한국구조물진단유지관리공학회 논문집 (Journal of the Korea institute for structural maintenance and inspection)

  • 제28권5호
  • /
  • Pages.20-29
  • /
  • 2024
  • /
  • 2234-6937(pISSN)
  • /
  • 2287-6979(eISSN)
한국구조물진단유지관리공학회 (Korea Institute for Structural Maintenance Inspection)
권호 표지
DOI QR코드

DOI QR Code

기계학습 기반 해양 노출 환경의 콘크리트 교량 데이터를 활용한 염화물 확산계수 예측모델 개발

Development of a Machine Learning-Based Model for the Prediction of Chloride Diffusion Coefficient Using Concrete Bridge Data Exposed to Marine Environments

  • 남우석 (부산대학교 사회환경시스템공학과, 국토안전관리원) ;
  • 임홍재 (부산대학교 사회환경시스템공학과)
  • 투고 : 2024.08.09
  • 심사 : 2024.09.11
  • 발행 : 2024.10.31
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

염화물 확산계수는 해양환경에 위치한 콘크리트 교량의 내구성 평가를 위한 중요한 지표 중 하나이다. 본 논문에서는 기존 연구에서 고려하지 않았던 해양 노출 환경(대기중, 비말대, 간만대)과 공용 중인 콘크리트 교량의 데이터를 활용해 염화물 확산계수 예측 모델을 개발하였다. 이를 위해 교량 하부구조에서 취득한 염화물 프로파일 데이터를 활용하였고 데이터 전처리 후 기계학습 모델인 RF, GBM, KNN을 하이퍼파라미터 튜닝을 통해 최적화 하였다. 콘크리트 물성치를 포함한 6개 변수(W/B, 시멘트 종류, 굵은골재 부피 비율, 공용연수, 강도, 노출 환경) 모델과 노출 환경을 고려하지 않은 5개 변수 모델, 정밀안전진단에서 취득 가능한 3개 변수(공용연수, 강도, 노출 환경) 모델을 개발하여 성능을 비교·검토 하였다. 그 결과 해양 환경에 위치한 콘크리트 교량의 경우 노출 환경을 고려함에 따라 염화물 확산계수 예측 모델의 성능을 향상시킬 수 있음을 확인하였으며, 또한 정밀안전진단 데이터만으로도 염화물 확산계수를 효과적으로 예측할 수 있음을 확인하였다.

The chloride diffusion coefficient is a critical indicator for assessing the durability of concrete marine substructures. This study develops a prediction model for the chloride diffusion coefficient using data from concrete bridges located in marine exposure zones (atmospheric, splash, tidal), an aspect that has not been considered in previous studies. Chloride profile data obtained from these bridge substructures were utilized. After data preprocessing, machine learning models, including Random Forest (RF), Gradient Boosting Machine (GBM), and K-Nearest Neighbors (KNN), were optimized through hyperparameter tuning. The performance of these models was developed and compared under three different variable sets. The first model uses six variables: water-to-binder (W/B) ratio, cement type, coarse aggregate volume ratio, service life, strength, and exposure environment. The second model excludes the exposure environment, using only the remaining five variables. The third model relies on just three variables: service life, strength, and exposure environment factors that can be obtained from precision safety diagnostics. The results indicate that including the exposure environment significantly enhances model performance for predicting the chloride diffusion coefficient in concrete bridges in marine environments. Additionally, the three variable model demonstrates that effective predictions can be made using only data from precision safety diagnostics.

키워드

과제정보

이 성과는 부산대학교 기본연구지원사업(2년)에 의하여 연구되었음.

참고문헌

  1. Metha, P. K., and Monteiro, P. N. (2009), Concrete-Structure, properties, and materials. 2nd edition, Prentice Hall, New-Jersey, 113-171.
  2. Metha, P. K., and Monteiro, P. N. (1993), Concrete-structure, Properties, and Materials. 2nd edition. Prentice Hall, 1-7.
  3. Liu, Y., and Weyers, R. E. (1998), Modeling the time-to-corrosion cracking in chloride contaminated reinforced concrete structures, Cement and Concrete Research, 28(9), 1357-1368. https://doi.org/10.1016/S0008-8846(98)00105-7
  4. Thamoas, M. D. A., and Bamforth, P. B. (1999), Modelling chloride diffusionin concrete Effect of fly ash and slag, Cem. Concr. Res., 29, 487-495. https://doi.org/10.1016/S0008-8846(98)00192-6
  5. Tang, L., and Sorensen. H. E. (1998), Evaluation of the Rapid Test Methods for Measuring the Chloride Diffusion Coefficients of Concrete, Nordtest Project No. 1388-98.
  6. Thomas, M. D. A., and Bentz, E. C. (2002), Computer Program for Predicting the Service Life and Life-cycle Costs of Reinforced Concrete Exposed to Chlorides (Life365 Manual), SFA, Love tsvi le, 2002, 8-11.
  7. Bentz, E. C. (2003), Probabilistic Modeling of Service Life for Structures subjected to Chlorides, ACI Materials Journal, Sep.-Oct., 391-397.
  8. Polder, R. B.,. van der Wegen, G., and Boutz, M. (2007), Performance based guideline for service life design of concrete for civil engineering structures - A proposal discussed in the Netherlands. International RILEM Workshop on Performance Based Evaluation and Indicators for Concrete Durability, 31-39.
  9. Kim, M. W. (2013), Prediction of Durability Life Cycle and Repair Cost Analysis using Chloride Diffusion Coefficient for Marine Concrete Structures. Doctoral dissertation, University of Seoul, Department of Civil Engineering.
  10. Yoon, Y. S., Lee, J. S., Min, J. Y., and Kwon, S. J. (2022), Behavior of apparent chloride diffusion coefficient of fly ash concrete under long-term marine exposure, Advances In Concrete Construction, 146, 369-380 (in Korean). https://doi.org/10.2139/ssrn.4051149
  11. HuiPing, Z., XiaoChao, L., Muhammad Nasir, A., Ahmed, A., Siyab, U. A., Fadi, A., and Ayaz. A. (2024), Analyzing chloride diffusion for durability predictions of concrete using contemporary machine learning strategies Materials Today Communications, 38.
  12. Tran, V. Q. (2022), Machine learning approach for investigating chloride diffusion coefficient of concrete containing supplementary cementitious materials. Construction and Building Materials, 328, 127103.
  13. Xuanrui, Y., Jiehong, L., Yang, Y., and Anxiang, S. (2024). Advancing service life estimation of reinforced concrete considering the coupling effects of multiple factors: Hybridized physical testing and machine learning approach, Journal of Building Engineering, 84.
  14. KALIS (2023). Detailed Guidelines for Safety and Maintenance of the Implementation(Bridge).
  15. Korean Standards Association (2017). KS F 2713: Test method for chloride content in concrete (in Korean).
  16. Weerdt, K. D., Wilson, W., MAchner, A., and Georget, F. (2023), Chloride profiles - What do they tell us and how should they be used? Cement and Concrete Research, 173.
  17. Makin, J. G. (2022), An Introduction to Modern Statistical Learning. arXiv preprint arXiv:2207.10185.
  18. Glasser, F. P., Marchand, J., and Samson, E. (2008). Durability of Concrete - Degradation Phenomena Involving Detrimental Chemical Reactions, Cement and Concrete Research, 38(2), 226-246. https://doi.org/10.1016/j.cemconres.2007.09.015
  19. Chan, J. Y., Leow, S. M., Bea, K. T., Cheng, W. K., Phoong, S. W., Hong, Z. W., and Chen, Y. L. (2022). Mitigating the Multicollinearity Problem and Its Machine Learning Approach: A Review Mathematics, MDPI, 10(8), 1283.
  20. Benesty, J., Chen, J., Huang, Y., and Cohen, I. (2009). Pearson correlation coefficient. In Noise reduction in speech processing, Springer, Berlin, Heidelberg
  21. O'Brien, R. M. (2007), A caution regarding rules of thumb for variance inflation factors, Quality & Quantity, 41(5), 673-690. https://doi.org/10.1007/s11135-006-9018-6
  22. Jolliffe, I. T. (2002), Principal Component Analysis. Springer-Verlag New York.
  23. Linjian, W. (2020), Study on Chloride Ion Diffusion Considering the Heterogeneity of Concrete Materials and the Blocking Effect of Reinforcement, Doctoral dissertation, Tianjin University.