Safety barrier performance assessment by integrating computational fluid dynamics and evacuation modeling for toxic gas leakage scenarios

Safety barrier performance assessment by integrating computational fluid dynamics and evacuation modeling for toxic gas leakage scenarios

Yuan, S. (TU Delft Safety and Security Science)
Cai, Jitao (China University of Mining and Technology (Beijing))
Reniers, G.L.L.M.E. (TU Delft Safety and Security Science; Universiteit Antwerpen; Katholieke Universiteit Leuven)
Yang, M. (TU Delft Safety and Security Science; Universiti Teknologi Malaysia)
Chen, Chao (Student TU Delft)
Wu, Jiansong (China University of Mining and Technology (Beijing))

2022

Toxic gas leakage represents a type of major process accident scenario threatening human life. Technical and non-technical safety barriers are employed to prevent toxic gas leakage accidents or mitigate the possible catastrophic consequences. Evacuation must be executed in severe toxic gas release scenarios. The performance assessment of technical safety barriers and evacuations in these accident scenarios, although very important, has never been investigated in previous studies. This paper proposes an approach integrating event tree analysis (ETA), computational fluid dynamics (CFD) simulation, and evacuation modeling (EM), for risk assessment of toxic gas leakage accidents in chemical plants. In the proposed approach, the spatiotemporal distribution of toxic gas is predicted by CFD simulations. A dynamic evacuation is determined by a cellular automaton (CA)-based model. Synergistic interventions resulting from technical safety barriers and evacuations are considered in the risk assessment. Considering safety barrier failures in the event tree analysis, individual fatality risks due to toxic gas leakage scenarios are calculated. For illustrative purposes, the proposed method is applied to a case of ammonia leakage. The results show that worse scenarios would be ignored without considering the failure probabilities of technical safety barriers, which can cause underestimated individual fatality risks. Timely gas detection & alarm has the potential to expedite the starting time of evacuations and thus may shorten the time that evacuees stay in the toxicity area to reduce individual fatality risks.

Chemical industry
Computational fluid dynamics
Evacuation modeling
Risk assessment
Safety barriers
Toxic gas leakage

http://resolver.tudelft.nl/uuid:8cd62ffe-9781-4288-adda-93d9beb8dff7

https://doi.org/10.1016/j.ress.2022.108719

0951-8320

Reliability Engineering & System Safety, 226

Institutional Repository

journal article

© 2022 S. Yuan, Jitao Cai, G.L.L.M.E. Reniers, M. Yang, Chao Chen, Jiansong Wu