All Issue

2025 Vol.41, Issue 5 Preview Page
Deep Learning-based Prediction of the Axial Bearing Capacity of PHC Auger-drilled Piles Using Dynamic Load Test Data

딥러닝 기반 동재하시험 자료를 활용한 PHC 매입말뚝의 연직지지력 예측

Kyu-Sun Kim1*
김 규선1*
1Member, Principal Professional, Construction Technology Team, Samsung C&T Corp.
1정회원, 삼성물산㈜ 건설부문 기반기술팀 프로
*Corresponding Author
31 October 2025. pp. 143-155
PDF XML Citation
Abstract

In this study, a multi-output regression deep learning model that predicts the axial bearing capacity of PHC auger-drilled piles was developed using the SDA method by decomposing it into shaft friction and toe bearing capacity. A dataset of 2,457 dynamic load tests conducted under various conditions in Korea was used to train the deep learning model. The model was constructed with eight input variables, including pile specifications, ground conditions, driving parameters, and construction conditions, and three output variables: shaft friction, toe bearing capacity, and total bearing capacity. The accuracy of the trained deep learning model was R2=0.786 for shaft friction, R2=0.740 for the toe bearing capacity, and R2=0.825 for the total bearing capacity. Analyses of the influence of input variables on the output variables revealed that elapsed time and pile length were the dominant factors for shaft friction, elapsed time and pile diameter for the toe bearing capacity, and pile diameter and elapsed time for the total bearing capacity. The deep learning model, capable of efficiently predicting the bearing capacity characteristics of PHC auger-drilled piles exhibiting complex behavioral patterns, is expected to assist on-site construction managers in conveniently predicting suitable driving parameters and testing schedules that satisfy quality requirements, thereby facilitating effective auger-drilled pile construction and quality management.

Keywords
Deep learning
Dynamic load test
PHC auger-drilled pile
Axial bearing capacity

본 연구에서는 SDA 공법으로 시공된 PHC 매입말뚝의 연직지지력을 주면마찰력과 선단지지력으로 분리하여 예측하는 다중출력 회귀 딥러닝 모델을 개발하였다. 국내 다양한 조건에서 수집된 2,457개의 동재하시험 데이터셋을 딥러닝 학습에 이용하였으며, 말뚝제원, 지반조건, 항타조건, 시공조건 등의 8개 입력변수와 주면마찰력, 선단지지력, 총지지력의 3개 출력변수로 모델을 구성하였다. 학습된 딥러닝 모델의 정확도는 각각 주면마찰력 R2=0.786, 선단지지력 R2=0.740, 총지지력 R2=0.825로 나타났으며, 출력변수에 대한 입력변수의 영향성 분석결과, 주면마찰력은 경과시간과 말뚝길이, 선단지지력은 경과시간과 말뚝직경, 총지지력은 말뚝직경과 경과시간이 주요 영향인자로 파악되었다. 복잡한 거동을 나타내는 매입말뚝의 지지력 특성을 효율적으로 예측할 수 있는 딥러닝 모델은 현장 시공관리자가 목표 품질수준에 부합하는 항타조건 및 시험시기를 간편하게 예측하여, 효과적인 매입말뚝 시공 및 품질관리를 가능하게 할 것으로 판단된다.

References
1

Benbouras, M. A., Petrişor, A.-I., Zedira, H., Ghelani, L., and Lefilef, L. (2021), “Forecasting the Bearing Capacity of the Driven Piles Using Advanced Machine-learning Techniques”, Applied Sciences, Vol.11, No.22, pp.10908, https://doi.org/10.3390/app112210908.

10.3390/app112210908
2

Chai, S.-G. (2007), Bearing Capacity of SDA Augered Piles in Various Grounds Depended on Water-Cement Ratio of Cement Milk, Ph.D. Dissertation, Chung-Ang University, Seoul, Korea (in Korean).

3

Cho, C.-W. (2003), “The Characteristics of the Set-up Effect of Driven Piles”, Journal of the Korean Geotechnical Society, Vol.19, No.4, pp.235-246 (in Korean).

4

Cho, C.-W. (2023), Piling Engineering Practice, 2nd Ed., Apub Press, Seoul, Korea (in Korean).

5

Hong, W.-P. and Chai, S.-G. (2007), “Estimation of Frictional Capacity of SDA Augered Piles in Various Grounds”, Journal of the Korean Society of Civil Engineers, Vol.27, No.4C, pp.279-292 (in Korean).

6

Khan, A., Khan, M., Khan, W. A., Afridi, M. A., Naseem, K. A., and Noreen, A. (2025), “Predicting Pile Bearing Capacity Using Gene Expression Programming with SHapley Additive exPlanation Interpretation”, Discover Civil Engineering, Vol.2, pp.58, https://doi.org/10.1007/s44290-025-00215-x.

10.1007/s44290-025-00215-x
7

Lee, I.-M. and Lee, J.-H. (1996), “Prediction of Pile Bearing Capacity Using Artificial Neural Networks”, Computers and Geotechnics, Vol.18, No.3, pp.189-200, https://doi.org/10.1016/0266-352X(95)00027-8.

10.1016/0266-352X(95)00027-8
8

Lee, S. and Chang, J.-W. (2006), “Evaluation of Bearing Capacity on PHC Auger-drilled Piles Using Artificial Neural Network”, Journal of The Korea Institute for Structural Maintenance and Inspection, Vol.10, No.6, pp.213-221 (in Korean), https://doi.org/10.11112/jksmi.2006.10.6.213.

10.11112/jksmi.2006.10.6.213
9

Lee, W, Lee, I.-M., Choi, Y., Lee, J.-H., and Kim, B.-C. (1994), “Construction Control and Analysis of Pile Foundation with Pile Driving Analyzer”, KGS Pile Foundation Technical Committee Seminar, pp.110-131 (in Korean).

10

Li, S., Hai, M., Zhang, Q., Zhou, B., Wang, M., and Zhao, Z. (2025), “Study on an Interpretable Prediction Model for Pile Bearing Capacity based on SHAP and BP Neural Networks”, Scientific Reports, Vol.15, pp.28134, https://doi.org/10.1038/s41598-025-13616-w.

10.1038/s41598-025-13616-w40750986PMC12316844
11

Lundberg, S. M. and Lee, S.-I. (2017), “A Unified Approach to Interpreting Model Predictions”, 31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, CA, USA.

12

Maizir, H. and Suryanita, R. (2018), “Evaluation of Axial Pile Bearing Capacity based on Pile Driving Analyzer (PDA) Test Using Neural Network”, IOP Conference Series: Earth and Environmental Science, Vol.106, pp.012037, https://doi.org/10.1088/1755-1315/106/1/012037.

10.1088/1755-1315/106/1/012037
13

Nguyen, T. T., Nguyen, K. L., Huynh, T. Q., and Tran, Q. (2025), “Influence of Feature Selection on Machine Learning Prediction of Pile Foundation – The Role of Soil-pile Interaction Knowledge and Application to Base Resistance”, Geodata and AI, Vol.3, pp.100019, https://doi.org/10.1016/j.geoai.2025.100019.

10.1016/j.geoai.2025.100019
14

Onyelowe, K. C., Hanandeh, S., Kamchoom, V., Ebid, A. M., Silva, F. D., Palta, J. L., Llamuca, J. L., and Avudaiappan, S. (2025), “Developing Advanced Datadriven Framework to Predict the Bearing Capacity of Piles on Rock”, Scientific Reports, Vol.15, pp.11051, https://doi.org/10.1038/s41598-025-96186-1.

10.1038/s41598-025-96186-140169682PMC11961753
15

Park, H.-I., Seok, J.-W., Hwang, D.-J., and Cho, C.-W. (2006), “A Study on Optimized Artificial Neural Network Model for the Prediction of Bearing Capacity of Driven Piles”, Journal of the Korean Geotechnical Society, Vol.22, No.6, pp.15-26 (in Korean), https://doi.org/10.7843/kgs.2006.22.6.15.

10.7843/kgs.2006.22.6.15
16

Park, J. (2017), “A Comparative Study on the Bearing Capacity of Dynamic Load Test and Static Load Test of PHC Bored Pile”, Journal of the Korean GEO-Environmental Society, Vol.18, No.9, pp.19-31 (in Korean), https://doi.org/10.14481/jkges.2017.18.9.19.

10.14481/jkges.2017.18.9.19
17

Pham, T. A., Ly, H.-B., Tran, V. Q., Giap, L. V., Vu, H.-L. T., and Duong, H.-A. T. (2020), “Prediction of Pile Axial Bearing Capacity Using Artificial Neural Network and Random Forest”, Applied Sciences, Vol.10, No.5, pp.1871, https://doi.org/10.3390/app10051871.

10.3390/app10051871
18

Puri, N., Prasad, H. D., and Jain, A. (2018), “Prediction of Geotechnical Parameters Using Machine Learning Techniques”, Procedia Computer Science, Vol.125, pp.509-517, https://doi.org/10.1016/j.procs.2017.12.066.

10.1016/j.procs.2017.12.066
19

Rausche, F., Goble, G., and Likins, G., Jr. (1985), “Dynamic Determination of Pile Capacity”, Journal of Geotechnical Engineering, Vol.111, No.3, pp.367-1383, https://doi.org/10.1061/(ASCE)0733-9410(1985)111:3(367).

10.1061/(ASCE)0733-9410(1985)111:3(367)
20

Seo, M. J., Park, J.-B., Park, M.-C., and Lee, J.-S. (2020), “Estimation of Load-settlement Curves of Embedded Piles Combining Results of End of Initial Driving and Restrike Dynamic Pile Tests”, Journal of the Korean Geotechnical Society, Vol.36, No.7, pp.15-28 (in Korean), https://doi.org/10.7843/kgs.2020.36.7.15.

10.7843/kgs.2020.36.7.15
21

Seo, S., Kim, G., Park, J.-B., Kim, J., Park, Y.-B., and Chung, M. (2025), “Leveraging Data-driven Machine Learning Techniques to Enhance Bearing Capacity Estimation in Prebored and Precast Piles”, Expert Systems with Applications, Vol.285, pp.128070, https://doi.org/10.1016/j.eswa.2025.128070.

10.1016/j.eswa.2025.128070
22

Shahin, M. A. (2010), “Intelligent Computing for Modeling Axial Capacity of Pile Foundations”, Canadian Geotechnical Journal, Vol.47, No.12, pp.1383-1394, https://doi.org/10.1139/T09-094.

10.1139/T09-094
23

Shahin, M. A. (2016), “State-of-the-art Review of Some Artificial Intelligence Applications in Pile Foundations”, Geoscience Frontiers, Vol.7, No.1, pp.33-44, https://doi.org/10.1016/j.gsf.2014.10.002.

10.1016/j.gsf.2014.10.002
24

Udengaard, J. T., Skouboe, E., Sørensen, K.K., Alibrandi, U., and Andersen, L.V. (2024), “Pile Capacity Prediction based on Machine Learning Using Support Vector Regression”, 19th Nordic Geotechnical Meeting, Gothenburg, Sweden.

Information
  • Publisher :The Korean Geotechnical Society
  • Publisher(Ko) :한국지반공학회
  • Journal Title :Journal of the Korean Geotechnical Society
  • Journal Title(Ko) :한국지반공학회 논문집
  • Volume : 41
  • No :5
  • Pages :143-155
  • Received Date : 2025-08-28
  • Revised Date : 2025-10-14
  • Accepted Date : 2025-10-15
  • DOI :https://doi.org/10.7843/kgs.2025.41.5.143
Journal Informaiton Journal of the Korean Geotechnical Society Journal of the Korean Geotechnical Society
Archives
: Vol.23~28 (2007~2012)
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close