Physics-informed dictionary learning of time-varying 3D settlements from sparse monitoring data and 2D numerical models with consideration of complex stratigraphy

Abstract

Digital twins of geotechnical structures replicate their physical counterparts, such as underground spaces developed from land reclamations. These spaces often exhibit intricate three-dimensional (3D) stratigraphic distributions, including irregular and interbedded soil layers. Developing a virtual 3D model, such as finite element model (FEM), with complex stratigraphy poses significant computational challenges due to the necessity for numerous stratum voxels, high-resolution meshing, and prohibitive analysis times. Incorporating field settlement data for model updating escalates the computational burden, as repeated evaluations of 3D FEM models are required for each model updating. To address this challenge, this study develops a novel approach for efficiently predicting time-varying 3D settlement from two-dimensional (2D) numerical models with sparsely measured monitoring data. Settlements from 2D FEM analyses, which account for complex stratigraphy, are compiled within a dictionary learning framework and combined with limited monitoring data to estimate time-varying settlements at multiple 2D cross-sections. The 2D settlements are then utilized to reconstruct high-resolution 3D settlements through 3D compressive sampling (3D-CS), eliminating a need for additional numerical model evaluations when integrating new monitoring data. The proposed approach is illustrated using a reclamation project in Hong Kong, China.

Keywords

Digital twin / Temporally varying 3D settlement / Sophisticated stratigraphy / Sparse dictionary learning / Compressive sampling

CRediT authorship contribution statement

Dan-Ni Zhang: Writing - original draft, Visualization, Validation, Software, Methodology, Investigation, Formal analysis, Data curation. Hua-Ming Tian: Writing - review & editing, Writing - original draft, Validation, Supervision, Methodology, Formal analysis, Conceptualization. Yu Wang: Writing - review & editing, Writing - original draft, Validation, Supervision, Project administration, Methodology, Funding acquisition, Conceptualization. Chao Shi: Writing - review & editing, Visualization, Supervision, Project administration. Kostas Senetakis: Writing - review & editing, Validation, Supervision, Project administration.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The work described in this paper was supported by grants from the Research Grant Council of the Hong Kong Special Administrative Region (Project Nos: 11203322 and 11207724), the Ministry of Education, Singapore, under its AcRF regular Tier 1 Grant (Project nos. RG69/23 and RG86/25), and the Ministry of National Development under its Cities of Tomorrow (CoT) grant (Project no. CoT-V3-2024-2). The financial support is gratefully acknowledged.

References

[1] Babanagar, N., Sheil, B., Ninic´ J., Zhang, Q., Hardy, S., 2025. Digital twins for urban underground space. Tunn. Undergr. Space. Technol. 155, 106140. https://doi.org/10.1016/j.tust.2024.106140.

[2] Brinkgreve, R.B.J., Kumarswamy, S., Swolfs, W.M., Foria, F., 2018. PLAXIS 2D Material Models Manual. PLAXIS BV.

[3] Candès, E.J., Romberg, J., Tao, T., 2006. Robust uncertainty principles: Exact signal reconstruction from highly incomplete frequency information. IEEE Trans. Inf. Theory 52 (2), 489-509. https://doi.org/10.1109/TIT.2005.862083.

[4] Catuneanu, O., 2006. Principles of Sequence Stratigraphy. Elsevier, Amsterdam, p. 375.

[5] Chen, Z., Song, B., Wang, Z., Cai, Y., 2000. Late Quaternary evolution of the sub-aqueous Yangtze Delta, China: Sedimentation, stratigraphy, palynology, and deformation. Mar. Geol. 162 (2-4), 423-441. https://doi.org/10.1016/S0025-3227(99)00064-X.

[6] Deng, Z.P., Pan, M., Niu, J.T., Jiang, S.H., 2022. Full probability design of soil slopes considering both stratigraphic uncertainty and spatial variability of soil properties. Bull. Eng. Geol. Environ. 81 (5), 195. https://doi.org/10.1007/s10064-022-02702-2.

[7] Dagger, R., Saftner, D., Mayne, P., 2018. Cone penetration test design guide for state geotechnical engineers. Minnesota Department of Transportation: Retrieved from the University of Minnesota Digital Conservancy.

[8] Dumitrescu, B., Irofti, P., 2018. Dictionary Learning Algorithms and Applications. Springer, Berlin, Germany.

[9] Golder, 2015. Hong Kong Boundary crossing facilities - report on Ground conditions for PCB excavation. Unpublished report to Hong Kong-Zhuhai-Macao Bridge Hong Kong Project Management Office.

[10] Guan, Z., Wang, Y., Stuedlein, A.W., 2022. Efficient three-dimensional soil liquefaction potential and reconsolidation settlement assessment from limited CPTs considering spatial variability. Soil Dynam. Earthquake Eng. 163, 107518. https://doi.org/10.1016/j.soildyn.2022.107518.

[11] Hu, Y., Wang, Y., 2020. Probabilistic soil classification and stratification in a vertical cross-section from limited cone penetration tests using random field and Monte Carlo simulation. Comput. Geotech. 124, 103634. https://doi.org/10.1016/j.compgeo.2020.103634.

[12] Huang, H.W., Zhang, D.M., 2016. Resilience analysis of shield tunnel lining under extreme surcharge: Characterization and field application. Tunn. Undergr. Space. Technol. 51, 301-312. https://doi.org/10.1016/j.tust.2015.10.044.

[13] Huang, Z., Zhang, D., Pitilakis, K., Tsinidis, G., Huang, H., Zhang, D., Argyroudis, S., 2022. Resilience assessment of tunnels: Framework and application for tunnels in alluvial deposits exposed to seismic hazard. Soil Dynam. Earthquake Eng. 162, 107456. https://doi.org/10.1016/j.soildyn.2022.107456.

[14] Javaid, M., Haleem, A., Suman, R., 2023. Digital twin applications toward industry 4.0: A review. Cognitive Robot. 3, 71-92. https://doi.org/10.1016/j.cogr.2023.04.003.

[15] Jiang, F., Ma, L., Broyd, T., Chen, K., 2021. Digital twin and its implementations in the civil engineering sector. Autom. Constr. 130, 103838. https://doi.org/10.1016/j.autcon.2021.103838.

[16] Jiao, Z., Ding, C., Chen, L., Zhang, F., 2018. Three-dimensional imaging method for array ISAR based on sparse Bayesian inference. Sensors 18 (10), 3563. https://doi.org/10.3390/s18103563.

[17] Kreutz-Delgado, K., Murray, J.F., Rao, B.D., Engan, K., Lee, T.W., Sejnowski, T.J., 2003. Dictionary learning algorithms for sparse representation. Neural Comput. 15 (2), 349-396. https://doi.org/10.1162/089976603762552951.

[18] Lin, Z., Chen, Y., Liu, X., Jiang, R., Shen, B., Xin-Xin, G., 2020. Optimized design for sparse arrays in 3-D imaging sonar systems based on perturbed Bayesian compressive sensing. IEEE Sens. J. 20 (10), 5554-5565. https://doi.org/10.1109/JSEN.2020.2971568.

[19] Liu, Q., Lei, Y., Yin, X., Lei, J., Pan, Y., Sun, L., 2023. Development and application of a novel probabilistic back-analysis framework for geotechnical parameters in shield tunneling based on the surrogate model and Bayesian theory. Acta Geotech. 18 (9), 4899-4921. https://doi.org/10.1007/s11440-023-01850-3.

[20] Lo, M.K., Leung, Y.F., 2019. Bayesian updating of subsurface spatial variability for improved prediction of braced excavation response. Can. Geotech. J. 56 (8), 1169-1183. https://doi.org/10.1139/cgj-2018-0409.

[21] Lyu, B., Hu, Y., Wang, Y., 2023. Data-driven development of three-dimensional subsurface models from sparse measurements using Bayesian compressive sampling: A benchmarking study. ASCE-ASME J. Risk Uncertain. Eng. Syst., Part a: Civil Eng. 9 (2), 04023010. https://doi.org/10.1061/AJRUA6.RUENG-935.

[22] Lyu, B., Wang, Y., 2023. Benchmarking 3D subsurface models from Bayesian compressive sampling using real data. In: Ching J., Najjar S., Wang L. (Eds.), Geo-Risk 2023:Innovation in Data and Analysis Methods. Arlington, Virginia, pp. 428-437. https://doi.org/10.1061/9780784484975.045.

[23] Ma, Y., Tsao, D., Shum, H.Y., 2022. On the principles of parsimony and self-consistency for the emergence of intelligence. Front. Inform. Technol. Elect. Eng. 23 (9), 1298-1323. https://doi.org/10.48550/arXiv.2207.04630.

[24] Mairal, J., Bach, F., Ponce, J., Sapiro, G., 2009. Online dictionary learning for sparse coding,in:Proceedings of the 26th Annual International Conference on Machine Learning, pp. 689-696. https://doi.org/10.1145/1553374.1553463.

[25] Medina-Cetina, Z., Son, J., Moradi, M., 2019. Bayesian stratigraphy integration of geophysical, geological, and geotechnical surveys data, in:Offshore Technology Conference. https://doi.org/10.4043/29674-MS.

[26] Miro, S., König, M., Hartmann, D., Schanz, T., 2015. A probabilistic analysis of subsoil parameters uncertainty impacts on tunnel-induced ground movements with a back-analysis study. Comput. Geotech. 68, 38-53. https://doi.org/10.1016/j.compgeo.2015.03.012.

[27] Niazi, F., 2021. CPT-Based Geotechnical Design Manual, Volume 1:CPT Interpretation—Estimation of Soil Properties (Joint Transportation Research Program Publication No. FHWA/IN/JTRP-2021/22). West Lafayette, IN: Purdue University. https://doi.org/10.5703/1288284317346.

[28] Pati, Y.C., Rezaiifar, R., Krishnaprasad, P.S., 1993. Orthogonal matching pursuit:Recursive function approximation with applications to wavelet decomposition. In: Proceedings of 27th Asilomar conference on signals, systems and computers, Pacific Grove, CA, USA, vol. 1, pp. 40-44. https://doi.org/10.1109/ACSSC.1993.342465.

[29] Pilikos, G., 2020. The relevance vector machine for seismic Bayesian compressive sensing. Geophysics 85 (4), WA279-WA292. https://doi.org/10.1190/geo2019-0200.1.

[30] Qian, C., Liu, X., Ripley, C., Qian, M., Liang, F., Yu, W., 2022. Digital twin—Cyber replica of physical things: Architecture, applications and future research directions. Future Internet 14 (2), 64. https://doi.org/10.3390/fi14020064.

[31] Qian, Z.H., Shi, C., Wang, Y., Cao, Z.J., 2024. Nonparametric and continuous variable-based stratigraphic modelling from sparse boreholes using signed distance function and Bayesian compressive sensing. Can. Geotech. J. 62, 1-23. https://doi.org/10.1139/cgj-2024-0295.

[32] Qian, Z., Shi, C., 2024. Prior geological knowledge enhanced Markov random field for development of geological cross-sections from sparse data. Comput. Geotech. 173, 106587. https://doi.org/10.1016/j.compgeo.2024.106587.

[33] Robertson, P.K., 2009. Interpretation of cone penetration tests—a unified approach. Can. Geotech. J. 46 (11), 1337-1355. https://doi.org/10.1139/T09-065.

[34] Robertson, P.K., 2016. Cone penetration test (CPT)-based soil behaviour type (SBT) classification system—an update. Can. Geotech. J. 53 (12), 1910-1927. https://doi.org/10.13544/j.cnki.jeg.2020-263.

[35] Robertson, P.K., Wride, C.E., 1998. Evaluating cyclic liquefaction potential using the cone penetration test. Can. Geotech. J. 35 (3), 442-459. https://doi.org/10.1139/t98-017.

[36] Rocscience, Inc., 2021. CPT data interpretation theory manual. https://static.rocscience.cloud/.

[37] Rubinstein, R., Bruckstein, A.M., Elad, M., 2010. Dictionaries for sparse representation modeling. Proc. IEEE 98 (6), 1045-1057. https://doi.org/10.1109/JPROC.2010.2040551.

[38] Sadeghi, M., Babaie-Zadeh, M., Jutten, C., 2013. Dictionary learning for sparse representation: A novel approach. IEEE Signal Process. Lett. 20 (12), 1195-1198. https://doi.org/10.1109/LSP.2013.2285218.

[39] Salomon, D., 2007. Data compression:The Complete Reference. Springer Science & Business Media, New York.

[40] Salucci, M., Poli, L., Oliveri, G., 2019. Full-vectorial 3D microwave imaging of sparse scatterers through a multi-task Bayesian compressive sensing approach. J. Imag. 5 (1), 19. https://doi.org/10.3390/jimaging5010019.

[41] Shen, Y., El Naggar, M.H., Zhang, D., Huang, Z., Du, X., 2025. Optimal intensity measure for seismic performance assessment of shield tunnels in liquefiable and non-liquefiable soils. Undergr. Space 21, 149-163. https://doi.org/10.1016/j.undsp.2024.03.008.

[42] Shi, C., Wang, Y., 2022. Assessment of reclamation-induced consolidation settlement considering stratigraphic uncertainty and spatial variability of soil properties. Can. Geotech. J. 59 (7), 1215-1230. https://doi.org/10.1139/cgj-2021-0349.

[43] Shu, X., Ahuja, N., 2011. Imaging via three-dimensional compressive sampling (3DCS). In: 2011 International Conference on Computer Vision, Barcelona, Spain, pp. 439-446. https://doi.org/10.1109/ICCV.2011.6126273.

[44] Shuku, T., Phoon, K.K., 2023. Data-driven subsurface modelling using a Markov random field model. Georisk: Assessment and Management of Risk for Engineered. Systems and Geohazards 17 (1), 41-63. https://doi.org/10.1080/17499518.2023.2181973.

[45] Tian, H., Wang, Y., 2023. Data-driven and physics-informed Bayesian learning of spatiotemporally varying consolidation settlement from sparse site investigation and settlement monitoring data. Comput. Geotech. 157, 105328. https://doi.org/10.1016/j.compgeo.2023.105328.

[46] Tian, H.M., Wang, Y., 2024a. Optimal selection of dictionary atoms for sparse dictionary learning of time-varying monitoring data in two-dimensional geotechnical problems. Comput. Geotech. 165, 105953. https://doi.org/10.1016/j.compgeo.2023.105953.

[47] Tian, H.M., Wang, Y., 2024b. Interpretable machine learning for selection of site-specific soil constitutive models and consolidation settlement analysis. Comput. Geotech. 171, 106396. https://doi.org/10.1016/j.compgeo.2024.106396.

[48] Tian, H., Wang, Y., Zhang, D., 2025a. Real-time model updating and prediction of three-dimensional time-varying consolidation settlement using machine learning. J. Rock Mech. Geotech. Eng. 17 (9), 5954-5969. https://doi.org/10.1016/j.jrmge.2024.10.030.

[49] Tian, H.M., Wang, Y., Shi, C., 2025b. Machine learning-aided selection of CPT-based transformation models using field monitoring data from a specific project. Acta Geotech. 20 (1), 439-459. https://doi.org/10.1007/s11440-024-02475-w.

[50] Tian, B., Zhang, X., Li, L., Pu, L., Pu, L., Shi, J., Wei, S., 2021. Fast Bayesian compressed sensing algorithm via relevance vector machine for LASAR 3D imaging. Remote Sens. 13 (9), 1751. https://doi.org/10.3390/rs13091751.

[51] Tipping, M.E., 2001. Sparse Bayesian learning and the relevance vector machine. J. Mach. Learn. Res. 1, 211-244. https://doi.org/10.1162/15324430152748236.

[52] Tošic´ I., Frossard, P., 2011. Dictionary learning. IEEE Signal Process. Mag. 28 (2), 27-38. https://doi.org/10.1109/MSP.2010.939537.

[53] Wang, C., Tang, Z., Xu, H., 2024. WSSGCN: Wide sub-stage graph convolutional networks. Neurocomputing 602, 128273. https://doi.org/10.1016/j.neucom.2024.128273.

[54] Wang, Y., Tian, H.M., 2024. Digital geotechnics: from data-driven site characterisation towards digital transformation and intelligence in geotechnical engineering. Georisk: Assess. Manage. Risk Eng. Syst. Geohazards 18 (1), 8-32. https://doi.org/10.1080/17499518.2023.2278136.

[55] Wang, Y., Zhao, T., 2016. Interpretation of soil property profile from limited measurement data: A compressive sampling perspective. Can. Geotech. J. 53 (9), 1547-1559. https://doi.org/10.1139/cgj-2015-0545.

[56] Wright, J., Ma, Y., 2022. High-dimensional data analysis with low-dimensional models:Principles, computation, and applications. Cambridge University Press. https://doi.org/10.1017/9781108779302.

[57] Yang, H., Liu, Z., 2024. A fused sampling method integrating geotechnical and geophysical data for assessing three-dimensional soil liquefaction-induced damage capacity. Comput. Geotech. 166, 106024. https://doi.org/10.1016/j.compgeo.2023.106024.

[58] Wang, Y., Zhao, T., Hu, Y., Phoon, K.K., 2019. Simulation of random fields with trend from sparse measurements without detrending. J. Eng. Mech. 145 (2), 04018130. https://doi.org/10.1061/(ASCE)EM.1943-7889.0001560.

[59] Yan, W., Shen, P., Zhou, W.H., Ma, G., 2023. A rigorous random field-based framework for 3D stratigraphic uncertainty modelling. Eng. Geol. 323, 107235. https://doi.org/10.1016/j.enggeo.2023.107235.

[60] Yang, Z., Chen, J., Feng, Y., Li, X., 2025. Probabilistic analysis of offshore single pile composite foundation considering rotated anisotropy spatial variability of multi-layered soils. Comput. Geotech. 182, 107159. https://doi.org/10.1016/j.compgeo.2025.107159.

[61] Zhang, Z., Liu, X., Wei, S., Gan, H., Liu, F., Li, Y., Liu, F., 2019. Electrocardiogram reconstruction based on compressed sensing. IEEE Access 7, 37228-37237. https://doi.org/10.1109/ACCESS.2019.2905000.

[62] Zhao, C., Gong, W., Li, T., Juang, C.H., Tang, H., Wang, H., 2021. Probabilistic characterization of subsurface stratigraphic configuration with modified random field approach. Eng. Geol. 288, 106138. https://doi.org/10.1016/j.enggeo.2021.106138.

[63] Zhao, T., Wang, Y., 2020. Non-parametric simulation of non-stationary non-gaussian 3D random field samples directly from sparse measurements using signal decomposition and Markov Chain Monte Carlo (MCMC) simulation. Reliability Eng. Syst. Safety 203, 107087. https://doi.org/10.1016/j.ress.2020.107087.

[64] Zheng, H., Mooney, M., Gutierrez, M., 2023. Updating model parameters and predictions in SEM tunnelling using a surrogate-based Bayesian approach. Géotechnique 74 (13), 1855-1867. https://doi.org/10.1680/jgeot.22.00299.