Article Contents
Article Contents

# Multi-point Taylor series to solve differential equations

• * Corresponding author: Zézé
• The use of Taylor series is an effective numerical method to solve ordinary differential equations but this fails when the sought function is not analytic or when it has singularities close to the domain. These drawbacks can be partially removed by considering multi-point Taylor series, but up to now there are only few applications of the latter method in the literature and not for problems with very localized solutions. In this respect, a new numerical procedure is presented that works for an arbitrary cloud of expansion points and it is assessed from several numerical experiments.

Mathematics Subject Classification: Primary: 58F15, 58F17; Secondary: 53C35.

 Citation:

• Figure 1.  Discretization points in the case of three subdomains and two point-Taylor series. The expansion points are $x_i$ and the additional collocation points are $y_j$ (2 per subdomain)

Figure 2.  Example 1: $f(x) = g(x) = 1, L = 10$. Comparison of a one-point Taylor series, one-point Taylor series in two subdomains and two-point Taylor series

Figure 3.  Example 1, 2-point Taylor series (n = 2). Convergence with the degree p

Figure 4.  Example 1, degree p = 4. Convergence with the number expansion points

Figure 5.  Example 1, 2-point Taylor series, c $\pm \frac{10}{3}$, p = 6. Distribution of the residual and of the error in the interval

Figure 6.  Example 1, 2-point Taylor series, p = 6. Maximal value of the residual and of the error according to the location of the expansion points

Figure 7.  Example 2, 4-point Taylor series, expansion points located in ($\pm 0.8; \pm 1.8$). Convergence with the degree

Figure 8.  Example 3, Solution with 3 subdomains, Taylor 4-point and $p = 5$, $(L = 10, a = 1)$, $u_{max} = 4.718$

Figure 9.  Example 3, Solution with 3 subdomains, Taylor 4-point and $p = 5$, $(L = 10, a = 0.1)$, $u_{max} = 5.0788$

Figure 10.  Example 3, Solution with 3 subdomains, Taylor 4-point and $p = 4$, $(L = 10, a = 0.1)$, $u_{max} = 5.109$

Figure 11.  Example 3, residual with 3 subdomains, $p = 5$, $n = 4$. Case of a smooth distribution of points

•  W. Aggoune , H. Zahrouni  and  M. Potier-Ferry , Asymptotic numerical methods for unilateral contact, International Journal for Numerical Methods in Engineering, 68 (2006) , 605-631.  doi: 10.1002/nme.1714. H. Ben Dhia , Multiscale mechanical problems: The arlequin method, Comptes Rendus de l'Academie des Sciences Series IIB Mechanics Physics Astronomy, 12326 (1998) , 899-904. C. Chester , B. Friedman  and  F. Ursell , An extension of the method of steepest descents, Mathematical Proceedings of the Cambridge Philosophical Society, 53 (1957) , 599-611.  doi: 10.1017/S0305004100032655. G. Corliss  and  Y. F. Chang , Solving ordinary differential equations using taylor series, ACM Transact Math Software, 8 (1982) , 114-144.  doi: 10.1145/355993.355995. F. Costabile  and  A. Napoli , Solving bvps using two-point taylor formula by a symbolic software, Journal of Computational and Applied Mathematics, 210 (2007) , 136-148.  doi: 10.1016/j.cam.2006.10.081. M. Gasca  and  T. Sauer , Polynomial interpolation in several variables, Computational Mathematics, 12 (2000) , 377-410.  doi: 10.1023/A:1018981505752. M. Giesbrecht , G. Labahn  and  W. S. Lee , Symbolic-numeric sparse interpolation of multivariate polynomials, Journal of Symbolic Computation, 44 (2009) , 943-959.  doi: 10.1016/j.jsc.2008.11.003. U. Haussler-Combe  and  C. Korn , An adaptive approach with the element-free-galerkin method, Computer Methods in Applied Mechanics and Engineering, 162 (1998) , 203-222.  doi: 10.1016/S0045-7825(97)00344-7. E. J. Kansa , Multiquadrics - a scattered data approximation scheme with applications to computational fluid-dynamics-ii solutions to parabolic, hyperbolic and elliptic partial differential equations, Computers & Mathematics with Applications, 19 (1990) , 147-161.  doi: 10.1016/0898-1221(90)90271-K. J. L. Lopez  and  N. M. Temme , New series expansions of the gauss hypergeometric function, Advances in Computational Mathematics, 39 (2013) , 349-365.  doi: 10.1007/s10444-012-9283-y. J. L. Lopez  and  E. Perez Sinusia , New series expansions for the confluent hypergeometric function m (a, b, z), Applied Mathematics and Computation, 235 (2014) , 26-31.  doi: 10.1016/j.amc.2014.02.099. J. L. Lopez , P. Pagola  and  E. Perez Sinusia , New series expansions of the 3f2 function, Journal of Mathematical Analysis and Applications, 421 (2015) , 982-995.  doi: 10.1016/j.jmaa.2014.07.065. J. L. Lopez  and  N. M. Temme , Two-point taylor expansions of analytic functions, Studies in Applied Mathematics, 109 (2002) , 297-311.  doi: 10.1111/1467-9590.00225. J. L. Lopez  and  N. M. Temme , Multi-point taylor expansions of analytic functions, Transactions of the American Mathematical Society, 356 (2004) , 4323-4342.  doi: 10.1090/S0002-9947-04-03619-0. J. L. Lopez , E. Perez Sinusia  and  N. M. Temme , Multi-point taylor approximations in one- dimensional linear boundary value problems, Applied Mathematics and Computation, 207 (2009) , 519-527.  doi: 10.1016/j.amc.2008.11.015. J. L. Lopez  and  E. Perez Sinusia , Two-point taylor approximations of the solutions of two-dimensional boundary value problems, Applied Mathematics and Computation, 218 (2012) , 9107-9115.  doi: 10.1016/j.amc.2012.02.060. V. P. Nguyen , T. Rabczuk , S. P. A. Bordas  and  M. Duflot , Meshless methods: A review and computer implementation aspects, Mathematics and Computers in Simulation, 79 (2008) , 763-813.  doi: 10.1016/j.matcom.2008.01.003. A. Quarteroni, Méthodes Numériques Pour le Calcul Scientifique. Programmes en Matlab, Collection IRIS, 2000. J. N. Reddy, Mechanics of laminated composite plates and shells: theory and analysis, CRC press, Boca Raton, 2004. P. Rentrop , A taylor series method for the numerical solution of two-point boundary value problems, Numerische Mathematik, 31 (1978) , 359-375.  doi: 10.1007/BF01404566. Y. Tampango , M. Potier-Ferry , Y. Koutsawa  and  S. Belouettar , Convergence analysis and detection of singularities within a boundary meshless method based on taylor series, Engineering Analysis with Boundary Elements, 36 (2012) , 1465-1472.  doi: 10.1016/j.enganabound.2012.03.014. Y. Tampango , M. Potier-Ferry , Y. Koutsawa  and  S. Tiem , Coupling of polynomial approximations with application to a boundary meshless method, International Journal for Numerical Methods in Engineering, 95 (2013) , 1094-1112.  doi: 10.1002/nme.4549. J. Yang , H. Hu  and  M. Potier-Ferry , Solving large scale problems by Taylor Meshless method, International Journal for Numerical Methods in Engineering, 112 (2017) , 103-124.  doi: 10.1002/nme.5508. D. S. Zeze , M. Potier-Ferry  and  N. Damil , A boundary meshless method with shape functions computed from the pde, Engineering Analysis with Boundary Elements, 34 (2010) , 747-754.  doi: 10.1016/j.enganabound.2010.03.008.

Figures(11)