Dynamic risk analysis of hydrogen gas leakage using Bow-tie technique and Bayesian network

The power plants of each country can be considered as one of the most important factors in economic development and growth of that country. Accidents at power plants are very dangerous and make their accessibility difficult. So, it is critical to be able to predict and appropriately assess the relevant risks. In this study, we focus on hydrogen gas leakage from chlorination unit as the study scenario to evaluate potential risk of accidents. We then use the Bow-tie technique and Bayesian Network analysis to determine the type and the relationship between the effective causes of the catastrophic accidents. According to Bayesian Network, decrease in flow rate in the ventilation system of the storage tank was identified as the most probable base event, and erupted fire/sudden fire/explosion were identified as the most probable consequences of occurrence of the top event. The use of Bayesian network could reduce parameter uncertainty through probability updating.