KSCE Journal of Civil and Environmental Engineering Research (대한토목학회논문집)

Korean Society of Civil Engeneers (대한토목학회)
권호 표지
DOI QR코드

DOI QR Code

Evaluation of a Thermal Conductivity Prediction Model for Compacted Clay Based on a Machine Learning Method

기계학습법을 통한 압축 벤토나이트의 열전도도 추정 모델 평가

  • 윤석 (한국원자력연구원 방사성폐기물처분연구부) ;
  • 방현태 (한밭대학교 토목공학과) ;
  • 김건영 (한국원자력연구원 방사성폐기물처분연구부) ;
  • 전해민 (한밭대학교 건설환경공학과)
  • Received : 2020.05.25
  • Accepted : 2020.11.11
  • Published : 2021.04.01
Download PDF
⟨ Previous Next ⟩

Abstract

The buffer is a key component of an engineered barrier system that safeguards the disposal of high-level radioactive waste. Buffers are located between disposal canisters and host rock, and they can restrain the release of radionuclides and protect canisters from the inflow of ground water. Since considerable heat is released from a disposal canister to the surrounding buffer, the thermal conductivity of the buffer is a very important parameter in the entire disposal safety. For this reason, a lot of research has been conducted on thermal conductivity prediction models that consider various factors. In this study, the thermal conductivity of a buffer is estimated using the machine learning methods of: linear regression, decision tree, support vector machine (SVM), ensemble, Gaussian process regression (GPR), neural network, deep belief network, and genetic programming. In the results, the machine learning methods such as ensemble, genetic programming, SVM with cubic parameter, and GPR showed better performance compared with the regression model, with the ensemble with XGBoost and Gaussian process regression models showing best performance.

완충재는 고준위 방사성 폐기물을 처분하기 위한 공학적 방벽 시스템에서 중요한 구성요소 중 하나이며 사용 후 핵연료가 담긴 처분용기와 암반사이에 채워지는 물질이기 때문에 지하수 유입으로부터 처분용기를 보호하고, 방사성 핵종 유출을 저지하는 중요한 역할을 수행한다. 따라서 공학적 방벽 시스템의 처분용기로부터 발생하는 고온의 열량은 완충재를 통하여 전파되기에 완충재의 열전도도는 처분시스템의 안전성 평가에 매우 중요하다. 본 연구에서는 국내에서 생산되는 압축 벤토나이트 완충재의 열전도도 예측을 위한 경험적 회귀 모델의 정합성을 검증하고 정확도를 높이기 위해 예측모델의 구축에 기계학습법을 적용해 보았다. 벤토나이트의 건조밀도, 함수비 및 온도 값을 바탕으로 열전도도를 예측하고자 하였으며, 이때 다항 회귀, 결정 트리, 서포트 벡터 머신, 앙상블, 가우시안 프로세스 회귀, 인공신경망, 심층 신뢰 신경망, 유전 프로그래밍과 같은 기계학습 기법을 적용하였다. 기계학습 기법을 이용하여 예측한 결과, 부스팅 기반의 앙상블 기법, 유전 프로그래밍, 3차 함수 기반의 SVM, 가우시안 프로세스 회귀의 기계학습기법을 활용한 모델이 선형 회귀 분석 기법에 비해 좋은 성능을 보였으며, 특히 앙상블의 부스팅 기법과 가우시안 프로세스 회귀 기법을 사용한 모델들이 가장 좋은 성능을 보였다.

Keywords

References

  1. Bang, H. T., Yoon, S. and Jeon, H. (2020). "Application of machine learning methods to predict a thermal conductivity model for compacted bentonite." Annals of Nuclear Energy, Vol. 142, 107395. https://doi.org/10.1016/j.anucene.2020.107395
  2. Cho, W. J. (2019). Bentonite barrier material for radioactive waste disposal, Korea Atomic Energy Research Institute Report, KAERI/GP-535/2019 (in Korean).
  3. Cho, W. J., Lee, J. W. and Kwon, S. K. (2011). "An empirical model for the thermal conductivity of compacted bentonite and a bentonite-sand mixture." Heat and Mass Transfer, Vol. 47, pp. 1385-1393. https://doi.org/10.1007/s00231-011-0800-1
  4. Choi, H. J., Kim, K. S., Cho, W. J., Lee, J. O. and Choi, J. W. (2014). HLW long-term management system development: Development of engineered barrier system performance, Korea Atomic Energy Research Institute Report, KAERI/TR-3859/2014 (in Korean).
  5. Eberhart, R. C. and Shi, Y. (2011). Computational intelligence: concepts to implementations, Elsevier, Amsterdam, Nederland.
  6. Fischer, A. and Igel, C. (2014). "Training restricted Boltzmann machines: An introduction." Pattern Recognition, Vol. 47, No. 1, pp. 25-39. https://doi.org/10.1016/j.patcog.2013.05.025
  7. Geron, A. (2017). Hands-on machine learning with Scikit-Learn and TensorFlow, O'Reilly Media, Inc., California, USA.
  8. Japan Nuclear Cycle Development Institute (JNC) (1999). H12 project to establish technical basis for HLW disposal in Japan, Supporting Report 2. Japan Nuclear Cycle Development Institute.
  9. Karnland, O. (2010). Chemical and mineralogical characterization of the bentonite buffer for the acceptance conctrol procedure in a KBS-3 repository, Svensk Karn-branslehantering AB Report, SKB TR-10-60.
  10. Lee, J. O., Cho, W. J. and Kwon, S. K. (2011). "Thermal-hydro-mechanical properties of refernece bentonite buffer for a Korean HLW repository." Tunnel and Underground Space, Vol. 21, No. 4, pp. 264-273 (in Korean). https://doi.org/10.7474/TUS.2011.21.4.264
  11. Lee, J. O., Lee, M, S., Choi, H. J., Lee, J. Y. and Kim, I. Y. (2014). Establishment of the concept of buffer for an HLW repository: An approach, Korea Atomic Energy Research Institute Report, KAERI/TR-5824 (in Korean).
  12. Madsen, F. T. (1988). "Clay mineralogical investigation related to nuclear waste disposal." Clay Mincerals, Vol. 33, No. 1, pp. 109-129. https://doi.org/10.1180/000985598545318
  13. Mitchell, T. (1997). Machine learning, McGraw Hill, New York, USA.
  14. Polikar, R. (2006). "Ensemble based systems in decision making." IEEE Circuits and systems magazine, Vol. 6, No. 3, pp.21-45. https://doi.org/10.1109/MCAS.2006.1688199
  15. Rasmussen, C. E. and Williams, C. K. I. (2006). Gaussian processes for machine learning, MIT Press, Cambridge, Massachusetts.
  16. Smola, A. J. and Scholkopf, B. (2004). "A tutorial on support vector regression." Statistics and Computing, Vol. 14, No. 3, pp. 199-222. https://doi.org/10.1023/B:STCO.0000035301.49549.88
  17. Tang, A. M., Cui, Y, J. and Le, T. T. (2008). "A study on the thermal conductivity of compacted bentonite." Applied Clay Science, Vol. 41, pp. 181-189. https://doi.org/10.1016/j.clay.2007.11.001
  18. Vapnik, V. (1995). The nature of statistical learning theory, Springer, New York, USA.
  19. Villar, M. V., Martin, P. L. and Barcala, J. M. (2006). "Modification of physical, mechanical and hydraulic properties of bentonite by thermo-hydraulic gradients." Engineering Geology, Vol. 81, No. 3, pp. 284-297. https://doi.org/10.1016/j.enggeo.2005.06.012
  20. Xu, Y., Sun, D., Zeng, Z. and Lv, H. (2019). "Temperature dependence of apparent thermal conductivity of compacted bentonites as buffer material for high-level radioactive waste repository." Applied Clay Science, Vol. 174, pp. 10-14. https://doi.org/10.1016/j.clay.2019.03.017
  21. Yoo, M., Choi, H. J., Lee, M. S. and Lee, S. Y. (2016). "Measurement of properties of domestic bentonite for a buffer of an HLW repository." Journal of the Korean Radioactive Waste Society, Vol. 14, No. 2, pp. 135-147 (in Korean). https://doi.org/10.7733/jnfcwt.2016.14.2.135
  22. Yoon, S., Cho, W. H, Lee, C. S. and Kim, G. Y. (2018). "Thermal conductivity of Korean compacted bentonite buffer materials for a nuclear waste repository." Energies, Vol. 11, No. 9, 2269. https://doi.org/10.3390/en11092269
  23. Yoon, S. and Kim, K. Y. (2021). "Measuring thermal conductivity and water suction for variably saturated bentonite." Nuclear Engineering and Technology, Vol. 53, pp. 1041-1048. https://doi.org/10.1016/j.net.2020.08.017