Construction of a multilayer solution of the Duffing equation based on analytical modification of numerical methods

Context and relevance. The purpose of the article is to study a new approach to constructing approximate solutions of parametric differential equations. The new approach combines the analytical modification of iterative formulas of conventional numerical methods with physically informed neural networks. The new method allows us to build a hierarchy of approximate solutions with high accuracy with a small neural network size. For problems with parameters, the method allows you to build parametric models with solutions in the form of functions. For these models, the task parameters are among the input variables. Methods and materials. The operation of the method is illustrated by the example of the Duffing equation. An analytical modification of the Heun's method is used to construct an approximate parametric solution. Our approach allows us to build mathematical models of three levels. As models of the first level, we investigate the method of analytical modification with the selection of the optimal parameter. As second-level models, we build models in which the parameter is a neural network, which is input to the task parameter. These models are somewhat more complex, but also more accurate. The models of the third level are considered models in which the parameter of the Heun's method is a neural network, the input of which is a task parameter and an independent variable. Results. The result is a hierarchy of adaptive models of varying complexity and accuracy, which makes it easy to adapt these models to the observational data of the object under study. Conclusions. The proposed method for constructing a multilayer neural network solution for parametric ordinary differential equations based on an analytical modification of numerical methods can be applied to for other differential equations, systems of differential equations, partial differential equations, et al.

1. Вержбицкий, В.М. (2005). Численные методы. Математический анализ и обыкновенные дифференциальные уравнения. (400 с.). М.: Оникс 21 век.
   Verzhbitsky, V.M. (2005). Numerical methods. Mathematical analysis and ordinary differential equations. (400 p.). Moscow: Onyx 21st century. (In Russ.).
2. Вержбицкий, В.М. (2021). Численные методы математической физики. (210 c.). Москва, Берлин: Директ-Медиа.
   Verzhbitsky, V.M. (2021). Numerical Methods of Mathematical Physics. (210 p.). Moscow, Berlin: Direct-Media. (In Russ.).
3. Кольцова, Э.М. (2026). Численные методы решения уравнений математической физики и химии. В: Э.М. Кольцова, А.С. Скичко, А.В. Женса (Ред.), Учебник для вузов (210 c.). М: Изд. Юрайт.
   Koltsova, E.M. (2026). Numerical methods for solving equations of mathematical physics and chemistry. In: E.M. Koltsova, A.S. Skichko, A.V. Zhensa (Ed.), Textbook for universities (220 p.). Moscow: Yurait Publishing House (In Russ.).
4. Guo, J., Zhu, H., Yang, Y., Guo, C. (2025). Advances in physics-informed neural networks for solving complex partial differential equations and their engineering applications: A systematic review. Engineering applications of artificial intelligence. 161:112044, doi:10.1016/j.engappai.2025.112044
5. Hairer, E., Wanner, G., Norsett, S.P. (1993). Solving Ordinary Differential Equations I. Springer, Berlin. https://doi.org/10.1007/978-3-540-78862-1
6. Jiang, P., Zhou, Q., Shao, X. (2020). Surrogate Model-Based Engineering Design and Optimization. Springer, Tracts in Mechanical Engineering. doi:10.1007/978-981-15-0731-1
7. Kembhavi, A., Kumar, R. (2023). Advances and Applications in Numerical Analysis. Perceptions of success and its relationship to the level of achievement in fencing. International Journal of Contemporary Research in Multidisciplinary, 2(5): 66-70.
8. Kuznetsov, S.V. (2025). On the extreme stiffness of degenerate Duffing oscillator: A feature overlooked by Ueda. Mechanics Research Communications. vol. 148:104504, doi:1016/j.mechrescom.2025.104504
9. Lagaris, I.E., Likas, A., Fotiadis, D.I. (1998). “Artificial neural networks for solving ordinary and partial differential equations,” IEEE Transactions on Neural Networks, vol. 9, pp. 987–1000.
10. Lazovskaya, T., Tarkhov, D. (2016). ‘Multilayer neural network models based on grid methods.’, in IOP Conference Series: Materials Science and Engineering; 158(1):012061
11. Liu, G.-R. (2025). PINN with Python. An Introduction Scientech Publisher LLC. 293 p.
12. Luo, K., Zhao, J., Wang, Y., Li, J., Wen, J., Liang, J., Soekmadji, H., Liao, S. (2025). Physics-informed neural networks for PDE problems: a comprehensive review Artif Intell Rev 58( 323)
13. Manzoni, L., Papetti, D.M., Cazzaniga, P., Spolaor, S., Mauri, G., Besozzi, D., Nobile, M.S. (2020). Surfing on Fitness Landscapes: A Boost on Optimization by Fourier Surrogate Modeling. Entropy, 22( 285).
14. Marcus, J. (2024). Neuer Machine Learning for Engineers: Introduction to Physics-Informed, Explainable Learning Methods for AI in Engineering Applications. 258 p.
15. Michaloglou, A., Papadimitriou, I., Gialampoukidis, I. et al. (2025). Physics-Informed Neural Networks in Materials Modeling and Design: A Review. Arch Computat Methods Eng. doi:1007/s11831-025-10448-9
16. Mohammed, J.K. (2025). Recent Advances in Approximate Methods for Predicting Nonlinear Vibrations in Cantilever Beams and Plates: A Review Eurasian journal of science and Engineering 11(2).
17. Piccirillo, V., Balthazar, J.M., Lima, J.J., Tusset, A.M. (2025). Stabilization of unstable periodic orbits in a Duffing system using an optimal time-delayed feedback control and pseudospectral method Chaos, Solitons & Fractals. 194:116201.
18. Raissi, M., Perdikaris, P., Karniadakis, G.E. (2017). ‘Physics Informed Deep Learning (Part I): Data-driven Solutions of Nonlinear Partial Differential Equations.’ arXiv 1711.10561.
19. Raissi, M., Perdikaris, P., Karniadakis, G.E. (2017). ‘Physics Informed Deep Learning (Part II): Data-driven Discovery of Nonlinear Partial Differential Equations’, arXiv:1711.10566
20. Raissi, M., Perdikaris, P., Karniadakis, G.E. (2019). ‘Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations’, Comput. Phys., (378), pp. 686–707
21. Ranft, S., von der Linden, W. (2021). "Bayesian Surrogate Analysis and Uncertainty Propagation". Physical Sciences Forum. 3 (1): 6. arXiv:2101.04038.
22. Ren, Z., Zhou, S., Liu, D., Liu, O. (2025). Physics-Informed Neural Networks: A Review of Methodological Evolution, Theoretical Foundations, and Interdisciplinary Frontiers Toward Next-Generation Scientific Computing Zhuang Zhao, Ye Wang, Weijian Zhang, Zhenggang Ba, Lin Sun, Physics-informed neural networks in heat transfer-dominated multiphysics systems: A comprehensive review, Engineering Applications of Artificial Intelligence. vol. 157:111098.
23. Tarkhov, D., Vasilyev, A. (2005). ‘New neural network technique to the numerical solution of mathematical physics problems. I: Simple problems Optical Memory and Neural Networks’, Information Optics, (14). pp. 59–72. II: Complicated and nonstandard problems, Optical Memory and Neural Networks’, Information Optics. (14), pp. 97-122.
24. Tarkhov, D., Vasilyev, A. (2020). Semi-Empirical Neural Network Modeling and Digital Twins Development, Academic Press: Cambridge, MA, USA. 281 p. ISBN: 978-0-12-815651-3.
25. Tarkhov, D., Lazovskaya, T., Malykhina, G. (2023). Constructing Physics-Informed Neural Networks with Architecture Based on Analytical Modification of Numerical Methods by Solving the Problem of Modelling Processes in a Chemical Reactor. Sensors, 23:663. https://doi.org/10.3390/s23020663
26. Umer, M., Olejnik, P. (2025). Approximate Analytical Approaches to Nonlinear Differential Equations: A Review of Perturbation, Decomposition and Coefficient Methods in Engineering Archives of Computational Methods in Engineering. vol. 32, 2767–2798p.
27. Zhou, M., Mei, G., Xu, N. (2023). Enhancing Computational Accuracy in Surrogate Modeling for Elastic–Plastic Problems by Coupling S-FEM and Physics-Informed Deep Learning. Mathematics. 11(9), 2016. https://doi.org/10.3390/math11092016