Citation: | Jin Wen, Yun-Long Liu, Xue-Juan Ren, Donal O'Regan. A REGULARIZATION METHOD FOR BACKWARD PROBLEMS OF SINGULARLY PERTURBED PARABOLIC AND FRACTIONAL DIFFUSION EQUATIONS[J]. Journal of Applied Analysis & Computation, 2025, 15(4): 2212-2237. doi: 10.11948/20240424 |
In this paper, backward problems of singularly perturbed parabolic and fractional diffusion equations are studied from the additional temperature data at fixed time $ t=T$. We analyze the ill-posedness of these two inverse problems, and apply the quasi-reversibility regularization method to solve these problems. Then we obtain the convergence rates of logarithmic and Hölder types for the backward problems. Finally, several one- and two-dimensional numerical examples are given to verify the effectiveness and feasibility of the proposed method.
[1] | A. Y. Bomba and O. A. Fursachik, Inverse singularly perturbed problems of convection-diffusion type in quadrangular curvilinear domains, Mat. Metodi Fiz.-Mekh. Polya, 2009, 52(3), 59–66. |
[2] | X. Cai, D. L. Cai, R. Q. Wu and K. H. Xie, High accuracy non-equidistant method for singular perturbation reaction-diffusion problem, Applied Mathematics and Mechanics, 2009, 30, 175–182. doi: 10.1007/s10483-009-0205-8 |
[3] | P. Carter, A stabilizing effect of advection on planar interfaces in singularly perturbed reaction-diffusion equations, SIAM J. Appl. Math., 2024, 84(3), 1227–1253. doi: 10.1137/23M1610872 |
[4] | D. Chaikovskii, A. Liubavin and Y. Zhang, Asymptotic expansion regularization for inverse source problems in two-dimensional singularly perturbed nonlinear parabolic PDEs, CSIAM Trans. Appl. Math., 2023, 4(4), 721–757. doi: 10.4208/csiam-am.SO-2022-0017 |
[5] | D. Chaikovskii and Y. Zhang, Convergence analysis for forward and inverse problems in singularly perturbed time-dependent reaction-advection-diffusion equations, J. Comput. Phys., 2022, 470, Paper No. 111609, 32. |
[6] | J. Dardé, Iterated quasi-reversibility method applied to elliptic and parabolic data completion problems, Inverse Probl. Imaging, 2016, 10(2), 379–407. doi: 10.3934/ipi.2016005 |
[7] | A. M. Denisov, Approximate solution of an inverse problem for a singularly perturbed integro-differential heat equation, Comput. Math. Math. Phys., 2023, 63(5), 837–844. doi: 10.1134/S0965542523050081 |
[8] | A. M. Denisov, Approximation of the solution of an inverse problem for a singularly perturbed system of partial differential equations, Differ. Equ., 2023, 59(6), 762–768. doi: 10.1134/S0012266123060058 |
[9] | J. R. Dorroh and X. P. Ru, The application of the method of quasi-reversibility to the sideways heat equation, J. Math. Anal. Appl., 1999, 236(2), 503–519. doi: 10.1006/jmaa.1999.6462 |
[10] | N. V. Duc, N. V. Thang and N. T. Thành, The quasi-reversibility method for an inverse source problem for time-space fractional parabolic equations, J. Differential Equations, 2023, 344, 102–130. doi: 10.1016/j.jde.2022.10.029 |
[11] | L. Eldén, Approximations for a Cauchy problem for the heat equation, Inverse Problems, 1987, 3, 263–273. doi: 10.1088/0266-5611/3/2/009 |
[12] | S. Haber, Boundary Value Problem for a Singularly Perturbed Differential Equation, ProQuest LLC, Ann Arbor, MI, Thesis (Ph. D.)–Massachusetts Institute of Technology, 1955. |
[13] | S. Haber and N. Levinson, A boundary value problem for a singularly perturbed differential equation, Proc. Amer. Math. Soc., 1955, 6, 866–872. doi: 10.1090/S0002-9939-1955-0074634-3 |
[14] | S. I. Kabanikhin, Inverse and Ill-posed Problems: Theory and Applications, Walter de Gruyter GmbH & Co. KG, Berlin, 2012, vol. 55, xvi+459. |
[15] | S. Karasuljić and H. Ljevaković, On construction of a global numerical solution for a semilinear singularly-perturbed reaction diffusion boundary value problem, Mat. Bilten, 2020, 44(2), 131–148. |
[16] | A. A. Kilbas, H. M. Srivastava and J. J. Trujillo, Theory and Applications of Fractional Differential Equations, in: North-Holland Mathematics Studies, Elsevier Science B.V., Amsterdam, 2006, vol. 204, p. xvi+523. |
[17] | N. Kumar, S. Toprakseven, N. Singh Yadav and J. Y. Yuan, A Crank-Nicolson WG-FEM for unsteady 2D convection-diffusion equation with nonlinear reaction term on layer adapted mesh, Appl. Numer. Math., 2024, 201, 322–346. doi: 10.1016/j.apnum.2024.03.013 |
[18] | R. Lattès and J.-L. Lions, The Method of Quasi-Reversibility. Applications to Partial Differential Equations, Modern Analytic and Computational Methods in Science and Mathematics, American Elsevier Publishing Co., Inc., New York, 1969. |
[19] | T. T. Le and L. H. Nguyen, T.-P. Nguyen and W. Powell, The quasi-reversibility method to numerically solve an inverse source problem for hyperbolic equations, J. Sci. Comput., 2021, 87(3), Paper No. 90, 23. |
[20] | Y. S. Li and T. Wei, An inverse time-dependent source problem for a time-space fractional diffusion equation, Appl. Math. Comput., 2018, 336, 257–271. |
[21] | T. Linss, Maximum-norm error analysis of a non-monotone FEM for a singularly perturbed reaction-diffusion problem, BIT, 2007, 47(2), 379–391. doi: 10.1007/s10543-007-0118-z |
[22] | G. Q. Liu and Y. C. Su, A uniformly convergent difference scheme for the singular perturbation problem of a high order elliptic differential equation, Applied Mathematics and Mechanics, 1996, 17(5), 413-421. doi: 10.1007/BF00131089 |
[23] | D. V. Lukyanenko, A. A. Borzunov and M. A. Shishlenin, Solving coefficient inverse problems for nonlinear singularly perturbed equations of the reaction-diffusion-advection type with data on the position of a reaction front, Commun. Nonlinear Sci. Numer. Simul., 2021, 99, Paper No. 105824, 10. |
[24] | D. V. Lukyanenko, V. B. Grigorev, V. T. Volkov and M. A. Shishlenin, Solving of the coefficient inverse problem for a nonlinear singularly perturbed two-dimensional reaction-diffusion equation with the location of moving front data, Comput. Math. Appl., 2019, 77(5), 1245–1254. doi: 10.1016/j.camwa.2018.11.005 |
[25] | D. V. Lukyanenko, M. A. Shishlenin and V. T. Volkov, Solving of the coefficient inverse problems for a nonlinear singularly perturbed reaction-diffusion-advection equation with the final time data, Commun. Nonlinear Sci. Numer. Simul., 2018, 54, 233–247. doi: 10.1016/j.cnsns.2017.06.002 |
[26] | D. V. Lukyanenko, M. A. Shishlenin and V. T. Volkov, Asymptotic analysis of solving an inverse boundary value problem for a nonlinear singularly perturbed time-periodic reaction-diffusion-advection equation, J. Inverse Ill-Posed Probl., 2019, 27(5), 745–758. doi: 10.1515/jiip-2017-0074 |
[27] | D. V. Lukyanenko, V. T. Volkov, N. N. Nefedov, L. Recke and K. Schneider, Analytic-numerical approach to solving singularly perturbed parabolic equations with the use of dynamic adapted meshes, Model. Anal. Inf. Sist., 2016, 23(3), 334–341. doi: 10.18255/1818-1015-2016-3-334-341 |
[28] | Z. Mao and L. B. Liu, A moving grid algorithm for a strongly coupled system of singularly perturbed convection-diffusion problems, Math. Appl. (Wuhan), 2018, 31(3), 653–660. |
[29] | M. A. Mohye, J. B. Munyakazi and T. G. Dinka, A second order numerical method for two-parameter singularly perturbed time-delay parabolic problems, J. Math. Model., 2023, 11(4), 745–766. |
[30] | I. Podlubny, Fractional Differential Equations, An Introduction to Fractional Derivatives, Fractional Differential Equations, to Methods of their Solution and Some of their Applications, in: Mathematics in Science and Engineering, Academic Press, Inc., San Diego, CA, 1999, vol. 198, p. xxiv+340. |
[31] | Z. Qian, C.-L. Fu and R. Shi, A modified method for a backward heat conduction problem, Appl. Math. Comput., 2007, 185(1), 564–573. |
[32] | A. Seal, S. Natesan and S. Toprakseven, A dimensional-splitting weak Galerkin finite element method for 2d time-fractional diffusion equation, J. Sci. Comput., 2024, 98(3), Paper No. 56, 26. |
[33] | G. I. Shishkin and L. P. Shishkina, A stable standard difference scheme for a singularly perturbed convection-diffusion equation under computer perturbations, Tr. Inst. Mat. Mekh., 2014, 20(1), 322–333. |
[34] | C. M. Song, H. Li and J. N. Li, Initial boundary value problem for the singularly perturbed Boussinesq-type equation, Discrete Contin. Dyn. Syst., 2013, 709–717. |
[35] | S. Toprakseven, A weak Galerkin finite element method for time fractional reaction-diffusion-convection problems with variable coefficients, Appl. Numer. Math., 2021, 168, 1–12. doi: 10.1016/j.apnum.2021.05.021 |
[36] | S. Toprakseven and S. Dinibutun, A weak Galerkin finite element method for parabolic singularly perturbed convection-diffusion equations on layer-adapted meshes, Electron. Res. Arch., 2024, 32(8), 5033–5066. doi: 10.3934/era.2024232 |
[37] | S. Toprakseven and N. Srinivasan, An efficient weak Galerkin FEM for third-order singularly perturbed convection-diffusion differential equations on layer-adapted meshes, Appl. Numer. Math., 2024, 204, 130–146. doi: 10.1016/j.apnum.2024.06.009 |
[38] | R. Vulanović and L. Teofanov, A modification of the Shishkin discretization mesh for one-dimensional reaction diffusion problems, Applied Mathematics and Computation, 2013, 220, 104–116. doi: 10.1016/j.amc.2013.05.055 |
[39] | J. G. Wang and T. Wei, An iterative method for backward time-fractional diffusion problem, Numer. Methods Partial Differential Equations, 2014, 30(6), 2029–2041. doi: 10.1002/num.21887 |
[40] | J. G. Wang and T. Wei, Quasi-reversibility method to identify a space-dependent source for the time-fractional diffusion equation, Appl. Math. Model., 2015, 39(20), 6139–6149. doi: 10.1016/j.apm.2015.01.019 |
[41] | J. Wen, L. M. Huang and Z. X. Liu, A modified quasi-reversibility method for inverse source problem of Poisson equation, Inverse Probl. Sci. Eng., 2021, 29(12), 2098–2109. doi: 10.1080/17415977.2021.1902516 |
[42] | J. Wen, Z. Y. Li and Y. P. Wang, Solving the backward problem for time-fractional wave equations by the quasi-reversibility regularization method, Adv. Comput. Math., 2023, 49(6), Paper No. 80, 26. |
[43] | J. Wen, Y. L. Liu and D. O'Regan, An efficient iterative quasi-reversibility method for the inverse source problem of time-fractional diffusion equations, Numerical Heat Transfer Part B-Fundamentals, 2024. |
[44] | J. Wen, X. J. Ren and S. J. Wang, Simultaneous determination of source term and initial value in the heat conduction problem by modified quasi-reversibility regularization method, Numerical Heat Transfer Part B-Fundamentals, 2022, 82(3–4), 112–127. |
[45] | J. Wen, X. J. Ren and S. J. Wang, Simultaneous determination of source term and the initial value in the space-fractional diffusion problem by a novel modified quasi-reversibility regularization method, Physica Scripta, 2023, 98(2), 025201. doi: 10.1088/1402-4896/acaa68 |
[46] | J. Wen, C. W. Yue, Z. X. Liu and D. O'Regan, A fractional Landweber iteration method for simultaneous inversion in a time-fractional diffusion equation, J. Appl. Anal. Comput., 2023, 13(6), 3374–3402. |
[47] | F. Yang and C. L. Fu, Two regularization methods to identify time-dependent heat source through an internal measurement of temperature, Math. Comput. Modelling, 2011, 53(5–6), 793–804. |
[48] | F. Yang and C. L. Fu, A mollification regularization method for the inverse spatial-dependent heat source problem, J. Comput. Appl. Math., 2014, 255, 555–567. doi: 10.1016/j.cam.2013.06.012 |
[49] | Z. Q. Zhang and T. Wei, Identifying an unknown source in time-fractional diffusion equation by a truncation method, Applied Mathematics and Computation, 2013, 219(11), 5972–5983. doi: 10.1016/j.amc.2012.12.024 |
Exact solution and its approximation for Example 5.1 when
Exact solution and its approximation for Example 5.2 when
Exact solution and its approximation for Example 5.3 when
Numerical results for source term in Example 5.4.
The comparison of the numerical effects between the exact source term and its computed approximations for Example 5.4.
Exact solution and its approximation for Example 5.5 when
Exact solution and its approximation for Example 5.6 when
Exact solution and its approximation for Example 5.7 when
Numerical results for source term in Example 5.8 when
The comparison of the numerical effects between the exact source term and its computed approximations for Example 5.8 when