A regression-based approach to the prediction of crest settlement of embankment dams under earthquake shaking

Document Type : Article


Department of Civil Engineering, Shahrekord University, Shahrekord, Iran.


Settlement and deformation of the embankment dams are among the major damages caused by earthquakes, eventually leading to dam instability. Therefore, accurate assessment of the seismic settlement of embankment dams is of particular concern. This study aims to evaluate the settlement of embankment dams subjected to earthquake loads using regression-based methods. A wide-ranging real data on crest settlement of embankment dams caused by earthquakes was collected and analyzed. Yield acceleration of dam, maximum horizontal earthquake acceleration, fundamental period of dam body, predominant period of earthquake, and earthquake magnitude were considered as the most influential parameters affecting the seismic crest settlement of embankment dams. Using support vector regression method as well as multiple linear regression method, two models were developed to estimate earthquake induced crest settlement of embankment dams. Subsequently, sensitivity analysis was conducted in order to assess the behavior of the proposed models under different conditions. Finally, the accuracy of the proposed models was compared with the existing relationship for estimation of earthquake induced crest settlement of embankment dams. Although both MLR- and SVR-based models have an acceptable accuracy in estimation of the crest settlement of embankment dams under earthquake loading, the SVR-based model has a higher accuracy.


1. Meehan, C.L. and Vahedifard, F. Evaluation of  simpli_ed methods for predicting earthquake-induced  slope displacements in earth dams and embankments",  Engineering Geology, 152(1), pp. 180{193 (2013).  2. Newmark, N.M. E_ects of earthquakes on dams  and embankments", Geotechnique, 15(2), pp. 139{160  (1965).  3. Garini, E., Gazetas, G., and Anastasopoulos, I.  Asymmetric 'Newmark' sliding caused by motions  containing severe 'directivity' and 'ing' pulses",  Geotechnique, 61(9), pp. 733{756 (2011).  678 H. Javdanian et al./Scientia Iranica, Transactions A: Civil Engineering 27 (2020) 671{681  4. Makdisi, F.I. and Seed, H.B. Simpli_ed procedure for  estimating dam and embankment earthquake-induced  deformations", Journal of Geotechnical Engineering,  104(7), pp. 849{867 (1978).  5. Sarma, S.K. Seismic stability of earth dams and embankments",  Geotechnique, 25(4), pp. 743{761 (1975).  6. Bray, J.D., Macedo, J., and Travasarou, T. Simpli  _ed procedure for estimating seismic slope displacements  for subduction zone earthquakes", Journal  of Geotechnical and Geoenvironmental Engineering,  144(3), 04017124 (2017).  7. Hynes-Gri_n, M.E. and Franklin, A.G. Rationalizing  the seismic coe_cient method", Misc. Paper GL-84-13.  U.S. Army Waterway Experiment Station, Vicksburg,  Mississippi (1984).  8. Kramer, S.L. and Smith, M.W. Modi_ed Newmark  model for seismic displacements of compliant slopes",  Journal of Geotechnical and Geoenvironmental Engineering,  123(7), pp. 635{644 (1997).  9. Bray, J.D. and Rathje, E.M. Earthquake-induced  displacements of solid-waste land_lls", Journal of  Geotechnical and Geoenvironmental Engineering,  124(3), pp. 242{253 (1998).  10. Rathje, E.M. and Bray, J.D. Nonlinear coupled  seismic sliding analysis of earth structures", Journal  of Geotechnical and Geoenvironmental Engineering,  126(11), pp. 1002{1014 (2000).  11. Bray, J.D. and Travasarou, T. Simpli_ed procedure  for estimating earthquake-induced deviatoric slope displacements",  Journal of Geotechnical and Geoenvironmental  Engineering, 133(4), pp. 381{392 (2007).  12. Jibson, R.W. Regression models for estimating coseismic  landslide displacement", Engineering Geology,  91(2), pp. 209{218 (2007).  13. Ebrahimian, B. Numerical analysis of nonlinear dynamic  behavior of earth dams", Frontiers of Architecture  and Civil Engineering in China, 5(1), pp. 24{40  (2011).  14. Park, D.S. and Kim, N.R. Safety evaluation of cored  rock_ll dams under high seismicity using dynamic  centrifuge modeling", Soil Dynamics and Earthquake  Engineering, 97, pp. 345{363 (2017).  15. Kim, M.K., Lee, S.H., Choo, Y.W., and Kim, D.S.  Seismic behaviors of earth-core and concrete-faced  rock-_ll dams by dynamic centrifuge tests", Soil Dynamics  and Earthquake Engineering, 31(11), pp. 1579{  1593 (2011).  16. Singh, R., Roy, D., and Das, D. A correlation for  permanent earthquake-induced deformation of earth  embankments", Engineering Geology, 90(3), pp. 174{  185 (2007).  17. Singh, R. and Roy. D. Estimation of earthquakeinduced  crest settlements of embankments", American  Journal of Engineering and Applied Sciences, 2(3), pp.  515{525 (2009).  18. Jafarian, Y., Haddad, A., and Javdanian, H. Predictive  model for normalized shear modulus of cohesive  soils", Acta Geodynamica et Geomaterialia, 11(1), pp.  89{100 (2014).  19. Javdanian, H., Jafarian, Y., and Haddad, A. Predicting  damping ratio of _ne-grained soils using soft computing  methodology", Arabian Journal of Geosciences,  8(6), pp. 3959{3969 (2015).  20. Javdanian, H., Haddad, A., and Jafarian, A. Evaluation  of dynamic behavior of _ne-grained soils using  group method of data handling", Transportation Infrastructure  Engineering, 1(3), pp. 77{92 (2015).  21. Javdanian, H. Assessment of shear sti_ness ratio of  cohesionless soils using neural modeling", Modeling  Earth Systems and Environment, 3(3), pp. 1045{1053  (2017).  22. Javdanian, H. The E_ect of geopolymerization on  the uncon_ned compressive strength of stabilized _negrained  soils", International Journal of Engineering-  Transactions B: Applications, 30(11), pp. 1673{1680  (2017).  23. Javdanian, H. and Lee, S. Evaluating uncon_ned  compressive strength of cohesive soils stabilized with  geopolymer: a computational intelligence approach",  Engineering with Computers, 35(1), pp. 191{199  (2019).  24. Javdanian, H., Haddad, A., and Mehrzad, B. Experimental  and numerical investigation of the bearing  capacity of adjacent footings on reinforced soil", Electronic  Journal of Geotechnical Engineering, 17(R), pp.  2597{2617 (2012).  25. Javdanian, H. Evaluation of soil liquefaction potential  using energy approach: experimental and statistical  investigation", Bulletin of Engineering Geology and the  Environment (2017). DOI: 10.1007/s10064-017-1201-6  26. Yaghmaei-Sabegh, S. Earthquake ground-motion duration  estimation by using of general regression neural  network", Scientia Iranica, 25(5), pp. 2425-2439  (2017). DOI: 10.24200/sci.2017.4217  27. Najafzadeh, M. and Barani, G.A. Comparison of  group method of data handling based genetic programming  and back propagation systems to predict scour  depth around bridge piers", Scientia Iranica, 18(6),  pp. 1207{1213 (2011).  28. Najafzadeh, M., Barani, G.A., and Hessami-Kermani,  M.R. Group method of data handling to predict  scour depth around vertical piles under regular waves",  Scientia Iranica, 20(3), pp. 406{413 (2013).  29. Najafzadeh, M. and Lim, S.Y. Application of improved  neuro-fuzzy GMDH to predict scour depth at  sluice gates", Earth Science Informatics, 8(1), pp. 187{  196 (2015).  30. Najafzadeh, M., Shiri, J., and Rezaie-Balf, M. New  expression-based models to estimate scour depth at  clear water conditions in rectangular channels", Marine  Georesources & Geotechnology, 36(2), pp. 227{  235 (2018).  H. Javdanian et al./Scientia Iranica, Transactions A: Civil Engineering 27 (2020) 671{681 679  31. Najafzadeh, M., Tafarojnoruz, A., and Lim, S.Y. Prediction  of local scour depth downstream of sluice gates  using data-driven models", ISH Journal of Hydraulic  Engineering, 23(2), pp. 195{202 (2017).  32. Shooshpasha, I., Amiri, I., and MolaAbasi, H. An  investigation of friction angle correlation with geotechnical  properties for granular soils using GMDH type  neural networks", Scientia Iranica, 22(1), pp. 157{164  (2015).  33. Habibagahi, G. and Taherian, M. Prediction of collapse  potential for compacted soils using arti_cial  neural networks", Scientia Iranica, 11(1), pp. 1{20  (2004).  34. Kovacevic, M., Bajat, B., and Gajic, B. Soil type  classi_cation and estimation of soil properties using  support vector machines", Geoderma, 154(3), pp. 340{  347 (2010).  35. Pasolli, L., Notarnicola, C., and Bruzzone, L. Estimating  soil moisture with the support vector regression  technique", IEEE Geoscience and Remote Sensing  Letters, 8(6), pp. 1080{1084 (2011).  36. Lee, D., Kim, G., and Lee, K.E. Soil moisture  prediction using a support vector regression", Journal  of the Korean Data and Information Science Society,  24(2), pp. 401{408 (2013).  37. Elbisy, M.S. Support vector machine and regression  analysis to predict the _eld hydraulic conductivity of  sandy soil", KSCE Journal of Civil Engineering, 19(7),  pp. 2307{2316 (2015).  38. Soltani, N. and Bagheripour, M.H. Seismic wave  scatter study in valleys using coupled 2D _nite element  approach and absorbing boundaries", Scientia Iranica,  24(1), pp. 110{120 (2017).  39. Jafarian, Y., Javdanian, H., and Haddad, A. Straindependent  dynamic properties of Bushehr siliceouscarbonate  sand: Experimental and comparative  study", Soil Dynamics and Earthquake Engineering,  107, pp. 339{349 (2018).  40. Jafarian, Y., Javdanian, H., and Haddad, A. Dynamic  properties of calcareous and siliceous sands under  isotropic and anisotropic stress conditions", Soils and  Foundations, 58(1), pp. 172{184 (2018).  41. Gazetas, P. and Dakoulas, P. Seismic analysis and design  of rock_ll dams", Soil Dynamics and Earthquake  Engineering, 11, pp. 27{61 (1991).  42. Vapnik, V. An overview of statistical learning theory",  IEEE Transactions on Neural Networks, 10(5),  pp. 988{999 (1999).  43. Smola, A.J. and Scholkopf, B. A tutorial on support  vector regression", Statistics and Computing, 14(3),  pp. 199{222 (2004).  44. Ustun, B., Melssen. W.J., Oudenhuijzen, M., and  Buydens, L.M.C. Determination of optimal support  vector regression parameters by genetic algorithms and  simplex optimization", Analytica Chemica Acta, 544,  pp. 292{305 (2005).  45. Vapnik, V., The Nature of Statistical Learning Theory,  Springer Science & Business Media (2013).  46. Cherkassky, V. and Ma, Y. Practical selection of SVM  parameters and noise estimation for SVM regression",  Neural Networks, 17(1), pp. 113{126 (2004).  47. Zarif Sanayei, H.R., Talebbeydokhti, N., and Moradkhani,  H. 3D estimation of metal elements in sediments  of the Caspian Sea with moving least square and  radial basis function interpolation methods", Scientia  Iranica, 22(5), pp. 1661{1673 (2015).  48. Fattahi, H. Prediction of earthquake induced displacements  of slopes using hybrid support vector  regression with particle swarm optimization", Int J  Optim Civil Eng, 5(3), pp. 267{282 (2015).  49. Swaisgood, J.R. Embankment dam deformations  caused by earthquakes", In Paci_c Conference on  Earthquake Engineering, Paper No. 14 (2003).