Estimation of Uncertainty in Risk Assessment of Hydrogen Applications

Hydrogen technologies such as hydrogen fuelled vehicles and refuelling stations are being tested in practice in a number of projects (e.g. HyFleet-Cute and Whistler project) giving valuable information on the reliability and maintenance requirements. In order to establish refuelling stations the permitting authorities request qualitative and quantitative risk assessments (QRA) to show the safety and acceptability in terms of failure frequencies and respective consequences.

For new technologies not all statistical data might be established or are available in good quality causing assumptions and extrapolations to be made. Therefore, the QRA results will contain varying degrees of uncertainty as some components are well established while others are not. The paper describes a methodology to evaluate the degree of uncertainty in data for hydrogen applications based on the bias concept of the total probability and the NUSAP concept to quantify uncertainties of new not fully qualified hydrogen technologies and implications to risk management.