Journal of the Korean Geosynthetics Society (한국지반신소재학회논문집)

Korean Geosynthetics Society (한국지반신소재학회)
권호 표지
DOI QR코드

DOI QR Code

Comparison of Error Rate and Prediction of Compression Index of Clay to Machine Learning Models using Orange Mining

오렌지마이닝을 활용한 기계학습 모델별 점토 압축지수의 오차율 및 예측 비교

  • Yoo-Jae Woong (Dept. of Civil Engineering, Korea Maritime and Ocean Univ.) ;
  • Woo-Young Kim (Dept. of Civil Engineering, Korea Maritime and Ocean Univ.) ;
  • Tae-Hyung Kim (Dept. of Civil Engrg., Korea Maritime and Ocean Univ.)
  • 유재웅 ;
  • 김우영 ;
  • 김태형
  • Received : 2024.07.01
  • Accepted : 2024.08.08
  • Published : 2024.09.30
Download PDF
⟨ Previous Next ⟩

Abstract

Predicting ground settlement during the improvement of soft ground and the construction of a structure is an crucial factor. Numerous studies have been conducted, and many prediction equations have been proposed to estimate settlement. Settlement can be calculated using the compression index of clay. In this study, data on water content, void ratio, liquid limit, plastic limit, and compression index from the Busan New Port area were collected to construct a dataset. Correlation analysis was conducted among the collected data. Machine learning algorithms, including Random Forest, Neural Network, Linear Regression, Ada Boost, and Gradient Boosting, were applied using the Orange mining program to propose compression index prediction models. The models' results were evaluated by comparing RMSE and MAPE values, which indicate error rates, and R2 values, which signify the models' significance. As a result, water content showed the highest correlation, while the plastic limit showed a somewhat lower correlation than other characteristics. Among the compared models, the AdaBoost model demonstrated the best performance. As a result of comparing each model, the AdaBoost model had the lowest error rate and a large coefficient of determination.

연약지반을 개량하고 그 위에 구조물을 시공하는 데 있어 지반 침하량을 예측하는 것은 매우 중요한 일이다. 침하량을 예측하기 위해 과거로부터 많은 연구들이 진행되었고 많은 예측 식이 제시되었다. 침하량은 점토의 압축지수를 통해 산정할 수 있다. 본 연구에서는 부산항 신항의 함수비, 간극비, 액성한계, 소성한계, 압축지수의 데이터를 수집하여 데이터 셋을 구축하고, 구축된 데이터 셋을 통해 각 데이터 사이의 상관분석을 실시하였다. 오렌지 마이닝 프로그램을 이용하여 기계학습 알고리즘인 Random Forest, Neural Network, Linear Regression, AdaBoost, Gradient Boosting을 적용하여 압축지수 예측모델을 제시하였다. 각 모델의 결과는 오차율을 나타내는 지표 중 하나인 RMSE 값과 MAPE 값 그리고 모델의 유의미함을 나타내는 R2 값을 비교하여 평가하였다. 그 결과, 함수비가 가장 큰 상관성을 보이며, 소성한계의 경우 다른 특성들보다 다소 낮은 상관성을 나타냈다. 각 모델을 비교한 결과 AdaBoost 모델이 가장 오차율이 낮고, 결정 계수 값이 크게 도출되었다.

Keywords

References

  1. Azzouz, A. S., Krizek, R. J. and Corotis, R. B. (1976), Regression Analysis of Soil Compressibility. Soils and Foundation, Japan Society of Soil Mechanics and Foundations Engineering, Tokyo, Japan. 
  2. Bak, J. Y., Kim, M. S. and Kim, J. (2024), "Study on the Inverse Estimation of Strength of Sub-laminate of CFRP Composite Based on Artificial Neural Network Surrogate Model", Transactions of KSAE, Vol.32, No.6, pp.519-527. 
  3. Barron, R.A. (1948), "Consolidation of fine-grained soils by drain wells. Transactions", ASCE, Vol.113, No.2346, pp.718-754. 
  4. Breiman, L. (2001), "Random forests. Machine Learning", Kluwer Academic Publishers, Vol.45, pp.5-32. 
  5. Gibson, R.E., England, G.L. and Hussey, M.J.L. (1967), "The theory of one-dimensional consolidation of saturated clays", Geotechnique, Vol.17, No.3, pp.261-273. 
  6. Guo, H., Wang. J. K., Ao. W. and He. J. L. (2018), "SGBELM: An Advanced Stochastic Gradient Boosting-Based Ensemble Scheme for Extreme Learning Machine", Computational Intelli-Gence and Neuroscience, Vol.2018. 
  7. Kang, S. U., Kim, D. H., Kim, T. H., Chung, C. G. and Joo, I. G. (2024), "A Study on the Calculation of Consolidation Constants using Moisture Content of Sedimentary Clay in Busan and Gyeongnam Regions", Journal of the Korean Geosynthetics Society, Vol.23 No.1, pp.39-47. 
  8. Kim, P. S., Jeon, S. S. and Lee, S. H. (2023), "A Study on the Application of Machine Learning to Predict Keirin Competition Ranking and Sports Betting Methods", Korea Service Management Society, Vol.24, No.2, pp.157-192. 
  9. Kim, T. H. (2021), "A research report of design and construction of soft ground improvement work and maintenance improvement plan", Korean Society of Civil Engineers, pp.170 
  10. Kim, W. M., Kang, E. S. and Paik, S. H. (2022), "A Study of the Use of Orange Platform in Artificial Intelligence and Data Science Education", The Korean Assocaiation of Computer Education, Vol.26, No.1 pp.213-216. 
  11. Lee, J. B. (2023), "Exploring the Possibility of Automatic Scoring for Graphical Responses Using a Convolutional Neural Network", New Physics: Sae Mulli, Vol.73, No.2, pp.138-149 
  12. Lee, S. Y., Kim, J. Y., Kang, J. M., Baek, W. J. and Yoon. H. J. (2022), "Comparison of Performance of Machin Learning Models for Predicting Compression Index Based on Clay Properties", Proceedings of the Korean Society of Agricultural Engineers Conference, Vol.22, No.4, pp.127-134. 
  13. Lim, J.-C. (2019), "A final report on research on causes of settlement and prediction of long-term settlement in the Ungdong hinterland complex of Busan Port New Port", Korean Society of Civil Engineers, p.340. 
  14. Mayne, P. W. (1980), "Cam-clay Predictions of Undrained Strength", J. Geotech. Engrg. Div., ASCE, Vol.106, No.11, pp.1219-1242. 
  15. Mikasa, M. (1963), The consolidation of soft clay a new consolidation theory and its application, Kajima Institution Publishing Co., Ltd., Tokyo, Japan, pp.126-140. 
  16. Nishida, Y. (1956), "A Brief Note on Compression Index of Soil", J. Soil Mech. and Found Eng, ASCE, Vol.82, No.3, pp.1-14. 
  17. Terzaghi, K. and Peck, R. (1967), Soil Mechanics in Engineering Practice, John Wiley and Sons, Inc, New York N.Y. 
  18. Yoon, G. L. and Kim, B. T. (2003), "Formula of compression index prediction for marine clay in Korea", Journal of civil and Environmental Engineering Research, Vol.23, No.3C, pp.169-176.