This study introduces a novel hybrid computational framework that synergistically integrates neural networks with symbolic computation to address nonlinear partial differential equations (PDEs). By combining the robust nonlinear approximation capabilities of neural networks with the analytical precision of symbolic computation, our proposed symbolic ansatz method using neural network architecture (SANNA) achieves superior accuracy and efficiency compared to conventional numerical techniques. With in this framework, we design three distinct neural network architectures—each incorporating varied trial functions—and further integrate the bilinear neural network method (BNNM). To validate the effectiveness of our methodology, we apply it to the (3+1)-dimensional HB equation, a prototypical nonlinear model with significance in soliton theory and wave dynamics. The approach yields multiple novel analytical solutions, including periodic traveling waves and strongly localized nonlinear modes, all exhibiting clear mathematical interpretability and physical relevance. These results highlight the method's potential for applications in fluid dynamics, ocean engineering, and geophysical flow modeling.
Citation: Jianglong Shen, Min Liu, Jingbin Liang, Runfa Zhang. New solutions to the (3+1)-dimensional HB equation using bilinear neural networks method and symbolic ansatz method using neural network architecture[J]. AIMS Mathematics, 2025, 10(12): 30307-30330. doi: 10.3934/math.20251331
This study introduces a novel hybrid computational framework that synergistically integrates neural networks with symbolic computation to address nonlinear partial differential equations (PDEs). By combining the robust nonlinear approximation capabilities of neural networks with the analytical precision of symbolic computation, our proposed symbolic ansatz method using neural network architecture (SANNA) achieves superior accuracy and efficiency compared to conventional numerical techniques. With in this framework, we design three distinct neural network architectures—each incorporating varied trial functions—and further integrate the bilinear neural network method (BNNM). To validate the effectiveness of our methodology, we apply it to the (3+1)-dimensional HB equation, a prototypical nonlinear model with significance in soliton theory and wave dynamics. The approach yields multiple novel analytical solutions, including periodic traveling waves and strongly localized nonlinear modes, all exhibiting clear mathematical interpretability and physical relevance. These results highlight the method's potential for applications in fluid dynamics, ocean engineering, and geophysical flow modeling.
| [1] |
S. Tang, X. Feng, W. Wu, H. Xu, Physics-informed neural networks combined with polynomial interpolation to solve nonlinear partial differential equations, Comput. Math. Appl., 132 (2023), 48–62. https://doi.org/10.1016/j.camwa.2022.12.008 doi: 10.1016/j.camwa.2022.12.008
|
| [2] |
J. Yong, X. Luo, S. Sun, C. Ye, Deep mixed residual method for solving PDE-constrained optimization problems, Comput. Math. Appl., 176 (2024), 510–524. https://doi.org/10.1016/j.camwa.2024.11.009 doi: 10.1016/j.camwa.2024.11.009
|
| [3] |
G. Lei, Z. Lei, L. Shi, C. Zeng, D. Zhou, Solving PDEs on spheres with physics-informed convolutional neural networks, Appl. Comput. Harmon. Anal., 74 (2025), 101714. https://doi.org/10.1016/j.acha.2024.101714 doi: 10.1016/j.acha.2024.101714
|
| [4] |
Z. Zhao, Y. Zhang, Periodic wave solutions and asymptotic analysis of the Hirota-Satsuma shallow water wave equation, Math. Method. Appl. Sci., 38 (2015), 4262–4271. https://doi.org/10.1002/mma.3362 doi: 10.1002/mma.3362
|
| [5] |
J. Chu, Y. Liu, W. Ma, Integrability and multiple-rogue and multi-soliton wave solutions of the (3+1)-dimensional Hirota-Satsuma-Ito equation, Mod. Phys. Lett. B, 39 (2025), 2550060. https://doi.org/10.1142/S0217984925500605 doi: 10.1142/S0217984925500605
|
| [6] |
J. Shen, R. Zhang, J. Huang, J. Liang, Neural network-based symbolic computation algorithm for solving (2+1)-dimensional Yu-Toda-Sasa-Fukuyama equation, Mathematics, 13 (2025), 3006. https://doi.org/10.3390/math13183006 doi: 10.3390/math13183006
|
| [7] |
W. Huang, C. Zhou, X. Lü, J. Wang, Dispersive optical solitons for the Schrödinger-Hirota equation in optical fibers, Mod. Phys. Lett. B, 35 (2021), 2150060. https://doi.org/10.1142/S0217984921500603 doi: 10.1142/S0217984921500603
|
| [8] |
Y. Wang, C. Temuer, Y. Yang, Integrability for the generalised variable-coefficient fifth-order Korteweg-de Vries equation with Bell polynomials, Appl. Math. Lett., 29 (2014), 13–19. https://doi.org/10.1016/j.aml.2013.10.007 doi: 10.1016/j.aml.2013.10.007
|
| [9] |
Y. Zhou, W. Ma, Applications of linear superposition principle to resonant solitons and complexitons, Comput. Math. Appl., 73 (2017), 1697–1706. https://doi.org/10.1016/j.camwa.2017.02.015 doi: 10.1016/j.camwa.2017.02.015
|
| [10] |
L. Gao, Y. Zi, Y. Yin, W. Ma, X Lü, Bäcklund transformation, multiple wave solutions and lump solutions to a (3+1)-dimensional nonlinear evolution equation, Nonlinear Dyn., 89 (2017), 2233–2240. https://doi.org/10.1007/s11071-017-3581-3 doi: 10.1007/s11071-017-3581-3
|
| [11] |
B. Ren, W. Ma, J. Yu, Characteristics and interactions of solitary and lump waves of a (2+1)-dimensional coupled nonlinear partial differential equation, Nonlinear Dyn., 96 (2019), 717–727. https://doi.org/10.1007/s11071-019-04816-x doi: 10.1007/s11071-019-04816-x
|
| [12] |
Z. Lan, Y. Gao, J. Yang, C. Su, C. Zhao, Z. Gao, Solitons and Bäcklund transformation for a generalized (3+1)-dimensional variable-coefficient B-type Kadomtsev-Petviashvili equation in fluid dynamics, Appl. Math. Lett., 60 (2016), 96–100. https://doi.org/10.1016/j.aml.2016.03.021 doi: 10.1016/j.aml.2016.03.021
|
| [13] |
Z. Lan, Y. Gao, C. Zhao, J. Yang, C. Su, Dark soliton interactions for a fifth-order nonlinear Schrödinger equation in a Heisenberg ferromagnetic spin chain, Superlattices and Microstructures, 100 (2016), 191–197. https://doi.org/10.1016/j.spmi.2016.09.022 doi: 10.1016/j.spmi.2016.09.022
|
| [14] |
L. Gao, X. Zhao, Y. Zi, J. Yu, X. Lü, Resonant behavior of multiple wave solutions to a Hirota bilinear equation, Comput. Math. Appl., 72 (2016), 1225–1229. https://doi.org/10.1016/j.camwa.2016.06.008 doi: 10.1016/j.camwa.2016.06.008
|
| [15] |
W. Ma, Y. Zhang, Y. Tang, J. Tu, Hirota bilinear equations with linear subspaces of solutions, Appl. Math. Comput., 218 (2012), 7174–7183. https://doi.org/10.1016/j.amc.2011.12.085 doi: 10.1016/j.amc.2011.12.085
|
| [16] |
W. Li, A. Jiao, W. Liu, Z. Guo, High-order rational-type solutions of the analogous (3+1)-dimensional Hirota-bilinear-like equation, Math. Biosci. Eng., 20 (2023), 19360–19371. https://doi.org/10.3934/mbe.2023856 doi: 10.3934/mbe.2023856
|
| [17] |
Y. Ye, L. Wang, Z. Chang, J. He, An efficient algorithm of logarithmic transformation to Hirota bilinear form of KdV-type bilinear equation, Appl. Math. Comput., 218 (2011), 2200–2209. https://doi.org/10.1016/j.amc.2011.07.036 doi: 10.1016/j.amc.2011.07.036
|
| [18] |
H. Ismael, H. Nabi, T. Sulaiman, N. Shah, S. Eldin, H. Bulut, Hybrid and physical interaction phenomena solutions to the Hirota bilinear equation in shallow water waves theory, Results Phys., 53 (2023), 106978. https://doi.org/10.1016/j.rinp.2023.106978 doi: 10.1016/j.rinp.2023.106978
|
| [19] |
H. Zheng, W. Ma, X. Gu, Hirota bilinear equations with linear subspaces of hyperbolic and trigonometric function solutions, Appl. Math. Comput., 220 (2013), 226–234. https://doi.org/10.1016/j.amc.2013.06.019 doi: 10.1016/j.amc.2013.06.019
|
| [20] |
T. Yin, Z. Xing, J. Pang, Modified Hirota bilinear method to (3+1)-D variable coefficients generalized shallow water wave equation, Nonlinear Dyn., 111 (2023), 9741–9752. https://doi.org/10.1007/s11071-023-08356-3 doi: 10.1007/s11071-023-08356-3
|
| [21] |
M. Dong, S. Tian, X. Yan, L. Zou, Solitary waves, homoclinic breather waves and rogue waves of the (3+1)-dimensional Hirota bilinear equation, Comput. Math. Appl., 75 (2018), 957–964. https://doi.org/10.1016/j.camwa.2017.10.037 doi: 10.1016/j.camwa.2017.10.037
|
| [22] |
G. Zhu, H. Wang, Z. Mou, Y. Lin, Various solutions of the (2+1)-dimensional Hirota-Satsuma-Ito equation using the bilinear neural network method, Chinese J. Phys., 83 (2023), 292–305. https://doi.org/10.1016/j.cjph.2023.03.016 doi: 10.1016/j.cjph.2023.03.016
|
| [23] |
M. Alquran, R. Alhami, Analysis of lumps, single-stripe, breather-wave, and two-wave solutions to the generalized perturbed-KdV equation by means of Hirota's bilinear method, Nonlinear Dyn., 109 (2022), 1985–1992. https://doi.org/10.1007/s11071-022-07509-0 doi: 10.1007/s11071-022-07509-0
|
| [24] |
U. Mandal, S. Malik, S. Kumar, A. Das, A generalized (2+1)-dimensional Hirota bilinear equation: integrability, solitons and invariant solutions, Nonlinear Dyn., 111 (2023), 4593–4611. https://doi.org/10.1007/s11071-022-08036-8 doi: 10.1007/s11071-022-08036-8
|
| [25] |
J. Guo, Y. Yao, H. Wang, T. Gu, Pre-training strategy for solving evolution equations based on physics-informed neural networks, J. Comput. Phys., 489 (2023), 112258. https://doi.org/10.1016/j.jcp.2023.112258 doi: 10.1016/j.jcp.2023.112258
|
| [26] |
F. Li, J. Wang, Y. Yang, Exact and Data-Driven lump wave solutions for the (3+1)-dimensional Hirota-Satsuma-Ito-like equation, Symmetry, 16 (2024), 1469. https://doi.org/10.3390/sym16111469 doi: 10.3390/sym16111469
|
| [27] |
A. Bihlo, R. O. Popovych, Physics-informed neural networks for the shallow-water equations on the sphere, J. Comput. Phys., 456 (2022), 111024. https://doi.org/10.1016/j.jcp.2022.111024 doi: 10.1016/j.jcp.2022.111024
|
| [28] |
Y. Zhang, M. Wang, F. Zhang, Z. Chen, A solution method for differential equations based on Taylor PINN, IEEE Access, 11 (2023), 145020–145030. https://doi.org/10.1109/ACCESS.2023.3331330 doi: 10.1109/ACCESS.2023.3331330
|
| [29] |
M. R. Admon, N. Senu, A. Ahmadian, Z. A. Majid, S. Salahshour, A new modern scheme for solving fractal-fractional differential equations based on deep feedforward neural network with multiple hidden layer, Math. Comput. Simulat., 218 (2024), 311–333. https://doi.org/10.1016/j.matcom.2023.11.002 doi: 10.1016/j.matcom.2023.11.002
|
| [30] |
M. L. Shahab, H. Susanto, Neural networks for bifurcation and linear stability analysis of steady states in partial differential equations, Appl. Math. Comput., 483 (2024), 128985. https://doi.org/10.1016/j.amc.2024.128985 doi: 10.1016/j.amc.2024.128985
|
| [31] |
C. Yi, Y. Chen, X. Lan, Comparison on neural solvers for the Lyapunov matrix equation with stationary & nonstationary coefficients, Appl. Math. Model., 37 (2013), 2495–2502. https://doi.org/10.1016/j.apm.2012.06.022 doi: 10.1016/j.apm.2012.06.022
|
| [32] |
Z. Lan, N-soliton solutions, Bäcklund transformation and Lax pair for a generalized variable-coefficient cylindrical Kadomtsev-Petviashvili equation, Appl. Math. Lett., 158 (2024), 109239. https://doi.org/10.1016/j.aml.2024.109239 doi: 10.1016/j.aml.2024.109239
|
| [33] |
Y. Shen, B. Tian, Bilinear auto-Bäcklund transformations and soliton solutions of a (3+1)-dimensional generalized nonlinear evolution equation for the shallow water waves, Appl. Math. Lett., 122 (2021), 107301. https://doi.org/10.1016/j.aml.2021.107301 doi: 10.1016/j.aml.2021.107301
|
| [34] |
S. Kumar, B. Mohan, Bilinearization and new center-controlled N-rogue solutions to a (3+1)-dimensional generalized KdV-type equation in plasmas via direct symbolic approach, Nonlinear Dyn., 112 (2024), 11373–11382. https://doi.org/10.1007/s11071-024-09626-4 doi: 10.1007/s11071-024-09626-4
|
| [35] |
K. Wang, H. Zhu, F. Shi, X. Liu, G. Wang, G. Li, Lump wave, breather wave and other abundant wave solutions to the (2+1)-dimensional Sawada-Kotera-Kadomtsev Petviashvili equation of fluid mechanics, Pramana, 99 (2025), 40. https://doi.org/10.1007/s12043-024-02884-2 doi: 10.1007/s12043-024-02884-2
|
| [36] |
K. Hosseini, F. Alizadeh, E. Hinçal, M. Ilie, M. S. Osman, Bilinear Bäcklund transformation, Lax pair, Painlevé integrability, and different wave structures of a 3D generalized KdV equation, Nonlinear Dyn., 112 (2024), 18397–18411. https://doi.org/10.1007/s11071-024-09944-7 doi: 10.1007/s11071-024-09944-7
|
| [37] |
J. Manafian, Novel solitary wave solutions for the (3+1)-dimensional extended Jimbo-Miwa equations, Comput. Math. Appl., 76 (2018), 1246–1260. https://doi.org/10.1016/j.camwa.2018.06.018 doi: 10.1016/j.camwa.2018.06.018
|
| [38] |
R. Zhang, S. Bilige, Bilinear neural network method to obtain the exact analytical solutions of nonlinear partial differential equations and its application to p-gBKP equation, Nonlinear Dyn., 95 (2019), 3041–3048. https://doi.org/10.1007/s11071-018-04739-z doi: 10.1007/s11071-018-04739-z
|
| [39] |
R. Zhang, M. Li, A. Cherraf, S. Vadyala, The interference wave and the bright and dark soliton for two integro-differential equation by using BNNM, Nonlinear Dyn., 111 (2023), 8637–8646. https://doi.org/10.1007/s11071-023-08257-5 doi: 10.1007/s11071-023-08257-5
|
| [40] |
M. Isah, A. Yokus, D. Kaya, Exploring the influence of layer and neuron configurations on Boussinesq equation solutions via a bilinear neural network framework, Nonlinear Dyn., 112 (2024), 13361–13377. https://doi.org/10.1007/s11071-024-09708-3 doi: 10.1007/s11071-024-09708-3
|
| [41] |
S. Zhang, G. Zhu, W. Huang, H. Wang, C. Yang, Y. Lin, Symbolic computation of analytical solutions for nonlinear partial differential equations based on bilinear neural network method, Nonlinear Dyn., 113 (2025), 7121–7137. https://doi.org/10.1007/s11071-024-10715-7 doi: 10.1007/s11071-024-10715-7
|
| [42] |
M. Sharifi, S. Shojaei, Y. Ighris, I. El Glili, A. Mohammadi, A. Chamkha, et al., Statistical and machine learning approaches integrated with CFD for effective thermal conductivity prediction of NEPCMs in natural convection between horizontal concentric cylinders, J. Energy Storage, 141 (2026), 119340. https://doi.org/10.1016/j.est.2025.119340 doi: 10.1016/j.est.2025.119340
|
| [43] |
M. Sharifi, A. Mohammadi, A. Chamkha, A. Aly, Hydrothermal and entropy generation analysis of mixed convection heat transfer in Couette-Poiseuille flow of a trihybrid nanofluid over a backward-facing step, Int. J. Therm. Sci., 220 (2026), 110283. https://doi.org/10.1016/j.ijthermalsci.2025.110283 doi: 10.1016/j.ijthermalsci.2025.110283
|
| [44] |
A. Hyder, H. Budak, A. Aly, S. Abdelsalam, Exploring fractional Bullen-type inequalities via second-derivative bounds and artificial neural networks, Eng. Appl. Artif. Intel., 162 (2025), 110283. https://doi.org/10.1016/j.engappai.2025.112619 doi: 10.1016/j.engappai.2025.112619
|
Figures(9)
Jianglong Shen, Min Liu, Jingbin Liang, Runfa Zhang. New solutions to the (3+1)-dimensional HB equation using bilinear neural networks method and symbolic ansatz method using neural network architecture[J]. AIMS Mathematics, 2025, 10(12): 30307-30330. doi: 10.3934/math.20251331
DownLoad: