Journal of the Computational Structural Engineering Institute of Korea (한국전산구조공학회논문집)

Computational Structural Engineering Institute of Korea (한국전산구조공학회)
권호 표지
DOI QR코드

DOI QR Code

Bayesian Network-based Probabilistic Safety Assessment for Multi-Hazard of Earthquake-Induced Fire and Explosion

베이지안 네트워크를 이용한 지진 유발 화재･폭발 복합재해 확률론적 안전성 평가

  • Se-Hyeok Lee (Department of Structural Engineering Research, Korea Institute of Civil Engineering and Building Technology) ;
  • Uichan Seok (Department of Civil and Environmental Engineering, Seoul National University) ;
  • Junho Song (Department of Civil and Environmental Engineering, Seoul National University)
  • 이세혁 (한국건설기술연구원 구조연구본부) ;
  • 석의찬 (서울대학교 건설환경공학부) ;
  • 송준호 (서울대학교 건설환경공학부)
  • Received : 2024.05.21
  • Accepted : 2024.06.03
  • Published : 2024.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

Recently, seismic Probabilistic Safety Assessment (PSA) methods have been developed for process plants, such as gas plants, oil refineries, and chemical plants. The framework originated from the PSA of nuclear power plants, which aims to assess the risk of reactor core damage. The original PSA method was modified to adopt the characteristics of a process plant whose purpose is continuous operation without shutdown. Therefore, a fault tree, whose top event is shut down, was constructed and transformed into a Bayesian Network (BN), a probabilistic graph model, for efficient risk-informed decision-making. In this research, the fault tree-based BN from the previous research is further developed to consider the multi-hazard of earthquake-induced fire and explosion (EQ-induced F&E). For this purpose, an event tree describing the occurrence of fire and explosion from a release is first constructed and transformed into a BN. And then, this BN is connected to the previous BN model developed for seismic PSA. A virtual plot plan of a gas plant is introduced as a basis for the construction of the specific EQ-induced F&E BN to test the proposed BN framework. The paper demonstrates the method through two examples of risk-informed decision-making. In particular, the second example verifies how the proposed method can establish a repair and retrofit strategy when a shutdown occurs in a process plant.

최근 원자력 지진 PSA(Probabilistic Safety Assessment)를 토대로 산업시설물의 지진 PSA를 수행하는 연구가 진행되었다. 해당 연구는 원자력 발전소와 산업시설물의 차이를 파악하고, 최종적으로 운영정지를 목표로 하는 고장수목(Fault Tree)를 구축한 후 시각적 확률도구인 베이지안 네트워크(Bayesian Network, BN)으로 변환하였다. 본 연구는 선행연구를 기반으로 지진으로 유발된 구조손상으로 인해 발생 가능한 화재･폭발에 대해 PSA를 수행하고자 하였다. 이를 위해 화재･폭발을 사건수목(Event Tree)으로 표현하고, BN으로 변환하였다. 변환된 BN은 화재･폭발 모듈로서 선행연구에서 제시된 고장수목 기반 BN과 연계되어 최종적으로 지진 유발 화재･폭발 PSA를 수행할 수 있는 BN 기반 방법론이 개발되었다. 개발된 BN을 검증하기위해 수치예제로서 가상의 가스플랜트 Plot Plan을 생성하였고, 가스플랜트의 설비 종류가 구체적으로 반영된 대규모 BN을 구축하였다. 해당 BN을 이용하여 지진 규모에 따른 전체시스템의 운영정지 확률 및 하위시스템들의 고장확률 산정과 더불어 역으로 전체시스템이 운영 정지되었을 때 하위시스템들의 영향도 분석과 화재･폭발 가능성을 산정하여 다양한 의사결정을 수행할 수 있음을 제시함으로써 그 우수성을 확인하였다.

Keywords

Acknowledgement

본 연구는 국토교통부/국토교통과학기술진흥원의 지원으로 수행되었습니다(과제번호: RS-2021-KA163162).

References

  1. Abbiati, G., Brocardo, M., Filippo, R., Stojadinovic, B., Bursi, O.S. (2021) Seismic Fragility Analysis of a Coupled Tank-Piping System based on Artificial Ground Motions and Surrogate Modeling, J. Loss Prev. Process Ind., 72(2021), p.104575.
  2. Barber, D. (2012) Bayesian Reasoning and Machine Learning, Cambridge: Cambridge University Press.
  3. Bayes Fusion, LLC. (2023) GeNie (Version 4.0.3005) [Computer Software], Released on Jun 5, 2023, from www.bayesfusion.com.
  4. Bensi, M., Der Kiureghian, A., Straub, D. (2011) A Bayesian Network Methodology for Infrastructure Seismic Risk Assessment and Decision Support, Pacific Earthquake Engineering Research Center.
  5. Bensi, M., Der Kiureghian, A., Straub, D. (2013) Efficient Bayesian Network Modeling of Systems, Reliab. Eng. & Syst. Saf., 112(2013), pp.200~213. https://doi.org/10.1016/j.ress.2012.11.017
  6. Bursi, O.S., Filippo, R., Salandra., V., Pedot, M., Reza, M.S. (2018) Probabilistic Seismic Analysis of an LNG Subplant, J. Loss Prev. Process Ind., 53(2018), pp.45~60. https://doi.org/10.1016/j.jlp.2017.10.009
  7. Byun, J.-E., Zwirglmaier, K., Straub, D., Song, J. (2017) Matrixbased Bayesian Network for Efficient Memory Storage and Flexible Inference, Reliab. Eng. & Syst. Saf., 185(2017), pp. 533~545.
  8. EPRI (1994) Methodology for Developing Seismic Fragilities, TR-103959, Electric Power Research Institute.
  9. EPRI (2003) Seismic Probabilistic Risk Assessment Implementation Guide, TR-1002989, Electric Power Research Institute.
  10. Hauptmanns, U. (2004) Semi-Quantitative Fault Tree Analysis for Process Plant Safety using Frequency and Probability Ranges, J. Loss Prev. Process Ind., 17(2004), pp.339~345. https://doi.org/10.1016/j.jlp.2004.06.004
  11. Hwang, H.H.M., Chou, T. (1998) Evaluation of Seismic Performance of an Electric Substation using Event Tree/Fault Tree Technique, Probab. Eng. Mech., 13(2), pp.117~124. https://doi.org/10.1016/S0266-8920(97)00018-0
  12. KICT (2021) Development of Integrated Risk Management Package Technology based on Safety of Facilities, Annual Report, Korea Institute of Civil Engineering and Building Technology.
  13. Kim, M.-K., Choun, U.-S., Choi, I.-K., Oh, K.-H. (2009) Seismic Fragility Analysis of Substation Systems by Using the Fault Tree Method, J. Earthq. Eng. Soc. Korea, 13(2), pp.47~58.
  14. KOSHA (2021) KOSHA GUIDE, P-87-2021, Korea Occupational Safety & Health Agency.
  15. Kwag, S., Eem, S., Choi, E., Ha, J. G., Hahm, D. (2021) Suggestions for Enhancing Sampling-based Approach of Seismic Probabilistic Risk Assessment, J. Comput. Struct. Eng. Inst. Korea, 34(2), pp.77~84. https://doi.org/10.7734/COSEIK.2021.34.2.77
  16. Kwag, S., Gupta, A. (2016) Bayesian Network Technique in Probabilistic Risk Assessment for Multiple Hazards, Proceedings of 24th International Conference on Nuclear Engineering (ICONE 24), June 26-30, 2016, Charlotte, NC, US.
  17. Lee, S.-H., Mun, C., Park, S., Cho, J.-R., Song, J. (2023) Probabilistic Safety Assessment of Gas Plant Using Fault Tree-based Bayesian Network, J. Comput. Struct. Eng. Inst. Korea, 36(4), pp.273~282. https://doi.org/10.7734/COSEIK.2023.36.4.273
  18. Lee, S.-H., Song, J. (2016) Bayesian-Network-based System Identification of Spatial Distribution of Structural Parameters, Eng, Struct., 127(2016), pp.260~277. https://doi.org/10.1016/j.engstruct.2016.08.029
  19. Oh, J., Kwag, S. (2018) A Study on Seismic Probabilistic Safety Assessment for a Research Reactor, J. Comput. Struct. Eng. Inst. Korea, 31(1), pp.31~88. https://doi.org/10.7734/COSEIK.2018.31.1.31
  20. Vanem, E., Antao, P., Ostvik, I., Castillo, F. (2008) Analysing the Risk of LNG Carrier Operations, Reliab. Eng. & Syst. Saf., 93(2008), pp.1328~1344. https://doi.org/10.1016/j.ress.2007.07.007
  21. Zwirglmaier, K. (2016) Reliability Analysis with Bayesian Networks, Ph.D Thesis, Technische Universitat Munchen.