditions proposed by the authors have been investigated in paper [14], so they are omitted in this paper.…”
Section: Approaches To Solving the Sequence Of Linear Systemsmentioning
confidence: 99%
“…However, the effectiveness of the preconditioner decreases, as the difference between the first and the current matrix increases. To solve this problem, it was suggested to recompute matrix M when the convergence rate of solving the current linear system is too slow [13]. Existence of optimal threshold value was shown.…”
Section: Approaches To Solving the Sequence Of Linear Systemsmentioning
confidence: 99%