Skip to main content
Log in

A hybrid numerical scheme for singular perturbation delay problems with integral boundary condition

  • Original Research
  • Published:
Journal of Applied Mathematics and Computing Aims and scope Submit manuscript

Abstract

In this article, a singular perturbation delay problem of convection–diffusion (C–D) type having an integral boundary condition is considered. The analytical solution of the considered problem has a weak interior layer in addition to the boundary layer at the right end of the domain. Some a priori estimates are given on the exact solution which are useful for the error analysis. The numerical approximation is composed of a hybrid finite difference scheme on a generalized Shishkin mesh. For the proposed scheme, almost second order \(\varepsilon \)-uniform convergence is established. Numerical experiments are conducted to corroborate the theoretical results. A comparison with the existing scheme (J Appl Math Comput 63:813–828) is also performed.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2

Similar content being viewed by others

References

  1. Almomani, R., Almefleh, H.: On heat conduction problem with integral boundary condition. J. Emerg. Trends Eng. Appl. Sci. 3, 977–979 (2012)

    Google Scholar 

  2. Bahuguna, D., Abbas, S., Dabas, J.: Partial functional differential equation with an integral boundary condition and applications to population dynamics. Nonlinear Anal. 69, 2623–2635 (2008)

    MathSciNet  MATH  Google Scholar 

  3. Cannon, J.R.: The solution of the heat equation subject to the specification of energy. Q. Appl. Math. 21, 155–160 (1963)

    MathSciNet  MATH  Google Scholar 

  4. Choi, Y.S., Chan, K.Y.: A parabolic equation with nonlocal boundary conditions arising from electrochemistry. Nonlinear Anal. Theory Methods Appl. 18(4), 317–331 (1992)

    MathSciNet  MATH  Google Scholar 

  5. Cahlon, B., Kulkarni, D.M., Shi, P.: Stepwise stability for the heat equation with a nonlocal constraint. SIAM J. Numer. Anal. 32, 571–593 (1995)

    MathSciNet  MATH  Google Scholar 

  6. Formaggia, L., Nobil, F., Quarteroni, A., Venezian, A.: Multiscale modelling of the circulatory system: a preliminary analysis. Comput. Vis. Sci. 2, 75–83 (1999)

    Google Scholar 

  7. Jankowskii, T.: Differential equations with integral boundary conditions. J. Comput. Appl. Math. 147, 1–8 (2002)

    MathSciNet  Google Scholar 

  8. Hu, M., Wang, L.: Triple positive solutions for an impulsive dynamic equation with integral boundary condition on time scales. Int. J. Appl. Math. Stat. 31, 43–66 (2013)

    MathSciNet  Google Scholar 

  9. Turkyilmazoglu, M.: Parabolic partial differential equations with nonlocal initial and boundary values. Int. J. Comput. Methods 12(5), 1550024 (2015)

    MathSciNet  MATH  Google Scholar 

  10. Ashyralyev, A., Sharifov, Y.A.: Existence and uniqueness of solutions for nonlinear impulsive differential equations with two-point and integral boundary conditions. Adv. Differ. Equ. 173, 1–11 (2013)

    MathSciNet  MATH  Google Scholar 

  11. Bahuguna, D., Dabas, J.: Existence and uniqueness of a solution to a semilinear partial delay differential equation with an integral condition. Nonlinear Dyn. Syst. Theory 8, 7–19 (2008)

    MathSciNet  MATH  Google Scholar 

  12. Boucherif, A.: Second order boundary value problems with integral boundary condition. Nonlinear Anal. 70, 368–379 (2009)

    MathSciNet  MATH  Google Scholar 

  13. Li, H., Sun, F.: Existence of solutions for integral boundary value problems of second order ordinary differential equations. Bound. Value Probl. 2012(1), 147 (2012)

    MathSciNet  MATH  Google Scholar 

  14. Elsgolts, E.L.: Qualitative Methods in Mathematical Analysis in: Translations of Mathematical Monographs, vol. 12. American Mathematical Society, Providence (1964)

    Google Scholar 

  15. Culshaw, R.V., Ruan, S.: A delay differential equation model of HIV infection of \(CD4^{+}\) T-cells. Math. Biosci. 165, 27–39 (2000)

    MATH  Google Scholar 

  16. Glizer, V.Y.: Asymptotic analysis and solution of a finite-horizon \(H_\infty \) control problem for singularly perturbed linear systems with small state delay. J. Optim. Theory Appl. 117, 295–325 (2003)

    MathSciNet  MATH  Google Scholar 

  17. Longtin, A., Milton, J.: Complex oscillations in the human pupil light reflex with mixed and delayed feedback. Math. Biosci. 90, 183–199 (1988)

    MathSciNet  Google Scholar 

  18. Doolan, E.P., Miller, J.J.H., Schilders, W.H.A.: Uniform Numerical Methods for Problems with Initial and Boundary Layers. Boole Press, Dublin (1980)

    MATH  Google Scholar 

  19. Farrell, P.A., Hegarty, A.F., Miller, J.J.H., O’Riordan, E., Shishkin, G.I.: Robust Computational Techniques for Boundary Layers. Chapman-Hall, Boca Raton (2000)

  20. Miller, J.J.H., O’Riordan, E., Shishkin, G.I.: Fitted Numerical Methods for Singular Perturbation Problems. World Scientific Publishing Co., Singapore (1996)

  21. Roos, H.G., Stynes, M., Tobiska, L.: Numerical Methods for Singularly Perturbed Differential Equations, Convection–Diffusion and Flow Problems. Springer, Berlin (1996)

    MATH  Google Scholar 

  22. Kadalbajoo, M.K., Patidar, K.C.: Parameter-uniform fitted mesh method for singularly perturbed delay differential equations with layer behavior. Electron. Trans. Numer. Anal. 23, 180–201 (2006)

    MathSciNet  MATH  Google Scholar 

  23. Kadalbajoo, M.K., Sharma, K.K.: Numerical treatment of boundary value problems for second order singularly perturbed delay differential equations. Comput. Appl. Math. 24, 151–172 (2005)

    MathSciNet  MATH  Google Scholar 

  24. Lange, C.G., Miura, R.M.: Singular perturbation analysis of boundary value problems for differential-difference equations. SIAM J. Appl. Math. 42(3), 502–530 (1982)

    MathSciNet  MATH  Google Scholar 

  25. Nicaise, S., Xenophontos, C.: Robust approximation of singularly perturbed delay differential equations by the hp finite element method. Comput. Methods Appl. Math. 13, 21–37 (2013)

    MathSciNet  MATH  Google Scholar 

  26. Zarin, H.: On discontinuous Galerkin finite element method for singularly perturbed delay differential equations. Appl. Math. Lett. 38, 27–32 (2014)

    MathSciNet  MATH  Google Scholar 

  27. Turkyilmazoglu, M.: Analytic approximate solutions of parameterized unperturbed and singularly perturbed boundary value problems. Appl. Math. Model. 35, 3879–3886 (2011)

    MathSciNet  MATH  Google Scholar 

  28. Amiraliyev, G.M., Amiraliyev, I.G., Kuddu, M.: A numerical treatment for singularly perturbed differential equations with integral boundary condition. Appl. Math. Comput. 185, 574–582 (2007)

    MathSciNet  MATH  Google Scholar 

  29. Cakir, M., Amiraliyev, G.M.: A finite difference method for the singularly perturbed problem with nonlocal boundary condition. Appl. Math. Comput. 160, 539–549 (2005)

    MathSciNet  MATH  Google Scholar 

  30. Cen, Z., Cai, X.: A second order upwind difference scheme for a singularly perturbed problem with integral boundary condition in neutral network. In: 11th International Conference, KES 2007. XVII Italian Workshop on Neural Networks, Vietri sul Mare, Italy, Proceedings, vol. 2, pp. 175–181 (2007)

  31. Kudu, M., Amiraliyev, G.: Finite difference method for a singularly perturbed differential equations with integral boundary condition. Int. J. Math. Comput. 26, 72–79 (2015)

    MathSciNet  Google Scholar 

  32. Kumar, D., Kumari, P.: A parameter-uniform collocation scheme for singularly perturbed delay problems with integral boundary condition. J. Appl. Math. Comput. 63, 813–828 (2020)

    MathSciNet  MATH  Google Scholar 

  33. Sekar, E., Tamilselvan, A.: Singularly perturbed delay differential equations of convection–diffusion type with integral boundary condition. J. Appl. Math. Comput. 59(1), 701–722 (2020)

    MathSciNet  MATH  Google Scholar 

  34. Farrell, P.A., Hegarty, A.F., Miller, J.J.H., O’Riordan, E., Shishkin, G.I.: Singularly perturbed convection–diffusion problems with boundary and weak interior layers. J. Comput. Appl. Math. 166(1), 133–151 (2004)

  35. Mukherjee, K., Natesan, S.: Parameter-uniform hybrid numerical scheme for time-dependent convection-dominated initial-boundary-value problems. Computing 84(3–4), 209–230 (2009)

    MathSciNet  MATH  Google Scholar 

  36. Mukherjee, K., Natesan, S.: \(\varepsilon \)-uniform error estimate of hybrid numerical scheme for singularly perturbed parabolic problems with interior layers. Numer. Algorithms 58(1), 103–141 (2011)

    MathSciNet  MATH  Google Scholar 

  37. Stynes, M., Roos, H.G.: The midpoint upwind scheme. Appl. Numer. Math. 23(3), 361–374 (1997)

    MathSciNet  MATH  Google Scholar 

  38. Majumdar, A., Natesan, S.: An \(\varepsilon \)-uniform hybrid scheme for a singularly perturbed degenerate convection–diffusion problem. Int. J. Comput. Math. 96(7), 1313–1334 (2018)

    MathSciNet  MATH  Google Scholar 

  39. Gupta, V., Kumar, M., Kumar, S.: Higher order numerical approximation for time dependent singularly perturbed differential-difference convection–diffusion equations. Numer. Methods Partial Differ. Equ. 34, 357–380 (2018)

    MathSciNet  MATH  Google Scholar 

  40. Majumdar, A., Natesan, S.: A higher-order hybrid numerical scheme for singularly perturbed convection-diffusion problem with boundary and weak interior layers. Int. J. Math. Model. Numer. Optim. 10(1), 68–101 (2020)

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Pratima Rai.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendices

Appendix

Some properties of the barrier function \(\phi _{i}(m)\).

Lemma A.1

For \(0< m < r^{*}/2\), the following inequalities hold

$$\begin{aligned} {\mathcal {L}}_\varepsilon ^{M}\phi _{i}(m) \ge {\left\{ \begin{array}{ll} \dfrac{C}{\varepsilon + m h_i}\phi _{i}(m), &{} i=1, \ldots , M/4, \, M/2+1, \ldots , 3M/4, \\ \dfrac{C}{\varepsilon }\phi _{i}(m), &{} i=M/4+1, \ldots , M/2, \, 3M/4+1, \ldots , M-1. \end{array}\right. } \end{aligned}$$

Proof

Using the barrier function defined in Eq. (4.3), we obtain

$$\begin{aligned} \phi _{i+1}(m)-\phi _{i}(m) = \dfrac{m h_{i+1}}{\varepsilon + m h_{i+1}}\phi _{i+1}(m). \end{aligned}$$

We now apply the operator \({\mathcal {L}}_\varepsilon ^M\) on \(\phi _{i}(m)\) for \( i=1,\ldots ,M/4\) to obtain

$$\begin{aligned} {\mathcal {L}}_\varepsilon ^{M}\phi _{i}(m)&=-\varepsilon \delta ^2 \phi _{i}(m) + r_{i-1/2} D^{-} \phi _{i}(m) + s_{i-1/2} \phi _{i-1/2}(m) \\&=\bigg [ \dfrac{m}{\varepsilon + m h_{i}}\bigg (r_{i-1/2}-\dfrac{2mh_{i}}{h_{i}+h_{i+1}}\bigg ) + \dfrac{s_{i-1/2}}{2}\bigg (\dfrac{\varepsilon }{\varepsilon +mh_{i}} + 1\bigg )\bigg ]\phi _{i}(m) \\&\ge \dfrac{C}{\varepsilon +mh_{i}} \phi _{i}(m). \end{aligned}$$

Again, for \( i=M/4+1,\ldots ,M/2\), we have

$$\begin{aligned} {\mathcal {L}}_\varepsilon ^{M}\phi _{i}(m)&=-\varepsilon \delta ^2\phi _{i}(m) + r_{i} D^{0}\phi _{i}(m) + s_{i} \phi _{i}(m) \\&=\bigg [ \dfrac{m}{h_{i}+h_{i+1}}\bigg \{ \dfrac{h_{i}}{\varepsilon +mh_{i}}(r_i - 2m)+ \dfrac{h_{i+1} r_i}{\varepsilon }\bigg \} + s_{i}\bigg ]\phi _{i}(m) \\&\ge \dfrac{C}{\varepsilon } \phi _{i}(m) . \end{aligned}$$

It is easy to obtain that

$$\begin{aligned} \phi _{i-M/2}(m)=\bigg (\dfrac{\varepsilon }{\varepsilon + m {\tilde{h}}_1}\bigg )^{M/4}\bigg (\dfrac{\varepsilon }{\varepsilon + m {\tilde{h}}_2}\bigg )^{M/4}\phi _{i}(m), \,\, for \,\, i=M/2+1, \ldots , M-1. \end{aligned}$$

Now, applying the discrete operator \({\mathcal {L}}_\varepsilon ^M\) on \(\phi _{i}(m)\) for \(i= M/2+1,\ldots ,3M/4\), we obtain

$$\begin{aligned}&{\mathcal {L}}_\varepsilon ^{M}\phi _{i}(m) \\&\quad =-\varepsilon \delta ^2 \phi _{i}(m) + r_{i-1/2} D^{-}\phi _{i}(m) + s_{i-1/2}\, \phi _{i-1/2}(m) + t_{i-1/2} \,\phi _{i-M/2-1/2}(m) \\&\quad =\Bigg [ \dfrac{m}{\varepsilon + m h_{i}}\bigg (r_{i-1/2}-\dfrac{2mh_{i}}{h_{i}+h_{i+1}}\bigg )+\dfrac{s_{i-1/2}}{2}\bigg (\dfrac{\varepsilon }{\varepsilon +mh_{i}} + 1\bigg )\\&\qquad \qquad \quad +\dfrac{t_{i-1/2}}{2}\bigg (\dfrac{\varepsilon }{\varepsilon +mh_{i}} + 1\bigg )\bigg (\dfrac{\varepsilon }{\varepsilon + m {\tilde{h}}_1}\bigg )^{M/4}\bigg (\dfrac{\varepsilon }{\varepsilon + m {\tilde{h}}_2}\bigg )^{M/4}\Bigg ]\phi _{i}(m) \\&\quad =\Bigg [ \dfrac{m}{\varepsilon + m h_{i}}\bigg (r_{i-1/2}-\dfrac{2mh_{i}}{h_{i}+h_{i+1}}\bigg )+\bigg (\dfrac{s_{i-1/2}+t_{i-1/2}}{2}\bigg )\bigg (\dfrac{\varepsilon }{\varepsilon +mh_{i}} + 1\bigg )\\&\qquad \quad \quad +\dfrac{t_{i-1/2}}{2}\bigg (\dfrac{\varepsilon }{\varepsilon +mh_{i}} + 1\bigg )\bigg \{\bigg (\dfrac{\varepsilon }{\varepsilon + m {\tilde{h}}_1}\bigg )^{M/4}\bigg (\dfrac{\varepsilon }{\varepsilon + m {\tilde{h}}_2}\bigg )^{M/4}-1\bigg \}\Bigg ]\phi _{i}(m) \\&\quad \ge \dfrac{C}{\varepsilon +mh_{i}} \phi _{i}(m) \quad (using \,\,\, m < r^{*}/2). \end{aligned}$$

Again, for \(i=3M/4+1,\ldots ,M-1\), we have

$$\begin{aligned}&{\mathcal {L}}_\varepsilon ^{M}\phi _{i}(m) =-\varepsilon \delta ^2\phi _{i}(m) + r_{i} D^{0}\phi _{i}(m) + s_{i} \phi _{i}(m) + t_{i} \phi _{i-M/2}(m) \\&\quad =\Bigg [ \dfrac{m}{h_{i}+h_{i+1}}\bigg \{ \dfrac{h_{i}}{\varepsilon +mh_{i}}(r_i - 2m)+ \dfrac{h_{i+1} r_i}{\varepsilon }\bigg \} + s_{i} \\&\qquad +t_{i}\bigg (\dfrac{\varepsilon }{\varepsilon + m {\tilde{h}}_1}\bigg )^{M/4}\bigg (\dfrac{\varepsilon }{\varepsilon + m {\tilde{h}}_2}\bigg )^{M/4}\Bigg ]\phi _{i}(m) \\&\quad =\Bigg [ \dfrac{m}{h_{i}+h_{i+1}}\bigg \{ \dfrac{h_{i}}{\varepsilon +mh_{i}}(r_i - 2m)+ \dfrac{h_{i+1} r_i}{\varepsilon }\bigg \} + s_{i}+t_{i} \\&\qquad +t_{i}\bigg \{\bigg (\dfrac{\varepsilon }{\varepsilon + m {\tilde{h}}_1}\bigg )^{M/4}\bigg (\dfrac{\varepsilon }{\varepsilon + m {\tilde{h}}_2}\bigg )^{M/4}-1\bigg \}\Bigg ]\phi _{i}(m) \\&\quad \ge \dfrac{C}{\varepsilon } \phi _{i}(m). \end{aligned}$$

\(\square \)

Lemma A.2

For each i and \(0<m<r^{*}/2\), the mesh function \(\phi _{i}(m)\) satisfies

$$\begin{aligned} \exp \bigg (\dfrac{-m(1-y_i)}{\varepsilon }\bigg )\le \dfrac{\phi _i (m)}{\phi _{M/2} (m)},\quad 1\le i \le M/2-1, \quad \,\, \end{aligned}$$
(A.1)

and

$$\begin{aligned} \exp \bigg (\dfrac{-m(2-y_i)}{\varepsilon }\bigg )\le \dfrac{\phi _i (m)}{\phi _{M} (m)}, \quad M/2+1\le i \le M-1. \end{aligned}$$
(A.2)

Proof

We have,

$$\begin{aligned} \mathrm {e}^{-mh_j/\varepsilon }= (\mathrm {e}^{mh_j/\varepsilon })^{-1}\le \bigg (1+\dfrac{m h_j}{\varepsilon }\bigg )^{-1} \,\,\text {for each}\,\, j=1, \ldots , M. \end{aligned}$$

Multiplying the above inequalities for \(j= i+1,\ldots , M/2\), we obtain

$$\begin{aligned} \mathrm {e}^{-m(h_{i+1}+\cdots +h_{M/2})/\varepsilon }\le \prod _{j=i+1}^{M/2} \bigg (1+\dfrac{m h_j}{\varepsilon }\bigg )^{-1}, \end{aligned}$$

or,

$$\begin{aligned} \mathrm {e}^{-m(1-y_i)/\varepsilon }\le \dfrac{\displaystyle \prod _{j=i+1}^{M}\bigg (1+\dfrac{m h_j}{\varepsilon }\bigg )^{-1}}{\displaystyle \prod _{j=M/2+1}^{M}\bigg (1+\dfrac{m h_j}{\varepsilon }\bigg )^{-1}}=\dfrac{\phi _i (m)}{\phi _{M/2} (m)}, \quad for \quad 1\le i \le M/2-1. \end{aligned}$$

\(\square \)

Lemma A.3

For \(0<m<r^{*}/2\), the following inequalities hold true

$$\begin{aligned} \dfrac{\phi _{i} (m)}{\phi _{M/2} (m)}\le C L^{2m \tau _{0}(1-2i/M)} M^{-2m \tau _{0}(1-2i/M)}, \quad M/4\le i \le M/2, \end{aligned}$$

and

$$\begin{aligned} \dfrac{\phi _{i} (m)}{\phi _{M} (m)}\le C L^{2m \tau _{0}(2-2i/M)} M^{-2m \tau _{0}(2-2i/M)}, \quad 3M/4\le i \le M-1. \end{aligned}$$

Proof

For \(M/4 \le i \le M/2\), we have

$$\begin{aligned} \dfrac{\phi _{i} (m)}{\phi _{M/2} (m)}&=\prod _{j=i+1}^{M/2}\bigg (1+\dfrac{m h_j}{\varepsilon }\bigg )^{-1}\\&\le \exp \bigg (\dfrac{-m(y_{M/2}-y_{i})}{\varepsilon +m {\tilde{h}}_2}\bigg ), \quad where \,\, {\tilde{h}}_2=4\tau _{0}\varepsilon M^{-1} L \\&=\exp \bigg (\dfrac{-m(1-y_{i})}{\varepsilon +m {\tilde{h}}_2}\bigg ) \\&=\exp \bigg (\dfrac{-m(M/2-i) {\tilde{h}}_2 }{\varepsilon +m {\tilde{h}}_2}\bigg )\\&=\exp \bigg (\dfrac{-2m\tau _{0}(1-2i/M) L }{1+4m\tau _{0} M^{-1} L}\bigg ). \end{aligned}$$

Using \(e^{-L}\le L/M\), we get

$$\begin{aligned} \dfrac{\phi _{i} (m)}{\phi _{M/2} (m)} \le L^{\frac{2m\tau _{0}(1-2i/M)}{1+4m\tau _{0} M^{-1}\ln M}} M^{\frac{-2m\tau _{0}(1-2i/M)}{1+4m\tau _{0} M^{-1}\ln M}} \le C L^{2m \tau _{0}(1-2i/M)} M^{-2m \tau _{0}(1-2i/M)}. \end{aligned}$$

Using similar arguments, we get the desired inequality for \(3M/4\le i \le M-1\). \(\square \)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Sharma, A., Rai, P. A hybrid numerical scheme for singular perturbation delay problems with integral boundary condition. J. Appl. Math. Comput. 68, 3445–3472 (2022). https://doi.org/10.1007/s12190-021-01667-x

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12190-021-01667-x

Keywords

Mathematics Subject Classification

Navigation