Paper contribution

A machine learning-based tool for calibrating UDEC-BBM parameters for modelling intact rocks

Fengchang Bu; Michel Jaboyedoff; Ruoshen Lin

Proceedings
EUROCK 2026
Pages
1371-1376
Submitted
28 January 2026
Accepted
15 June 2026
Published
12 August 2026

Summary

Abstract

Despite the widespread prevalence of the DEM (discrete element method) in rock mechanics and rock engineering, existing trial-and-error and sensitivity analyses for DEM calibration suffer from inefficiency and subjectivity; therefore the calibration of DEM parameters remains a significant challenge. Given the considerable advantages of the DEM-based UDEC-BBM (universal distinct element code-bonded block model) in modelling intact rocks, we employed machine learning to calibrate the modelling parameters in UDEC-BBM. A comprehensive numerical database with 3,456 UDEC simulations was constructed for training machine learning models to establish the mapping from modelling parameters to modelled properties. Subsequently, the grid search was employed to back-calibrate modelling parameters according to the input target modelled properties (experimental data). This methodology has been implemented by the open-source software UPCal. To validate the UPCal, we collected experimental data from 99 rock types that served as modelled properties to be input into the UPCal to output the corresponding modelling parameters. The UPCal-calibrated modelling parameters were further imported into UDEC to perform another 1,485 simulations. These UDEC-simulated properties agree well with the experimental values, evidenced by the calibration accuracy as high as overall R2 = 0.94 for E (Young's modulus), 0.89 for ν (Poisson's ratio), 0.98 for UCS (unconfined compressive strength), 0.90 for BTS (Brazilian tensile strength), 0.76 for c (cohesion), and 0.73 for ϕ (internal friction angle), indicating the reliability and robustness of the UPCal-calibrated modelling parameters. UPCal achieves the rapid calibration of globally optimal UDEC-BBM modelling parameters using machine learning integrated with grid search, promoting the application of DEM in supporting both scientific research and engineering simulation practice.

Contributors

Authors

  1. Fengchang Bu University of Lausanne, Switzerland ORCID 0000-0003-1040-9486
  2. Michel Jaboyedoff University of Lausanne, Switzerland
  3. Ruoshen Lin University of Lausanne, Switzerland

Index terms

Keywords

Full text

Attachments

Works cited

References

  1. Barton N (2020) A review of mechanical over-closure and thermal over-closure of rock joints: Potential consequences for coupled modelling of nuclear waste disposal and geothermal energy development. Tunnelling and Underground Space Technology 99: 103379. https://doi.org/10.1016/j.tust.2020.103379.
  2. Boikov AV, Savelev RV, Payor VA (2018) DEM calibration approach: Random forest. Journal of Physics: Conference Series 1118 (1): 012009. https://doi.org/10.1088/1742-6596/1118/1/012009.
  3. Bu F, Jaboyedoff M, Derron MH, Xue L, Cui Y, et al. (2025) Parametric study to implement a water-weakening process in UDEC. Journal of Rock Mechanics and Geotechnical Engineering 17 (3): 1378–1386. https://doi.org/10.1016/j.jrmge.2024.05.030.
  4. Bu F, Lin R, Jaboyedoff M, Liu W, Xue L (2026a) Parametric calibration in bonded block models for simulating mechanical behaviours of intact rocks using machine learning. International Journal of Rock Mechanics and Mining Sciences 199: 106400. https://doi.org/10.1016/j.ijrmms.2026.106400.
  5. Bu F, Lin R, Jaboyedoff M, Liu W, Xue L (2026b) An adaptive UDEC calibration tool for modelling intact rocks based on machine learning. Journal of Rock Mechanics and Geotechnical Engineering. https://doi.org/10.1016/j.jrmge.2025.12.034.
  6. Bu F, Xue L, Zhai M, Huang X, Dong J, et al. (2022a) Evaluation of the characterization of acoustic emission of brittle rocks from the experiment to numerical simulation. Scientific Reports 12 (1): 498. https://doi.org/10.1038/s41598-021-03910-8.
  7. Bu F, Xue L, Zhai M, Xu C, Cui Y (2022b) Numerical investigation of the scale effects of rock bridges. Rock Mechanics and Rock Engineering 55:5671–5685. https://doi.org/10.1007/s00603-022-02952-2.
  8. Cho N, Martin CD, Sego DC (2007) A clumped particle model for rock. International Journal of Rock Mechanics and Mining Sciences 44 (7): 997–1010. https://doi.org/10.1016/j.ijrmms.2007.02.002.
  9. Cundall PA (1971) A computer model for simulating progressive, large-scale movements in block rock systems. Proceedings of the International Symposium on Rock Mechanics, Nancy, France.
  10. Diederichs MS (1999) Instability of hard rockmasses: The role of tensile damage and relaxation. Dissertation at University of Waterloo.
  11. Fabjan T, Mas Ivars D, Vukadin V (2015) Numerical simulation of intact rock behaviour via the continuum and Voronoi tesselletion models: A sensitivity analysis. Acta Geotechnica Slovenica 12 (2): 5–23.
  12. Fathipour-Azar H (2022) Machine learning-assisted distinct element model calibration: ANFIS, SVM, GPR, and MARS approaches. Acta Geotechnica 17 (4): 1207–1217. https://doi.org/10.1007/s11440-021-01303-9.
  13. Guo J, Xu G, Jing H, Kuang T (2013) Fast determination of meso-level mechanical parameters of PFC models. International Journal of Mining Science and Technology 23 (1): 157–162. https://doi.org/10.1016/j.ijmst.2013.03.007.
  14. Hudson JA, Harrison JP (2000) Engineering rock mechanics: An introduction to the principles. Elsevier.
  15. Itasca (2019) Universal distinct element code (UDEC) user’s guide, version 6.0. Itasca.
  16. Itasca (2025) Particle flow code (PFC) user’s guide, version 9.0.Itasca.
  17. Lan H, Martin CD, Hu B (2010) Effect of heterogeneity of brittle rock on micromechanical extensile behavior during compression loading. Journal of Geophysical Research: Solid Earth 115 (B1): B01202. https://doi.org/10.1029/2009JB006496.
  18. Lisjak A, Grasselli G (2014) A review of discrete modeling techniques for fracturing processes in discontinuous rock masses. Journal of Rock Mechanics and Geotechnical Engineering 6 (4): 301–314. https://doi.org/10.1016/j.jrmge.2013.12.007.
  19. Mayer JM, Stead D (2017) Exploration into the causes of uncertainty in UDEC Grain Boundary Models. Computers and Geotechnics 82: 110–123. https://doi.org/10.1016/j.compgeo.2016.10.003.
  20. Nicksiar M, Martin CD (2014) Factors affecting crack initiation in low porosity crystalline rocks. Rock Mechanics and Rock Engineering 47 (4): 1165–1181. https://doi.org/10.1007/s00603-013-0451-2.
  21. Potyondy DO, Cundall PA (2004) A bonded-particle model for rock. International Journal of Rock Mechanics and Mining Sciences 41 (8): 1329–1364. https://doi.org/10.1016/j.ijrmms.2004.09.011.
  22. Rasmussen LL (2021) Hybrid lattice/discrete element method for bonded block modeling of rocks. Computers and Geotechnics 130: 103907. https://doi.org/10.1016/j.compgeo.2020.103907.
  23. Shentu J, Lin B (2023) A novel machine learning framework for efficient calibration of complex DEM model: A case study of a conglomerate sample. Engineering Fracture Mechanics 279: 109044. https://doi.org/10.1016/j.engfracmech.2023.109044.
  24. Sinha S, Walton G (2020) A study on Bonded Block Model (BBM) complexity for simulation of laboratory-scale stress-strain behavior in granitic rocks. Computers and Geotechnics 118: 103363. https://doi.org/10.1016/j.compgeo.2019.103363.
  25. Stead D, Wolter A (2015) A critical review of rock slope failure mechanisms: The importance of structural geology. Journal of Structural Geology 74: 1–23.http://dx.doi.org/10.1016/j.jsg.2015.02.002.
  26. Tawadrous AS, DeGagné D, Pierce M, Mas Ivars D (2009) Prediction of uniaxial compression PFC3D model micro-properties using artificial neural networks. International Journal for Numerical and Analytical Methods in Geomechanics 33 (18): 1953–1962. https://doi.org/10.1002/nag.809.
  27. Turichshev A, Hadjigeorgiou J (2017) Development of synthetic rock mass bonded block models to simulate the behaviour of intact veined rock. Geotechnical and Geological Engineering 35 (1): 313–335. https://doi.org/10.1007/s10706-016-0108-5.
  28. Westbrink F, Elbel A, Schwung A, Ding SX (2021) Optimization of DEM parameters using multi-objective reinforcement learning. Powder Technology 379: 602–616. https://doi.org/10.1016/j.powtec.2020.10.067.
  29. Zhai S, Zhan J, Ba Y, Chen J, Li Y, et al. (2019) PFC model parameter calibration using uniform experimental design and a deep learning network. IOP Conference Series: Earth and Environmental Science 304 (3): 032062. https://doi.org/10.1088/1755-1315/304/3/032062.