This investigation is concerned with the mitigation of hazards associated with ignited releases of hydrogen from compressed gas tanks aboard fuel cell vehicles. These releases are controlled through a safety valve installed on the tank known as a pressure relief device, but current designs use round nozzles and initially high mass flow rates for the release of hydrogen from the tank. Consequently, the separation distances from pressure relief device releases due to ignited hydrogen can approach 50 m for a "no harm" criterion. The present investigation is divided into four parts. The first body chapter extends previous work which simulated experimental releases from round and plane jet fires using a two-stage modelling approach. The same calculation domains as in the previous investigation are employed for consistency, with a one-equation turbulence model and with combustion simulated in the near-to-nozzle field. It was demonstrated that the new model yields results which are 20% more conservative than those found previously. The flow features of the combusting near-to-nozzle field are described qualitatively. The results from this investigation serve primarily as a model validation for subsequent chapters. The second chapter explores the efficacy of using a variable-aperture pressure relief device to reduce the initially high mass flow rate after a nozzle is opened. To that end, the motion of a spring-loaded throttle immediately upstream of the nozzle is modelled. Parameters for the spring coefficient and throttle geometry are chosen so that the initial mass flow rate does not exceed a prescribed value. Compared to a fixed nozzle with the same initial mass flow rate, the variable-aperture device reduced the blow-down time from 120 minutes to 21 minutes. A simple spring system (constant spring coefficient) was able to constrain the throttle motion appropriately. The use of a plane nozzle achieved a substantial reduction in a jet flame length with the same release conditions, but with an associated increase in the width of the flame. The third body chapter investigates the use of a radial jet to reduce the flame length further. To facilitate comparison, the radial jet had the same opening area as the earlier-studied round and plane jets. The radial jet flame took the shape of a thin, flat disc and reduced the flame length by 75% compared to the round nozzle (33% compared to the plane jet). The rate of increase of flame length with increasing pressure was compared with previous experimental results for round and plane jets, and it was observed that the power law correlation between flame length and inlet pressure had a lower exponent (0.3) compared to round jets (0.43). The fourth body chapter examines the effects of external air movement on the shape and size of the round hydrogen jet fire. Wind velocities of 5, 10, 20, and 30 m/s were examined in directions parallel to, orthogonal to, and directly opposing the direction of release. Deflections from side winds were measured and compared with experiments. A correlation was developed using the momentum flux ratio which collapsed the results from the highly underexpanded hydrogen jet fire onto the same curve as jet fires from other fuels at a variety of velocities, thus expanding the range of applicability of a previously stated correlation. Wind velocities in excess of 10 m/s reduced the flame length by half, but did not vary appreciably in the range examined. Co-flowing and counter-flowing wind yielded results qualitatively consistent with theoretical predictions. This thesis proposes two ways to manage hazards associated with releases of hydrogen from pressure relief devices. The use of a variable-aperture pressure relief device would allow the mass flow rate to be chosen a priori and to be set in such a way as to reduce hazards to life safety and to property for a known container volume and pressure. A radial jet pressure relief device reduced the flame length by 75% compared to a free jet and better managed high internal tank pressures than the uni-directional releases. Fina"y, the effects of wind on hydrogen jet fires were contextualised in a framework that captures other fuels at lower velocities. The particular wind velocity of 10 m/s was identified as a "threshold" above which further increases in the wind velocity do not decrease the flame length significantly, a result which was corroborated by contemporary experiments.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:701439 |
| Date | January 2016 |
| Creators | Yates, David |
| Contributors | Molkov, Vladimir ; Makarov, Dmitriy |
| Publisher | Ulster University |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0018 seconds