Paper contribution

Study on time-dependent deformation surrogate model for Reservoir Bank Slope Considering Water Effects

Chenyao Wei; Chenfeng Gao; Songyu Yue; Denghui Yu; Zhiyong Pang; Yaoru Liu

Proceedings
EUROCK 2026
Pages
1317-1322
Submitted
28 January 2026
Accepted
15 June 2026
Published
12 August 2026

Summary

Abstract

Numerical simulation has proven effective for evaluating slope stability, its computational cost becomes a bottleneck when repeated model calls are required in engineering analyses. Surrogate models offer a practical alternative. This paper presents a deep learning-based surrogate model for time-dependent deformation prediction of reservoir bank slopes. The deformation process is simulated using an elastic-viscoplastic creep constitutive model that accounts for water effects. A dataset was constructed by generating 7490 sets of mechanical parameters via Latin hypercube sampling and computing the corresponding slope deformations at monitoring points. A neural network framework combining multi-layer stacked GRUs (gated recurrent units) and fully connected layers was developed to capture the complex nonlinear relationship between rock mass parameters and slope time-dependent deformation. Results demonstrate that the surrogate model achieves a coefficient of determination R² > 0.97 in predicting time-dependent slope deformation. Given a set of mechanical parameters, it can rapidly output the corresponding deformation, thereby providing a reliable basis for the analysis and calculation of rock slopes.

Contributors

Authors

  1. Chenyao Wei Tsinghua University
  2. Chenfeng Gao Tsinghua University ORCID 0000-0001-6703-6604
  3. Songyu Yue Tsinghua University
  4. Denghui Yu Tsinghua University
  5. Zhiyong Pang Tsinghua University
  6. Yaoru Liu Tsinghua University

Index terms

Keywords

Full text

Attachments

Works cited

References

  1. Chen ZY, Wang XG, Haberfield C, Yin JH, Wang YJ (2001) A three-dimensional slope stability analysis method using the upper bound theorem - Part I: theory and methods.International journal of rock mechanics and mining sciences 38(3): 369-378.https://doi.org/10.1016/S1365-1609(01)00012-0.
  2. Lyu QC, Liu T, Bai JL, Zhang Q, Wei CY, Liu YR. (2025) Study on deformation mechanism of reservoir slope near dam during operation period. American Rock Mechanics Association,Santa Fe, New Mexico, USA, 8 June 2025.
  3. Song DQ, Che AL, Zhu RJ,Ge XR. (2019) Natural frequency characteristics of rock masses containing a complex geological structure and their effects on the dynamic stability of slopes. Rock mechanics and rock Engineering 52(11):4457-4473.https://doi.org/10.1007/s00603-019-01885-7.
  4. Zhong K, Tan XH, Liu SW, Lu ZT, Hou XL, Wang Q. (2025) Prediction of slope failure probability based on machine learning with genetic-ADASYN algorithm. Engineering Geology 346:107885.https://doi.org/10.1016/j.enggeo.2024.107885.
  5. Qiu XY, Chen JY, Shi XZ, Zhang S, Zhou J, Chen X. (2018) Deformation prediction and analysis of underground mining during stacking of dry gangue in open-pit based on response surface methodology. Journal of central south university25(2):406-417.https://doi.org/10.1007/s11771-018-3746-3.
  6. Song JT, Yuan S, Xu ZG, Li XR. (2023) Fast inversion method for seepage parameters of core earth-rock dam based on LHS-SSA-MKELM fusion surrogate model. Structures 55:160-168.https://doi.org/10.1016/j.istruc.2023.06.049.
  7. Kasama K, Furukawa Z, Hu LH. (2021) Practical reliability analysis for earthquake-induced 3D landslide using stochastic response surface method. Computers and geotechnics137:104303.https://doi.org/10.1016/j.compgeo.2021.104303.
  8. Chehade HA, Guo X, Dias D, Sadek M, Jenck O, Chehade FH. (2023) Reliability analysis for internal seismic stability of geosynthetic-reinforced soil walls. Geosynthetics international 30(3): 296-314.https://doi.org/10.1680/jgein.22.00250.
  9. Cabanzo CM, Tinoco J, Sousa HS, Coelho M, Matos JC. (2023) Adaptation of traditional risk-based methodology for slopes to probabilistic-based approach integrating surrogate models. Computers and geotechnics. 161:105577.https://doi.org/10.1016/j.compgeo.2023.105577.
  10. Dadashzadeh N, Duzgun HSB, Yesiloglu-Gultekin N. (2017) Reliability-Based Stability Analysis of Rock Slopes Using Numerical Analysis and Response Surface Method. Rock mechanics and rock engineering50(8):2119-2133.https://doi.org/10.1007/s00603-017-1206-2.
  11. Liao K, Zhao XY, Wu YP, Miao FS, Pan YT, Beer M. (2025) Efficient reliability analysis of slopes in spatially variable soils with active learning-assisted bootstrap polynomial chaos expansion. Computers and geotechnics 179:107022.https://doi.org/10.1016/j.compgeo.2024.107022.
  12. Zhang TT, Dias D. (2024) An efficient solution for probabilistic slope seismic stability analysis based on polynomial chao kriging metamodel. Geohazards5(2):530-546.https://doi.org/10.3390/geohazards5020027.
  13. Zhou Z, Li DQ, Xiao T, Gao ZJ, Du WQ. (2021) Response surface guided adaptive slope reliability analysis in spatially varying soils. Computers and geotechnics132:103966.https://doi.org/10.1016/j.compgeo.2020.103966.
  14. Zhuang WY, Liu YR, Zhang K, Lyv QC, Hou SK, Yang Q. (2025) A physics knowledge-based surrogate model framework for timedependent slope deformation: Considering water effect and sliding states. Journal of rock mechanics and geotechnical engineering 17(9):5416-5436.https://doi.org/10.1016/j.jrmge.2024.11.002.
  15. Zhang L, Liu YR, Yang Q. (2014) A creep model with damage based on internal variable theory and its fundamental properties.Mechanics of materials78:44-55.https://doi.org/10.1016/j.mechmat.2014.07.017.
  16. Yan MT, Zhang ZM, Gao SC, Cao SY. (2022) Predicting aerodynamic pressure on a square cylinder from wake velocity field by masked gated recurrent unit model. Physics of fluids 34:115101.https://doi.org/10.1063/5.0110491.
  17. Ma L, Huang C, Liu ZS, Morin KA, Aziz M, Meints C. (2021) The correlation between drainage chemistry and weather for full-scale waste rock piles based on artificial neural network. Journal of contaminant hydrology239:103793.https://doi.org/10.1016/j.jconhyd.2021.103793.