Statistics, Lessons Learnt and Recommendations from the Analysis of the Hydrogen Incidents and Accidents Database (HIAD 2.0)

The Hydrogen Incidents and Accidents Database (HIAD) is an international open communication platform collecting systematic data on hydrogen-related undesired incidents, which was initially developed in the frame of HySafe, an EC co-funded Network of Excellence in the 6th Frame Work Programme by the Joint Research Centre of the European Commission (EC-JRC). It was updated by JRC as HIAD 2.01 in 2016 with the support of the Fuel Cells and Hydrogen 2 Joint Undertaking (FCH 2 JU). Since the launch of the European Hydrogen Safety Panel2 (EHSP) initiative in 2017 by FCH 2 JU, the EHSP has worked closely with JRC to upload additional/new incidents to HIAD 2.0 and analyze them to gather statistics, lessons learnt and recommendations through Task Force 3. The first report to summarise the findings of the analysis was published by FCH 2 JU in September 2019. Since the publication of the first report, the EHSP and JRC have continuously worked together to enlarge HIAD 2.0 by adding newly occurred incidents as well as quality historic incidents which were not previously uploaded to HIAD 2.0. This has facilitated the number of validated incidents in HIAD 2.0 to increase from 272 in 2018 to 593 in March 2021. This number is also dynamic and continues to increase as new incidents are being continuously added by both EHSP and JRC; and validated by JRC. The overall quality of the published incidents has also been improved whenever possible. For example, additional information has been added to some existing incidents. Since mid-2020, EHSP Task Force TF3 has further analysed the 485 events, which were in the database as of July 2020. For completeness of the statistics, these include the events considered in our first report3 as well as the newly added/validated events since then. In this process, the EHSP has also re-visited the lessons learnt in the first report to harmonise the approaches of analysis and improve the overall analysis. The analysis has comprehensively covered statistics, lessons learnt and recommendations. The increased number of incidents has also made it viable to extract statistics from the available incidents at the time of the analysis, including previously available incidents. It should be noted that some incidents reported is of low quality therefore it was not included in the statistical analysis.