Advancing overpressure quantification methodology for hydrogen venting

A major safety problem at present is the lack of an agreed methodology to estimate blast loads from hydrogen (H2) vents. Of particular interest is the delayed ignition of the flammable clouds that form upon a controlled or uncontrolled release. This study aims to test and improve an existing methodo...

Descripción completa

Detalles Bibliográficos
Autor: Jarubenjaluk, Adisa
Tipo de recurso: tesis de maestría
Fecha de publicación:2025
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:dnet:upcommonspor::f1c8592f9813e50faf6f83badbac8d2a
Acceso en línea:https://hdl.handle.net/2117/460999
Access Level:acceso abierto
Palabra clave:Hydrogen -- Safety measures
Hydrogen
Hydrogen venting
Deflagration
Detonation
Unconfined and uncongested release
Hydrogen explosion
Overpressure
Hidrogen -- Mesures de seguretat
Hidrogen -- Inflamabilitat
Àrees temàtiques de la UPC::Enginyeria química
Descripción
Sumario:A major safety problem at present is the lack of an agreed methodology to estimate blast loads from hydrogen (H2) vents. Of particular interest is the delayed ignition of the flammable clouds that form upon a controlled or uncontrolled release. This study aims to test and improve an existing methodology to enable more accurate prediction of overpressure in hydrogen venting scenarios. Present deflagration model underestimates overpressures by an order of magnitude, whereas the detonation model results in overly conservative estimates albeit useful to retain as an upper limit. While in safety assessments, a certain degree of conservatism is preferred due to the non-negligible potential consequences, estimates should be realistic and in line with the physics that expected to play a role emphasising the need to improve deflagration modelling. Visible flame speed is identified as the most influential parameter in the deflagration model. Modelled estimates based on cloud radius tends to yield slower flame speeds. An alternative approach using the Reynolds number—reflecting the turbulence induced by pressurised gas release—was proposed. The refined methodology was tested against vertical hydrogen release experiments, demonstrating improved overpressure predictions, with 73% of the predictions falling within a factor of four overestimation.