Geodesic
Estimating the error in the classical Runge–Kutta methods
Žurnal vyčislitelʹnoj matematiki i matematičeskoj fiziki, Tome 54 (2014) no. 5, pp. 746-754 
S. I. Khashin. Estimating the error in the classical Runge–Kutta methods. Žurnal vyčislitelʹnoj matematiki i matematičeskoj fiziki, Tome 54 (2014) no. 5, pp. 746-754. http://geodesic.mathdoc.fr/item/ZVMMF_2014_54_5_a3/
@article{ZVMMF_2014_54_5_a3,
     author = {S. I. Khashin},
     title = {Estimating the error in the classical {Runge{\textendash}Kutta} methods},
     journal = {\v{Z}urnal vy\v{c}islitelʹnoj matematiki i matemati\v{c}eskoj fiziki},
     pages = {746--754},
     year = {2014},
     volume = {54},
     number = {5},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/ZVMMF_2014_54_5_a3/}
}
TY  - JOUR
AU  - S. I. Khashin
TI  - Estimating the error in the classical Runge–Kutta methods
JO  - Žurnal vyčislitelʹnoj matematiki i matematičeskoj fiziki
PY  - 2014
SP  - 746
EP  - 754
VL  - 54
IS  - 5
UR  - http://geodesic.mathdoc.fr/item/ZVMMF_2014_54_5_a3/
LA  - ru
ID  - ZVMMF_2014_54_5_a3
ER  - 
%0 Journal Article
%A S. I. Khashin
%T Estimating the error in the classical Runge–Kutta methods
%J Žurnal vyčislitelʹnoj matematiki i matematičeskoj fiziki
%D 2014
%P 746-754
%V 54
%N 5
%U http://geodesic.mathdoc.fr/item/ZVMMF_2014_54_5_a3/
%G ru
%F ZVMMF_2014_54_5_a3

Voir la notice de l'article provenant de la source Math-Net.Ru

It is well known that it is impossible to construct embedded firth-order methods for estimating the error in four-stage Runge–Kutta methods of order four. In this paper, a technique for error estimating with no additional calculations of the right-hand sides of equations is proposed. The proposed estimate is of fifth order and is based on the data provided by three successive steps of the method. The main results of the paper are formulas for evaluating the local error based on two and three steps of the method, respectively. The main conclusion of the paper is that an automatic stepsize control should not necessarily be based on embedded methods. Such a control can be implemented for an arbitrary method.

[1] Butcher J. C., Numerical methods for ordinary differential equations, 2nd ed., John Wiley, New York, 2008 | MR

[2] Wanner G., Hairer E., Nørsett S. P., Solving ordinary differential equations, v. I, Nonstiff Problems, 2Ed., Springer, Berlin, 2000

[3] Jackiewicz Z., Verner J. H., “Derivation and implementation of two-step Runge–Kutta pairs”, Appl. Math., 19 (2002), 227–248 | MR

[4] Verner J. H., “Numerically optimal Runge–Kutta pairs with interpolants”, Numerical Algorith., 53 (2010), 383–396 | DOI | MR

[5] Jackiewicz Z., General Linear Methods for Ordinary Differential Equations, Wiley, New York, 2009 | MR

[6] Chan T. M. H., Butcher J. C., “Multi-step zero approximations for stepsize control”, Appl. Numer. Math., 34 (2000), 167–177 | DOI | MR

[7] Shampine L. F., Watts H. A., “Comparing Error Estimators for Runge–Kutta methods”, Math. Comp., 25 (1971), 445–455 | DOI | MR