On modified Runge-Kutta trees and methods

被引:154
|
作者
Tsitouras, Ch [2 ]
Famelis, I. Th [1 ]
Simos, T. E. [3 ,4 ]
机构
[1] TEI Athens, Dept Math, GR-12210 Egaleo, Greece
[2] TEI Chalkis, GR-34400 Psahna, Greece
[3] King Saud Univ, Coll Sci, Dept Math, Riyadh 11451, Saudi Arabia
[4] Univ Peloponnese, Dept Comp Sci & Technol, Sci Computat Lab, GR-22100 Tripolis, Greece
关键词
Rooted trees; Integer partitions; Truncation error; Runge-Kutta methods; Initial value problems; Mathematica; PAIRS;
D O I
10.1016/j.camwa.2011.06.058
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
Modified Runge-Kutta (mRK) methods can have interesting properties as their coefficients may depend on the step length. By a simple perturbation of very few coefficients we may produce various function-fitted methods and avoid the overload of evaluating all the coefficients in every step. It is known that, for Runge-Kutta methods, each order condition corresponds to a rooted tree. When we expand this theory to the case of mRK methods, some of the rooted trees produce additional trees, called mRK rooted trees, and so additional conditions of order. In this work we present the relative theory including a theorem for the generating function of these additional mRK trees and explain the procedure to determine the extra algebraic equations of condition generated for a major subcategory of these methods. Moreover, efficient symbolic codes are provided for the enumeration of the trees and the generation of the additional order conditions. Finally, phase-lag and phase-fitted properties are analyzed for this case and specific phase-fitted pairs of orders 8(6) and 6(5) are presented and tested. (C) 2011 Elsevier Ltd. All rights reserved.
引用
收藏
页码:2101 / 2111
页数:11
相关论文
共 50 条
  • [31] On the Preservation of Lyapunov Functions by Runge-Kutta Methods
    Calvo, M.
    Laburta, M. P.
    Montijano, J. I.
    Randez, L.
    NUMERICAL ANALYSIS AND APPLIED MATHEMATICS, VOLS 1 AND 2, 2009, 1168 : 735 - 738
  • [32] Effective Order of Partitioned Runge-Kutta Methods
    Ahmad, Junaid
    Habib, Yousaf
    Shafiq, Saba
    Rehman, Shafiq Ur
    BULLETIN OF THE IRANIAN MATHEMATICAL SOCIETY, 2020, 46 (01) : 79 - 98
  • [33] Dissipativity of Runge-Kutta methods in Hilbert spaces
    Adrian T. Hill
    BIT Numerical Mathematics, 1997, 37 : 37 - 42
  • [34] Parallelization of Runge-Kutta Methods for Hardware Implementation
    Fedoseev, Petr
    Zhukov, Konstantin
    Kaplun, Dmitry
    Vybornov, Nikita
    Andreev, Valery
    COMPUTATION, 2022, 10 (12)
  • [35] On the preservation of invariants by explicit Runge-Kutta methods
    Calvo, M.
    Hernandez-Abreu, D.
    Montijano, J. I.
    Randez, L.
    SIAM JOURNAL ON SCIENTIFIC COMPUTING, 2006, 28 (03): : 868 - 885
  • [36] ENERGY-PRESERVING RUNGE-KUTTA METHODS
    Celledoni, Elena
    McLachlan, Robert I.
    McLaren, David I.
    Owren, Brynjulf
    Quispel, G. Reinout W.
    Wright, William M.
    ESAIM-MATHEMATICAL MODELLING AND NUMERICAL ANALYSIS-MODELISATION MATHEMATIQUE ET ANALYSE NUMERIQUE, 2009, 43 (04): : 645 - 649
  • [37] Runge-Kutta research at Toronto
    Hull, TE
    Enright, WH
    Jackson, KR
    APPLIED NUMERICAL MATHEMATICS, 1996, 22 (1-3) : 225 - 236
  • [38] Constructing Runge-Kutta methods with the use of artificial neural networks
    Anastassi, Angelos A.
    NEURAL COMPUTING & APPLICATIONS, 2014, 25 (01): : 229 - 236
  • [39] A modified Runge-Kutta method for increasing stability properties
    Hu, Guang-Da
    Wang, Zheng
    JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS, 2024, 441
  • [40] Construction of Exponentially Fitted Symplectic Runge-Kutta-Nystrom Methods from Partitioned Runge-Kutta Methods
    Monovasilis, T.
    Kalogiratou, Z.
    Simos, T. E.
    MEDITERRANEAN JOURNAL OF MATHEMATICS, 2016, 13 (04) : 2271 - 2285