2002
DOI: 10.1007/bf03167454
|View full text |Cite
|
Sign up to set email alerts
|

Derivation and implementation of Two-Step Runge-Kutta pairs

Abstract: Explicit Runge-Kutta pairs are known to provide efficient solutions to initial value differential equations with inexpensive derivative evaluations. Two-step Runge-Kutta methods strive to improve the efficiency by utilizing approximations to the solution and its derivatives from the previous step. This article suggests a strategy for computing embedded pairs of such two-step methods using a smaller number of function evaluations than that required for traditional Runge-Kutta methods of the same order. This lea… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
21
1

Year Published

2009
2009
2016
2016

Publication Types

Select...
5
1

Relationship

2
4

Authors

Journals

citations
Cited by 25 publications
(22 citation statements)
references
References 24 publications
0
21
1
Order By: Relevance
“…This suggests that these two problems selected may not adequately challenge the variable step format. Even so, these results show considerably more efficiency for these two problems than corresponding results reported in [13] for two-step Runge-Kutta (TSRK) pairs of the same orders. This suggests that more study of TSRK pairs is needed if such general linear pairs are to compete successfully with RK pairs.…”
Section: Numerical Experimentscontrasting
confidence: 59%
See 1 more Smart Citation
“…This suggests that these two problems selected may not adequately challenge the variable step format. Even so, these results show considerably more efficiency for these two problems than corresponding results reported in [13] for two-step Runge-Kutta (TSRK) pairs of the same orders. This suggests that more study of TSRK pairs is needed if such general linear pairs are to compete successfully with RK pairs.…”
Section: Numerical Experimentscontrasting
confidence: 59%
“…The derivation of other special types of general linear methods [1] such as two-step Runge-Kutta pairs [13] remains an area of interest, and some of these may be found that are competitive with currently implemented pairs. Accordingly, the testing provided here forms a basis on which to decide whether such algorithms are likely to lead to more effective codes for solving initial value problems of type (1).…”
Section: Discussionmentioning
confidence: 99%
“…In the Python code, we use the built-in Adams/BDF multistep method with automatic step size control provided by SciPy, which is based on the Netlib VODE implementation [98]. We have performed tests with other Runge-Kutta-based integration schemes [99], but the results did not change, while the runtime of the code increased significantly. This is expected as the ordinary differential equation near the BH becomes stiff, which is a situation not handled well by Runge-Kutta-based integrators.…”
Section: Appendix D: Pyholementioning
confidence: 99%
“…TSRK methods were introduced by Jackiewicz and Tracogna [18] and further investigated in [1,3,8,9,15,17,20,25], Conte et al (unpublished manuscript) and [26]. Continuous methods (1.1) provide an approximation to the solution y(t) of (1.2) on the whole interval of integration, and not only in the gridpoints {t n } as in the case of discrete TSRK methods (1.3).…”
Section: Introductionmentioning
confidence: 98%