Numerical method for solving Fredholm and Volterra integral equations using artificial neural networks
https://doi.org/10.22405/2226-8383-2024-25-5-126-139
Abstract
Many problems in mathematics, mechanics, physics and other engineering disciplines lead to equations in which the unknown function appears under the integral sign. Integral equations are useful mathematical tools in many fields, so they are studied in many different aspects, such as the existence of solutions, approximation of solutions, calculation of correction or incorrigibility, correction of solutions, etc. Many articles mention the so-called PINN (physics-informed neural networks, which can be translated as physically conditioned neural networks), which have found application for solving differential equations, both ordinary and partial derivatives, as
well as systems of differential equations. PINNs are also used to solve integral equations, but publications usually provide methods for solving a certain class of equations, for example, the Fredholm equation of the 2nd kind or the Volterra equation of the 2nd kind. This article will describe a general method for solving continuous integral equations using neural networks that generalizes them to both Fredholm and Volterra integral equations. The essence of the method
is that the desired function is approximated by a neural network, which is essentially a huge function with a large number of adjustable parameters, which are selected from the condition of minimal squared residual, for which the parameters of the neural network are adjusted using the L-BFGS optimization algorithm. The results of the ANN method are compared with the exact solution for several typical integral equations.
About the Authors
Tien Duc Nguyen
Kazan (Volga) Federal University (Kazan); College
of Industrial Techniques (Bac Giang, Vietnam).
Russian Federation
Russian Federation
postgraduate student
Ilshat Zufarovich Akhmetov
Kazan (Volga) Federal University
Russian Federation
Russian Federation
postgraduate student
Anis Fuatovich Galimyanov
Kazan (Volga) Federal University
Russian Federation
Russian Federation
candidate of physical and mathematical sciences
References
1. Hosseini, S. M., Shahmorad, S. 2025, “Numerical piecewise approximate solution of Fredholm integro-differential equations by the Tau method”, Applied Mathematical Modelling., Т. 29, no. 11, pp. 1005–1021.
2. Maleknejad, K., Nouri, K., Yousefi, M. 2007, “Discussion on convergence of Legendre polynomial for numerical solution of integral equations”, Applied Mathematics and Computation., Т. 193, no. 2. pp. 335–339.
3. Maleknejad, K., Aghazadeh, N., Rabbani, M. 2006, “Numerical solution of second kind Fredholm integral equations system by using a Taylor-series expansion method”, Applied Mathematics and Computation., Т. 175, no 2, pp. 1229–1234.
4. G¨ulsu, M., Sezer, M. 2005, “The approximate solution of high-order linear difference equations with variable coefficients in terms of Taylor polynomials”, Applied mathematics and
5. computation., Т. 168, no. 1, pp. 76–88.
6. Ghasemi, M., Kajani, M. T., Babolian, E. 2007, “Numerical solutions of the nonlinear
7. Volterra–Fredholm integral equations by using homotopy perturbation method”, Applied
8. Mathematics and Computation., Т. 188, no. 1, pp. 446–449.
9. Ghasemi, M., Kajani, M. T., Babolian, E. 2007, “Numerical solutions of the nonlinear
10. Volterra–Fredholm integral equations by using homotopy perturbation method”, Applied
11. Mathematics and Computation., Т. 188, no. 1, pp. 446–449.
12. Abbasbandy, S. 2006, “Numerical solutions of the integral equations: Homotopy perturbation method and Adomian’s decomposition method”, Applied Mathematics and Computation., Т. 173, no. 1, pp. 493–500.
13. Nasir, A. N. K., Tokhi, M. O. 2015, “Novel metaheuristic hybrid spiral-dynamic bacteriachemotaxis algorithms for global optimisation”, Applied Soft Computing., Т. 27, pp. 357–375.
14. Hu, Y. C. 2015, “Flow-based tolerance rough sets for pattern classification”, Applied Soft
15. Computing., Т. 27, pp. 322–331.
16. Livi, L., Rizzi, A., Sadeghian, A. 2015, “Granular modeling and computing approaches for
17. intelligent analysis of non-geometric data”, Applied Soft Computing., Т. 27, pp. 567–574.
18. Bildik, N., Konuralp, A., Yal¸cınba¸s, S. 2010, “Comparison of Legendre polynomial approximation and variational iteration method for the solutions of general linear Fredholm integrodifferential equations”, Computers and Mathematics with Applications., Т. 59, no. 6, pp. 1909– 1917.
19. Saneifard, R. 2022, “Extended artificial neural networks approach for solving two-dimensional fractional-order Volterra-type integro-differential equations”, Information Sciences., Т. 612, pp. 887–897.
20. Jadoon, I. 2020, “Integrated meta-heuristics finite difference method for the dynamics of
21. nonlinear unipolar electrohydrodynamic pump flow model”, Applied Soft Computing., Т. 97,
22. pp. 106–791.
23. Livi, L., Rizzi, A., Sadeghian, A. 2015, “Granular modeling and computing approaches for
24. intelligent analysis of non-geometric data”, Applied Soft Computing., Т. 27, pp. 567–574.
25. Lagaris, I. E., Likas, A., Fotiadis, D. I. 1998, “Artificial neural networks for solving ordinary and partial differential equations”. IEEE transactions on neural networks., Т. 9, no. 5, pp. 987–1000.
26. Koryagin, A., Khudorozkov, R., Tsimfer, S. 2019, “PyDEns: A python framework for solving differential equations with neural networks”, arXiv preprint arXiv:1909.11544.
27. Guan, Y. 2022, “Solving Fredholm integral equations using deep learning”, International Journal of Applied and Computational Mathematics., Т. 8, no 2, pp. 87.
28. Effati, S., Buzhabadi, R. 2012, “A neural network approach for solving Fredholm integral
29. equations of the second kind”, Neural Computing and Applications., Т. 21, pp. 843–852.
30. Jafarian, A., Measoomy, S., Abbasbandy, S. 2015, “Artificial neural networks based modeling for solving Volterra integral equations system”, Applied Soft Computing., Т. 27, pp. 391–398.
31. Zhang, J. B., Yu, D. M., Pan, X. M. 2022, “Physics-Informed Neural Networks For the Solution of Electromagnetic Scattering by Integral Equations”, 2022 International Applied Computational Electromagnetics Society Symposium (ACES-China)., IEEE, pp. 1–2.
Review
For citations:
Nguyen T., Akhmetov I.Z., Galimyanov A.F. Numerical method for solving Fredholm and Volterra integral equations using artificial neural networks. Chebyshevskii Sbornik. 2024;25(5):126-139. (In Russ.) https://doi.org/10.22405/2226-8383-2024-25-5-126-139
Views:
381
Email this article 