A CNN deep learning approach to scour depth estimation around complex bridge piers in steady flow environments

Abstract

Scour around complex bridge piers (CBP) caused by sediment erosion due to steady flow is a critical challenge in hydraulic engineering, often leading to structural instabilities and failures. The accurate estimation of maximum scour depth is crucial for ensuring bridge safety and optimizing design. Traditional empirical methods and physics-based models, while widely utilized, often struggle to capture the complex interactions between hydrodynamic forces, sediment transport, and varying pier geometries, leading to conservative or inaccurate predictions. This study presents a one-dimensional convolutional neural network (1D CNN) and long short-term memory (LSTM) deep learning models for predicting the maximum scour depth around CBP under steady current conditions in a clear-water environment. The proposed model leverages the ability of 1D CNNs to process high-dimensional input dataset, capturing intricate non-linear relationships between influential parameters, such as flow velocity, pier configurations, sediment properties, and water depth. The dataset was transformed into non-dimensional forms using the Buckingham Pi theorem to enhance model generalization. The 1D CNN model was trained and validated using an extensive dataset, and its performance was benchmarked against established empirical models, including FDOT, HEC-18, and Coleman’s equation. Results show that the proposed 1D CNN model significantly outperforms traditional approaches, achieving higher coefficient of determination (R ² = 0.85) values and lower root mean squared error (RMSE = 0.1125), mean absolute error (MAE = 0.1078), and scatter index (SI = 0.1149). Moreover, the model's bias (B = -0.0194) and standard error (SE = 0.1147) remain minimal across unseen datasets, demonstrating robust predictive capability. This research highlights the potential of deep learning as a reliable and precise tool for scour depth prediction, contributing to improved risk assessment and sustainable bridge design under steady flow environments.

Keywords

Complex bridge pier / 1D CNN / LSTM / Maximum scour depth

References

[1] AkhlaghiE, BabarsadMS, DerikvandE, AbediniM. Assessment the effects of different parameters to rate scour around single piers and pile groups: a review. Arch Comput Methods Eng, 2020, 27(1): 183-197

[2] AlemiM, PêgoJP, MaiaR. Numerical simulation of the turbulent flow around a complex bridge pier on the scoured bed. Eur J Mech B Fluids, 2019, 76: 316-331

[3] Al-JubouriM, RayRP, AbbasEH. Advanced numerical simulation of scour around bridge piers: effects of pier geometry and debris on scour depth. J Marine Sci Eng, 2024, 12(9): 1637

[4] AlyAM, DoughertyE. Bridge pier geometry effects on local scour potential: a comparative study. Ocean Eng, 2021, 234, ArticleID: 109326

[5] AminiA, MelvilleBW, AliTM. Local scour at piled bridge piers including an examination of the superposition method. Can J Civ Eng, 2014, 41(5): 461-471

[6] AminpourM, AlaieR, KhosraviS, KardaniN, MoridpourS, NazemM. Slope stability machine learning predictions on spatially variable random fields with and without factor of safety calculations. Comput Geotech, 2023, 153, ArticleID: 105094

[7] Arinta RR, Andi EWR (2019) Natural disaster application on big data and machine learning: a review. 2019 4th International Conference on Information Technology, Information Systems and Electrical Engineering, ICITISEE 2019, 249–254. https://doi.org/10.1109/ICITISEE48480.2019.9003984

[8] Arneson LA, Zevenbergen LW, Lagasse PF, Clopper PE, Associates A (2012) Evaluating Scour at Bridges: Fifth Edition. https://doi.org/10.21949/1503647

[9] Ataie-Ashtiani B, Baratian-Ghorghi Z, Beheshti AA (2010) Experimental investigation of clear-water local scour of compound piers. J Hydraulic Eng 136(6):343–351. https://doi.org/10.1061/(ASCE)0733-9429(2010)136:6(343)

[10] BaghbaniA, ChoudhuryT, CostaS, ReinerJ. Application of artificial intelligence in geotechnical engineering: a state-of-the-art review. Earth Sci Rev, 2022, 228, ArticleID: 103991

[11] BarrieA, WangC, LiangF, QiW. Experimental investigation on the mechanism of local scour around a cylindrical coastal pile foundation considering sloping bed conditions. Ocean Eng, 2024, 312, ArticleID: 119225

[12] BeheshtiAA, Ataie-AshtianiB. Experimental study of three-dimensional flow field around a complex bridge pier. J Eng Mech, 2010, 136(2): 143-154

[13] BeheshtiAA, Ataie-AshtianiB. Scour hole influence on turbulent flow field around complex bridge piers. Flow Turbul Combust, 2016, 97(2): 451-474

[14] CaiK, HeJ, LiQ, ShangguanW, LiL, HuH. Meta-LSTM in hydrology: advancing runoff predictions through model-agnostic meta-learning. J Hydrol, 2024, 639, ArticleID: 131521

[15] ChambelJ, Fazeres-FerradosaT, MirandaF, BentoAM, Taveira-PintoF, LomonacoP. A comprehensive review on scour and scour protections for complex bottom-fixed offshore and marine renewable energy foundations. Ocean Eng, 2024, 304, ArticleID: 117829

[16] ChenJ, QuY, ZhaoX, SunZ. Protection mechanisms, countermeasures, assessments and prospects of local scour for cross-sea bridge foundation: a review. Ocean Eng, 2023, 288, ArticleID: 116145

[17] Chencho Li J, Hao H (2024) Structural damage quantification using long short-term memory (LSTM) auto-encoder and impulse response functions. J Infrastructure Intelligence Resilience 3(2). https://doi.org/10.1016/J.IINTEL.2024.100086

[18] ChoiSU, ChoiB. Prediction of time-dependent local scour around bridge piers. Water Environ J, 2016, 30(1–2): 14-21

[19] ColemanSE. Clearwater local scour at complex piers. J Hydraul Eng, 2005, 131(4): 330-334

[20] DengL, CaiCS. Bridge scour: prediction, modeling, monitoring, and countermeasures—review. Pract Period Struct des Constr, 2010, 15(2): 125-134

[21] DeyS, RaikarRV. Characteristics of horseshoe vortex in developing scour holes at piers. J Hydraul Eng, 2007, 133(4): 399-413

[22] Díaz-CarrascoP, CroquerS, TamimiV, LaceyJ, PoncetS. Advances in numerical reynolds-averaged navier-stokes modelling of wave-structure-seabed interactions and scour. J Marine Sci Eng, 2021, 9(6): 611

[23] DipietrangeloF, NicassioF, ScarselliG. Structural health monitoring for impact localisation via machine learning. Mech Syst Signal Process, 2023, 183, ArticleID: 109621

[24] DuS, WuG, ZhuDZ, WangR, LuY, LiangB. Experimental study of local scour around submerged square piles in combined waves and current. Ocean Eng, 2022, 266, ArticleID: 113176

[25] DuttaD, AfzalMS. Scour around twin-piles under combined wave–current flows. Coast Eng, 2024, 189, ArticleID: 104477

[26] EiniN, JanizadehS, BateniSM, JunC, HeggyE, KirsM. Predicting equilibrium scour depth around non-circular bridge piers with shallow foundations using hybrid explainable machine learning methods. Results Eng, 2024, 24, ArticleID: 103492

[27] FerraroD, TafarojnoruzA, GaudioR, CardosoAH. Effects of pile cap thickness on the maximum scour depth at a complex pier. J Hydraul Eng, 2013, 139(5): 482-491

[28] Fletcher DF, McClure DD, Kavanagh JM, Barton GW (2017) CFD simulation of industrial bubble columns: numerical challenges and model validation successes. Appl Math Model 44:25–42. https://doi.org/10.1016/j.apm.2016.08.033

[29] Freitas CJ (2002) The issue of numerical uncertainty. Appl Math Model 26(2):237–248. https://doi.org/10.1016/S0307-904X(01)00058-0

[30] GaoW, GeS. A comprehensive review of slope stability analysis based on artificial intelligence methods. Expert Syst Appl, 2024, 239, ArticleID: 122400

[31] Hashem T, Yousefpour N (2024) Application of Long-Short Term Memory and Convolutional Neural Networks for Real-Time Bridge Scour Prediction. https://arxiv.org/abs/2404.16549v2

[32] HassanWH, HusseinHH, AlshammariMH, JalalHK, RasheedSE. Evaluation of gene expression programming and artificial neural networks in PyTorch for the prediction of local scour depth around a bridge pier. Results Eng, 2022, 13, ArticleID: 100353

[33] HochreiterS, SchmidhuberJ. Long short-term memory. Neural Comput, 1997, 9(8): 1735-1780

[34] HoffmansG, VerheijH. Jet scour. Marit Eng, 2015, 164(4): 185-193

[35] HongJH, GoyalMK, ChiewYM, ChuaLHC. Predicting time-dependent pier scour depth with support vector regression. J Hydrol, 2012, 468–469: 241-248

[36] HongS, KoSJ, WooSI, KwakTY, KimSR. Time-series forecasting of consolidation settlement using LSTM network. Appl Intell, 2024, 54(2): 1386-1404

[37] HosseiniSH, MovahednejadMH, RohaniA. Predicting the scouring depth around a rectangular pier with a round nose by artificial intelligence methods. Eng Appl Artif Intell, 2024, 133, ArticleID: 108649

[38] HouS, WeiJ, HouM, XuJ, HanL. A hydrological knowledge-informed LSTM model for monthly streamflow reconstruction using distributed data: Application to typical rivers across the Tibetan plateau. J Hydrol, 2025, 649, ArticleID: 132409

[39] KhosronejadA, KangS, SotiropoulosF. Experimental and computational investigation of local scour around bridge piers. Adv Water Resour, 2012, 37: 73-85

[40] KimT, ShahriarAR, LeeWD, GabrMA. Interpretable machine learning scheme for predicting bridge pier scour depth. Comput Geotech, 2024, 170, ArticleID: 106302

[41] KiranyazS, AvciO, AbdeljaberO, InceT, GabboujM, InmanDJ. 1D convolutional neural networks and applications: a survey. Mech Syst Signal Process, 2021, 151, ArticleID: 107398

[42] KumarS, GoyalMK, DeshpandeV, AgarwalM. Estimation of time dependent scour depth around circular bridge piers: application of ensemble machine learning methods. Ocean Eng, 2023, 270, ArticleID: 113611

[43] KutlerP. A perspective of theoretical and applied computational fluid dynamics. AIAA J, 2012, 23(3): 328-341

[44] LaiYG, LiuX, BombardelliFA, SongY. Three-dimensional numerical modeling of local scour: a state-of-the-art review and perspective. J Hydraul Eng, 2022, 148(11): 03122002

[45] LeeSO, SturmTW. Effect of sediment size scaling on physical modeling of bridge pier scour. J Hydraul Eng, 2009, 135(10): 793-802

[46] LeeIB, BitogJPP, HongSW, SeoIH, KwonKS, BartzanasT, KaciraM. The past, present and future of CFD for agro-environmental applications. Comput Electron Agric, 2013, 93: 168-183

[47] LiX, van PaassenL, TaoJ. Investigation of using mangrove-inspired skirt pile group as a scour countermeasure. Ocean Eng, 2022, 266, ArticleID: 113133

[48] LiG, ZhuH, JianH, ZhaW, WangJ, ShuZ, YaoS, HanH. A combined hydrodynamic model and deep learning method to predict water level in ungauged rivers. J Hydrol, 2023, 625, ArticleID: 130025

[49] LiS, YangJ, HeX. Modeling transient flow dynamics around a bluff body using deep learning techniques. Ocean Eng, 2024, 295, ArticleID: 116880

[50] LinY, LinC. Scour effects on lateral behavior of pile groups in sands. Ocean Eng, 2020, 208, ArticleID: 107420

[51] LinkO, HenríquezS, EttmerB. Physical scale modelling of scour around bridge piers. J Hydraul Res, 2019, 57(2): 227-237

[52] LiuF, LiJ, WangL. PI-LSTM: physics-informed long short-term memory network for structural response modeling. Eng Struct, 2023, 292, ArticleID: 116500

[53] MaH, LiB, ZhangS. Scour mechanism around a pipeline under different current-wave conditions using the CFD-DEM coupling model. Comput Geotech, 2024, 170, ArticleID: 106304

[54] Martin-VideJP, HidalgoC, BatemanA. Local scour at piled bridge foundations. J Hydraulic Eng, 1998, 124(4): 439-444

[55] MarulasiddappaSB, PatilAP, KuntojiG, PraveenKM, NagannaSR. Prediction of scour depth around bridge abutments using ensemble machine learning models. Neural Comput Appl, 2024, 36(3): 1369-1380

[56] Melville BW, Coleman SE (2000) Bridge scour. 550. https://books.google.com/books/about/Bridge_Scour.html?hl=vi&id=SoIA3P1zLAEC

[57] MerghadiA, YunusAP, DouJ, WhiteleyJ, ThaiPhamB, BuiDT, AvtarR, AbderrahmaneB. Machine learning methods for landslide susceptibility studies: a comparative overview of algorithm performance. Earth Sci Rev, 2020, 207, ArticleID: 103225

[58] MohammedTA, NoorMJMM, GhazaliAH, YusufB, SaedK. Physical modeling of local scouring around bridge piers in erodable bed. J King Saud Univ - Engineering Sciences, 2007, 19(2): 195-206

[59] MorenoM, MaiaR, CoutoL. Effects of relative column width and pile-cap elevation on local scour depth around complex piers. J Hydraul Eng, 2016, 142(2): 04015051

[60] Nandi Mr B, Das Dr S (2025) Developing new equations for maximum scour depth near tandem, side-by-side, and eccentric piers. Can J Civil Eng 1–15. https://doi.org/10.1139/CJCE-2024-0373

[61] NandiB, DasS. Equation for time-dependent local scour at pier-like structures with eccentric in-line arrangements. Proc Inst Civil Eng Water Manag, 2023, 177(6): 361-374

[62] NandiB, DasS. Identify most promising temporal scour depth formula for circular piers proposed over last six decades. Ocean Eng, 2023, 286, ArticleID: 115639

[63] NandiB, PatelG, DasS. Prediction of maximum scour depth at clear water conditions: multivariate and robust comparative analysis between empirical equations and machine learning approaches using extensive reference metadata. J Environ Manage, 2024, 354, ArticleID: 120349

[64] NandiB, AsceAM, DasS. Predicting max scour depths near two-pier groups using ensemble machine-learning models and visualizing feature importance with partial dependence plots and SHAP. J Comput Civ Eng, 2025, 39(2): 04025007

[65] NandiB, DasS, PaulS. Estimation of clear water flow induced maximum scour depth using random forest and XGBoost. Lecture Notes in Mechanical Engineering, 2025, 81–95: 81

[66] NguyenHHD, PradhanAMS, SongCH, LeeJS, KimYT. A hybrid approach combining physics-based model with extreme value analysis for temporal probability of rainfall-triggered landslide. Landslides., 2024, 22(1): 149-168

[67] NguyenTN, YunDH, KimYT. Development of an empirical equation that predicts local scour around a single cube-type artificial reef under a steady current. Ocean Eng, 2024, 308, ArticleID: 118332

[68] NguyenTN, YunDH, KimYT. Measurement and prediction of scour volume around a cubic artificial reef under steady flow conditions using stereo vision. Ocean Eng, 2024, 313, ArticleID: 119635

[69] NguyenHHD, NguyenTN, PhanTAT, NguyenGP, PhamTKH, HuynhNT. Uncertainty effect of saturated depth on large scale assessment of landslide: case study at Saka town, Hiroshima Prefecture. Japan Earth Science Informatics, 2025, 18(2): 1-14

[70] NguyenH-H-D, NguyenT-N, PhanT-A-T, HuynhN-T, HuynhQ-D, TrieuT-T. Explainable artificial intelligence model for the prediction of undrained shear strength. Theor Appl Mech Lett, 2025, 15(3) ArticleID: 100578

[71] NiX, XueL, AnC. Experimental investigation of scour around circular arrangement pile groups. Ocean Eng, 2021, 219, ArticleID: 108096

[72] OğuzK, BorA. Prediction of local scour around bridge piers using hierarchical clustering and adaptive genetic programming. Applied Artificial Intelligence, 2022, 36 1) ArticleID: 2001734

[73] PandeyM, KarbasiM, JameiM, MalikA, PuJH. A comprehensive experimental and computational investigation on estimation of scour depth at bridge abutment: emerging ensemble intelligent systems. Water Resour Manage, 2023, 37 9): 3745-3767

[74] PanfengB, SonglinZ, HongyuC, CaiweiL, PengtaoW, LichangQ. Structural monitoring data repair based on a long short-term memory neural network. Sci Rep, 2024, 14(1): 1-18

[75] PhamMV, KimYT. Debris flow detection and velocity estimation using deep convolutional neural network and image processing. Landslides, 2022, 19(10): 2473-2488

[76] PhamMV, HaYS, KimYT. Automatic detection and measurement of ground crack propagation using deep learning networks and an image processing technique. Measurement, 2023, 215, ArticleID: 112832

[77] PhamMV, SongCH, NguyenTN, LeeJS, KimYT. Real-time debris flow detection using deep convolutional neural network and Jetson Nano. E3S Web of Conferences, 2023, 415, ArticleID: 03021

[78] PhamM-V, KimY-T, HaY-S. Displacement measurement and 3D reconstruction of segmental retaining wall using deep convolutional neural networks and binocular stereovision. Struct Control Health Monit, 2024, 2024(1) ArticleID: 9912238

[79] PhamNL, TaQB, KimJT. Pseudo-damage simulation and CNN deep learning for damage identification of submerged structure-foundation systems. Struct Health Monit, 2024

[80] PizarroA, ManfredaS, TubaldiE. The science behind scour at bridge foundations: a review. Water., 2020, 12 2): 374

[81] PourghasemiHR, KariminejadN, AmiriM, EdalatM, ZarafsharM, BlaschkeT, CerdaA. Assessing and mapping multi-hazard risk susceptibility using a machine learning technique. Scientific Reports., 2020, 10(1): 1-11

[82] RodriguesM, MiguéisVL, FelixC, RodriguesC. Machine learning and cointegration for structural health monitoring of a model under environmental effects. Expert Syst Appl, 2024, 238, ArticleID: 121739

[83] ShahriarAR, OrtizAC, MontoyaBM, GabrMA. Bridge Pier Scour: an overview of factors affecting the phenomenon and comparative evaluation of selected models. Transportation Geotechnics, 2021, 28, ArticleID: 100549

[84] SharafatiA, HaghbinM, MottaD, YaseenZM. The application of soft computing models and empirical formulations for hydraulic structure scouring depth simulation: a comprehensive review, assessment and possible future research direction. Archives of Computational Methods in Engineering., 2019, 28(2): 423-447

[85] Sheppard DM, Renna R (2005) Bridge scour manual. Florida Department of Transportation, Florida. https://www.fdot.gov/docs/default-source/roadway/drainage/bridgescour/FDOT-Scour-Manual.pdf

[86] TehraniFS, CalvelloM, LiuZ, ZhangL, LacasseS. Machine learning and landslide studies: recent advances and applications. Natural Hazards, 2022, 114(2): 1197-1245

[87] Thanh-Nhan N, Thanh-Hai D, Thanh-Hoang P, Yun-Tae K (2023) Three-dimensional numerical modeling of flow-induced scour around a cubic Artificial Reef. International Conference on Civil and Environmental Engineering, 19th (Web), 329–332. https://jglobal.jst.go.jp/en/detail?JGLOBAL_ID=202402263971031709

[88] TolaS, TinocoJ, MatosJC, ObrienE. Scour detection with monitoring methods and machine learning algorithms—a critical review. Appl Sci, 2023, 13(3): 1661

[89] TruongHL, NguyenN, LindenberghR, FiskP, HuynhT. Towards a digital twin of a storage tank using laser scan data. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2021, XLVI-4-W4-2021(4/W4-2021): 119-124

[90] WangX, YeA, JiB. Fragility-based sensitivity analysis on the seismic performance of pile-group-supported bridges in liquefiable ground undergoing scour potentials. Eng Struct, 2019, 198, ArticleID: 109427

[91] WangC, YuQ, LawKH, McKennaF, YuSX, TacirogluE, ZsarnóczayA, ElhaddadW, CetinerB. Machine learning-based regional scale intelligent modeling of building information for natural hazard risk management. Autom Constr, 2021, 122, ArticleID: 103474

[92] Wang C, Wu Q, Liang J, Liang F, Yu X, (Bill). (2024a) Establishment and implementation of an artificial intelligent flume for investigating local scour around underwater foundations. Transportation Geotechnics 49:101433. https://doi.org/10.1016/J.TRGEO.2024.101433

[93] WangC, YangY, ChangJ, CaiG, HeH, WuM, LiuS. Prediction of the consolidation coefficient of soft soil based on machine learning models. Soil Mech Found Eng, 2024, 61(3): 223-229

[94] WengangZ, HanlongL, LinW, XingZ, YanmeiZ. Application of machine learning in slope stability assessment. Application of Machine Learning in Slope Stability Assessment, 2023, 1–201: 1

[95] XuG, JiC, XuY, YuE, CaoZ, WuQ, LinP, WangJ. Machine learning in coastal bridge hydrodynamics: a state-of-the-art review. Appl Ocean Res, 2023, 134, ArticleID: 103511

[96] YangY, ShaoD, WangY, DaiS. A data-driven approach to integrated equilibrium-temporal scour forecasting at complex-pier structures using hybrid neural networks. Ocean Eng, 2024, 303: 117739

[97] YanmazAM, AltinbilekHD. Study of time-depenbent local scour around bridge piers. J Hydraul Eng, 1991, 117(10): 1247-1268

[98] Yee-Meng ChiewB. Scour protection at bridge piers. J Hydraul Eng, 1992, 118(9): 1260-1269

[99] YuL, DaoPB, QiuL, UhlT, LeM-Q. Review of machine-learning techniques applied to structural health monitoring systems for building and bridge structures. Appl Sci, 2022, 12(21): 10754

[100] Zerouali B, Bailek N, Tariq A, Kuriqi A, Guermoui M, Alharbi AH, Khafaga DS, El-Sayed, & & El-Kenawy M (2024) Enhancing deep learning-based slope stability classification using a novel metaheuristic optimization algorithm for feature selection. Sci Rep. 14(1):1–19. https://doi.org/10.1038/s41598-024-72588-5

[101] ZhangP, YinZY, ZhengY, GaoFP. A LSTM surrogate modelling approach for caisson foundations. Ocean Eng, 2020, 204, ArticleID: 107263

[102] ZhangN, ShenSL, ZhouA, JinYF. Application of LSTM approach for modelling stress–strain behaviour of soil. Appl Soft Comput, 2021, 100, ArticleID: 106959

[103] ZhangD, WangD, PengQ, LinJ, JinT, YangT, SorooshianS, LiuY. Prediction of the outflow temperature of large-scale hydropower using theory-guided machine learning surrogate models of a high-fidelity hydrodynamics model. J Hydrol, 2022, 606, ArticleID: 127427

[104] ZhangF, ChenX, YanJ, GaoX. Countermeasures for local scour around offshore wind turbine monopile foundations: a review. Appl Ocean Res, 2023, 141, ArticleID: 103764

[105] ZhangW, GuX, HongL, HanL, WangL. Comprehensive review of machine learning in geotechnical reliability analysis: algorithms, applications and further challenges. Appl Soft Comput, 2023, 136, ArticleID: 110066

RIGHTS & PERMISSIONS

The Author(s)