{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,2]],"date-time":"2025-10-02T00:44:12Z","timestamp":1759365852425,"version":"build-2065373602"},"reference-count":42,"publisher":"MDPI AG","issue":"10","license":[{"start":{"date-parts":[[2025,10,1]],"date-time":"2025-10-01T00:00:00Z","timestamp":1759276800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.mdpi.com"],"crossmark-restriction":true},"short-container-title":["Computation"],"abstract":"<jats:p>The core challenge of the Unique Continuity (UC) problem lies in inferring solutions across an entire domain using limited observational data, holding significant practical implications for multiphysics coupled models. Recently, physics-informed neural networks (PINNs) have shown considerable promise in addressing the UC problem. However, the reliance on a fixed activation function and a fixed weighted loss function prevents PINNs from adequately representing the multiphysics characteristics embedded in coupled models. To overcome these limitations, we propose a novel dual adaptive neural network (DANN) algorithm. This approach integrates trainable adaptive activation functions and an adaptively weighted loss scheme, enabling the network to dynamically balance the observational data and governing physics. Our method is applicable not only to the UC problem but also to general forward problems governed by partial differential equations. Furthermore, we provide a theoretical foundation for the algorithm by deriving a generalization error estimate, discussing the potential causes of neural networks solving this problem. Extensive numerical experiments including 3D demonstrate the superior accuracy and effectiveness of the proposed DANN framework in solving the UC problem compared to standard PINNs.<\/jats:p>","DOI":"10.3390\/computation13100228","type":"journal-article","created":{"date-parts":[[2025,10,1]],"date-time":"2025-10-01T11:06:02Z","timestamp":1759316762000},"page":"228","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Dual Adaptive Neural Network for Solving Free-Flow Coupled Porous Media Models Under Unique Continuation Problem"],"prefix":"10.3390","volume":"13","author":[{"given":"Kunhao","family":"Liu","sequence":"first","affiliation":[{"name":"College of Mathematics and Data Science, Shaanxi University of Science and Technology, Xi\u2019an 710021, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4016-0901","authenticated-orcid":false,"given":"Jibing","family":"Wu","sequence":"additional","affiliation":[{"name":"Laboratory for Big Data and Decision, College of Systems Engineering, National University of Defense Technology, Changsha 410073, China"}]}],"member":"1968","published-online":{"date-parts":[[2025,10,1]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"2357","DOI":"10.1002\/num.22718","article-title":"A decoupled stabilized finite element method for the dual\u2013porosity\u2013Navier\u2013Stokes fluid flow model arising in shale oil","volume":"37","author":"Gao","year":"2021","journal-title":"Numer. Methods Partial Differ. Equ."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"104712","DOI":"10.1016\/j.advwatres.2024.104712","article-title":"Lattice Boltzmann simulation of dissolution patterns in porous media: Single porosity versus dual porosity media","volume":"188","author":"Kashani","year":"2024","journal-title":"Adv. Water Resour."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"598","DOI":"10.1080\/27690911.2022.2117913","article-title":"Groundwater flow in karstic aquifer: Analytic solution of dual\u2013porosity fractional model to simulate groundwater flow","volume":"30","author":"Yadav","year":"2022","journal-title":"Appl. Math. Sci. Eng."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"116798","DOI":"10.1016\/j.cam.2025.116798","article-title":"A local parallel fully mixed finite element method for superposed fluid and porous layers","volume":"472","author":"Li","year":"2026","journal-title":"J. Comput. Appl. Math."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"67","DOI":"10.1007\/s10915-023-02185-7","article-title":"Modeling and a Domain Decomposition Method with Finite Element Discretization for Coupled Dual-Porosity Flow and Navier\u2013Stokes Flow","volume":"95","author":"Hou","year":"2023","journal-title":"J. Sci. Comput."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"B710","DOI":"10.1137\/15M1044072","article-title":"A dual\u2013porosity\u2013Stokes model and finite element method for coupling dual\u2013porosity flow and free flow","volume":"38","author":"Hou","year":"2016","journal-title":"SIAM J. Sci. Comput."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"16","DOI":"10.1007\/s10915-021-01674-x","article-title":"A parallel robin\u2013robin domain decomposition method based on modified characteristic fems for the time-dependent dual\u2013porosity\u2013Navier\u2013Stokes model with the Beavers\u2013Joseph interface condition","volume":"90","author":"Cao","year":"2022","journal-title":"J. Sci. Comput."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"41","DOI":"10.1007\/s10915-024-02771-3","article-title":"A Stokes\u2013dual\u2013porosity\u2013poroelasticity model and discontinuous galerkin method for the coupled free flow and dual porosity poroelastic medium problem","volume":"102","author":"Li","year":"2025","journal-title":"J. Sci. Comput."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"111397","DOI":"10.1016\/j.jcp.2022.111397","article-title":"A new coupled multiphysics model and partitioned time-stepping method for the triple-porosity-Stokes fluid flow model","volume":"466","author":"Nasu","year":"2022","journal-title":"J. Comput. Phys."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1093\/imanum\/drab093","article-title":"Estimates on the generalization error of physics-informed neural networks for approximating PDEs","volume":"43","author":"Mishra","year":"2023","journal-title":"IMA J. Numer. Anal."},{"key":"ref_11","first-page":"131261","article-title":"Physics-informed neural network for diffusive wave model","volume":"637","author":"Hou","year":"2024","journal-title":"J. Appl. Math. Comput."},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Dieva, N., Aminev, D., Kravchenko, M., and Smirnov, N. (2024). Overview of the Application of Physically Informed Neural Networks to the Problems of Nonlinear Fluid Flow in Porous Media. Computation, 12.","DOI":"10.3390\/computation12040069"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Frey, R., and K\u00f6ck, V. (2022). Deep Neural Network Algorithms for Parabolic PIDEs and Applications in Insurance and Finance. Computation, 10.","DOI":"10.3390\/computation10110201"},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Seabe, P.L., Moutsinga, C.R.B., and Pindza, E. (2023). Forecasting Cryptocurrency Prices Using LSTM, GRU, and Bi-Directional LSTM: A Deep Learning Approach. Fractal Fract., 7.","DOI":"10.3390\/fractalfract7020203"},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Yuan, S.H., Liu, Y.Q., Yan, L.M., Zhang, R.F., and Wu, S.J. (2025). Neural Networks-Based Analytical Solver for Exact Solutions of Fractional Partial Differential Equations. Fractal Fract., 9.","DOI":"10.3390\/fractalfract9080541"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"A2603","DOI":"10.1137\/18M1229845","article-title":"fPINNs: Fractional physics-informed neural networks","volume":"41","author":"Pang","year":"2019","journal-title":"SIAM J. Sci. Comput."},{"key":"ref_17","unstructured":"Dockhorn, T. (2019). A discussion on solving partial differential equations using neural networks. arXiv."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Chen, N. (2023). Stochastic Methods for Modeling and Predicting Complex Dynamical Systems: Uncertainty Quantification, State Estimation, and Reduced-Order Models, Springer Nature.","DOI":"10.1007\/978-3-031-22249-8"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"686","DOI":"10.1016\/j.jcp.2018.10.045","article-title":"Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations","volume":"378","author":"Raissi","year":"2019","journal-title":"J. Comput. Phys."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"981","DOI":"10.1093\/imanum\/drab032","article-title":"Estimates on the generalization error of physics-informed neural networks for approximating a class of inverse problems for PDEs","volume":"42","author":"Mishra","year":"2022","journal-title":"IMA J. Numer. Anal."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"037114","DOI":"10.1063\/5.0190429","article-title":"Two-dimensional temperature field inversion of turbine blade based on physics-informed neural networks","volume":"36","author":"Mai","year":"2024","journal-title":"Phys. Fluids."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"025001","DOI":"10.1088\/0266-5611\/32\/2\/025001","article-title":"Carleman estimate for the Navier\u2013Stokes equations and an application to a lateral Cauchy problem","volume":"32","author":"Bellassoued","year":"2016","journal-title":"Inverse Probl."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"2437","DOI":"10.1137\/151004239","article-title":"Controllability of evolution equations with memory","volume":"55","author":"Zhang","year":"2017","journal-title":"SIAM J. Control Optim."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"1087","DOI":"10.1088\/0266-5611\/21\/3\/018","article-title":"A mixed formulation of quasi\u2013reversibility to solve the Cauchy problem for Laplace\u2019s equation","volume":"21","author":"Bourgeois","year":"2005","journal-title":"Inverse Probl."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Badra, M., Caubet, F., and Dard\u00e9, J. (2016). Stability estimates for Navier\u2013Stokes equations and application to inverse problems. arXiv.","DOI":"10.3934\/dcdsb.2016052"},{"key":"ref_26","first-page":"1339","article-title":"DGM: A deep learning algorithm for solving partial differential equations","volume":"375","author":"Sirignano","year":"2018","journal-title":"J. Sci. Comput."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"A3055","DOI":"10.1137\/20M1318043","article-title":"Understanding and mitigating gradient flow pathologies in physics-informed neural networks","volume":"43","author":"Wang","year":"2021","journal-title":"SIAM J. Sci. Comput."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"111722","DOI":"10.1016\/j.jcp.2022.111722","article-title":"Self-adaptive physics-informed neural networks","volume":"474","author":"McClenny","year":"2023","journal-title":"J. Comput. Phys."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Aghaee, A., and Khan, M.O. (2024). Performance of Fourier-based activation function in physics-informed neural networks for patient-specific cardiovascular flows. Comput. Meth. Prog. Biomed., 247.","DOI":"10.1016\/j.cmpb.2024.108081"},{"key":"ref_30","unstructured":"Wang, H.H., Lu, L., Song, S.J., and Gao, H. (2023). Learning specialized activation functions for physics-informed neural networks. arXiv."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"109136","DOI":"10.1016\/j.jcp.2019.109136","article-title":"Adaptive activation functions accelerate convergence in deep and physics-informed neural networks","volume":"404","author":"Jagtap","year":"2020","journal-title":"J. Comput. Phys."},{"key":"ref_32","unstructured":"Ramachandran, P., Zoph, B., and Le, Q.V. (2017). Swish: A self-gated activation function. arXiv."},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Sunkari, S., Sangam, A., Raman, R., and Rajalakshmi, R. (2024). A refined ResNet18 architecture with Swish activation function for Diabetic Retinopathy classification. Biomed. Signal Proces., 88.","DOI":"10.1016\/j.bspc.2023.105630"},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"10","DOI":"10.17977\/um018v2i12019p41-46","article-title":"Adam optimization algorithm for wide and deep neural network. Knowledge Engineering and Data Science","volume":"2","author":"Jais","year":"2019","journal-title":"Knowl. Eng. Data Sci."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"112454","DOI":"10.1016\/j.cam.2019.112454","article-title":"A modified BFGS type quasi-Newton method with line search for symmetric nonlinear equations problems","volume":"367","author":"Zhou","year":"2020","journal-title":"J. Comput. Appl. Math."},{"key":"ref_36","first-page":"30","article-title":"introduction to numerical analysis springer-verlag","volume":"12","author":"Stoer","year":"2002","journal-title":"Texts Appl. Math."},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Cao, L.L., Li, J., Chen, Z.X., and Du, G.Z. (2023). A local parallel finite element method for superhydrophobic proppants in a hydraulic fracturing system based on a 2D\/3D transient triple\u2013porosity Navier\u2013Stokes model. arXiv.","DOI":"10.2139\/ssrn.4706776"},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"243","DOI":"10.1016\/S0045-7949(00)00123-1","article-title":"The Inf\u2013Sup condition and its evaluation for mixed finite element methods","volume":"79","author":"Bathe","year":"2001","journal-title":"Comput. Struct."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"80","DOI":"10.1016\/j.neunet.2023.08.006","article-title":"Prediction of fluid flow in porous media by sparse observations and physics-informed PointNet","volume":"167","author":"Kashefi","year":"2023","journal-title":"Neural Netw."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.camwa.2024.12.024","article-title":"MC-CDNNs: The Monte Carlo-coupled deep neural networks approach for stochastic dual-porosity-Stokes flow coupled model","volume":"181","author":"Li","year":"2025","journal-title":"Comput. Math. Appl."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"010201","DOI":"10.1088\/1674-1056\/ac7554","article-title":"The coupled deep neural networks for coupling of the Stokes and Darcy\u2013Forchheimer problems","volume":"32","author":"Yue","year":"2023","journal-title":"Chin. Phys. B"},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"469","DOI":"10.1016\/j.apm.2020.01.057","article-title":"Towards a better understanding of wall-driven square cavity flows using the lattice Boltzmann method","volume":"82","author":"Bo","year":"2020","journal-title":"Appl. Math. Model."}],"container-title":["Computation"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2079-3197\/13\/10\/228\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,1]],"date-time":"2025-10-01T11:25:51Z","timestamp":1759317951000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2079-3197\/13\/10\/228"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,1]]},"references-count":42,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2025,10]]}},"alternative-id":["computation13100228"],"URL":"https:\/\/doi.org\/10.3390\/computation13100228","relation":{},"ISSN":["2079-3197"],"issn-type":[{"value":"2079-3197","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10,1]]}}}