Journal of the Korean Geotechnical Society (한국지반공학회논문집)

Korean Geotechnical Society (한국지반공학회)
권호 표지
DOI QR코드

DOI QR Code

Neural Network-Based Prediction of Dynamic Properties

인공신경망을 활용한 동적 물성치 산정 연구

  • Min, Dae-Hong (Dept. of Disaster Safety Engrg., Daejeon Univ.) ;
  • Kim, YoungSeok (Northern Infrastructure Specialized Team, Korea Institute of Civil Engrg. and Building Technology) ;
  • Kim, Sewon (Dept. of Geotechnical Engrg. Research, Korea Institute of Civil Engrg. and Building Technology) ;
  • Choi, Hyun-Jun (Northern Infrastructure Specialized Team, Korea Institute of Civil Engrg. and Building Technology) ;
  • Yoon, Hyung-Koo (Dept. of Disaster Safety Engrg., Daejeon Univ.)
  • 민대홍 (대전대학교 재난안전공학과) ;
  • 김영석 (한국건설기술연구원 북방인프라특화팀) ;
  • 김세원 (한국건설기술연구원 지반연구본부) ;
  • 최현준 (한국건설기술연구원 북방인프라특화팀) ;
  • 윤형구 (대전대학교 재난안전공학과)
  • Received : 2023.10.13
  • Accepted : 2023.10.26
  • Published : 2023.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

Dynamic soil properties are essential factors for predicting the detailed behavior of the ground. However, there are limitations to gathering soil samples and performing additional experiments. In this study, we used an artificial neural network (ANN) to predict dynamic soil properties based on static soil properties. The selected static soil properties were soil cohesion, internal friction angle, porosity, specific gravity, and uniaxial compressive strength, whereas the compressional and shear wave velocities were determined for the dynamic soil properties. The Levenberg-Marquardt and Bayesian regularization methods were used to enhance the reliability of the ANN results, and the reliability associated with each optimization method was compared. The accuracy of the ANN model was represented by the coefficient of determination, which was greater than 0.9 in the training and testing phases, indicating that the proposed ANN model exhibits high reliability. Further, the reliability of the output values was verified with new input data, and the results showed high accuracy.

동적 물성치는 지반의 상세한 거동을 예측하기 위한 필수인자이나, 샘플 채취와 추가적인 실험이 동반되는 한계가 있다. 본 연구의 목적은 정적 지반 물성치를 기반으로 동적 지반 물성치를 예측하는 것으로 인공신경망을 활용하고자 하였다. 정적 물성치는 점착력, 내부마찰각, 함수비, 비중 그리고 일축압축강도로 선정하였으며 출력 값인 동적물성치는 압축파 속도와 전단파 속도로 결정하였다. 인공신경망 적용시 결과값의 신뢰성을 높이기 위해 Levenberg-Marquardt와 Bayesian regularization 방법을 적용하였으며, 각 최적화 방법에 따른 신뢰성을 비교하였다. 인공신경망 모델의 정확도는 결정계수로 나타냈으며, train과 test 과정 모두 0.9 이상의 값을 보여 해당 연구에서 구축한 인공신경망의 신뢰성이 높은 것으로 나타났다. 또한, 구축된 인공신경망 모델의 검증을 위해 새로운 입력 데이터에 대해서도 출력값의 신뢰성을 검증하였으며, 그 결과 높은 정확도를 보였다.

Keywords

Acknowledgement

본 연구는 과학기술정보통신부의 한국연구재단 중견 연구(NRF-2020R1A2C2012113)와 국토교통부 국토교통과학기술진흥원 'AI 기반 가스·오일 플랜트 운영·유지 관리 핵심기술 개발(RS-2021-KA161932)' 사업의 지원으로 수행되었으며 이에 감사드립니다.

References

  1. Burden, F. and Winkler, D. (2009), Bayesian regularization of neural networks. Artificial neural networks: methods and applications, 23-42.
  2. Cho, N. J., Cha, W., and Kim, H. K. (2016), Non-destructive Detection of Underground Cavities Using Thermal Images. Electron, Journal of the Korean Geotechnical Society, Vol.21, No.16, pp. 5465-5476.
  3. Go, G. H. and Lee, S. J. (2021), A Study on Numerical Analysis for GPR Signal Characterization of Tunnel Lining Cavities, Journal of the Korean Geotechnical Society, Vol.37, No.10, pp.65-76. https://doi.org/10.7843/KGS.2021.37.10.65
  4. Gu, D., Shen, J., and Wang, X. (2018), Deep learning, Instructor.
  5. Hateffard, F., Dolati, P., Heidari, A., and Zolfaghari, A. A. (2019), Assessing the Performance of Decision Tree and Neural Network Models in Mapping Soil Properties, Journal of Mountain Science, Vol.16, No.8, pp.1833-1847. https://doi.org/10.1007/s11629-019-5409-8
  6. Hecht-Nielsen, R. (1992), Theory of the backpropagation neural network, In Neural networks for perception (pp.65-93), Academic Press.
  7. Jafari, M. K., Shafiei, A., and Razmkhah, A. (2002), Dynamic properties of fine grained soils in south of Tehran.
  8. Jang Byeong-su, Kim Young-seok, Kim Se-won, Choi Hyun-jun, and Yoon Hyeong-gu (2023), Evaluation of Underground Cavities Using Infrared Images, Journal of the Korean Society of Geotechnical Engineering, Vol.39, No.7, pp.69-76.
  9. Kalantari, A. R., Johari, A., Zandpour, M., and Kalantari, M. (2023), Effect of Spatial Variability of Soil Properties and Geostatistical Conditional Simulation on Reliability Characteristics and Critical Slip Surfaces of Soil Slopes, Transportation Geotechnics, 39, 100933.
  10. Ng, W., Minasny, B., Montazerolghaem, M., Padarian, J., Ferguson, R., Bailey, S., and McBratney, A. B. (2019), Convolutional Neural Network for Simultaneous Prediction of Several Soil Properties Using Visible/near-infrared, Mid-infrared, and their Combined Spectra, Geoderma, 352, pp.251-267. https://doi.org/10.1016/j.geoderma.2019.06.016
  11. Michel, C., Gueguen, P., Lestuzzi, P., and Bard, P. Y. (2010), Comparison between Seismic Vulnerability Models and Experimental Dynamic Properties of Existing Buildings in France, Bulletin of Earthquake Engineering, 8, pp.1295-1307. https://doi.org/10.1007/s10518-010-9185-7
  12. Menq, F. Y. (2003), Dynamic properties of sandy and gravelly soils, The University of Texas at Austin.
  13. Rainieri, C., Fabbrocino, G., and Cosenza, E. (2010), Some Remarks on Experimental Estimation of Damping for Seismic Design of Civil Constructions, Shock and Vibration, Vol.17, No.4-5, pp.383-395. https://doi.org/10.1155/2010/737452
  14. Ranganathan, A. (2004), The Levenberg-marquardt Algorithm, Tutoral on LM algorithm, Vol.11, No.1, pp.101-110.
  15. Rojas, R., and Rojas, R. (1996), The backpropagation algorithm, Neural networks: a systematic introduction, pp.149-182.
  16. Roweis, S. (1996), Levenberg-marquardt optimization, Notes, University Of Toronto, 52.
  17. Sapna, S., Tamilarasi, A., and Kumar, M. P. (2012), Backpropagation Learning Algorithm based on Levenberg Marquardt Algorithm, Comp Sci Inform Technol (CS and IT), 2, pp.393-398.  https://doi.org/10.5121/csit.2012.2438
  18. Yu, X. H., Chen, G. A., and Cheng, S. X. (1995), Dynamic Learning Rate Optimization of the Backpropagation Algorithm, IEEE Transactions on Neural Networks, Vol.6, No.3, pp.669-677. https://doi.org/10.1109/72.377972
  19. Wali, A. S. and Tyagi, A. (2020), Comparative Study of Advance Smart Strain Approximation Method Using Levenberg-marquardt and Bayesian Regularization Backpropagation Algorithm, Materials Today: Proceedings, 21, pp.1380-1395. https://doi.org/10.1016/j.matpr.2020.01.178
  20. Zhang, J. R., Zhang, J., Lok, T. M., and Lyu, M. R. (2007), A Hybrid Particle Swarm Optimization-back-propagation Algorithm for Feedforward Neural Network Training, Applied mathematics and computation, Vol.185, No.2, pp.1026-1037.  https://doi.org/10.1016/j.amc.2006.07.025