Search

Hydrogen is a promising energy carrier that will become competitive in the near future. The present study modeled hydrogen leaks and diffusion in an actual size underground parking garage with the numerical model validated by scale experimental data. The results show that the hydrogen concentration distributions are not uniform in the gas mixture layer along the ceiling and the initial front velocity of the gas mixture layer decays with horizontal distance from the leaking car. The vertical filling front velocity for times after 600 s remain constant in the near field but increases linearly with distance in the far field. The corner walls did not significantly affect the far-field concentration distributions and the ventilation layout with vents in the garage corners provided better hydrogen removal. These results can be used to predict the hydrogen concentration buildup in large confined spaces and to help design underground parking garage ventilation systems.

Previous studies have shown that the two-layer model more accurately predicts hydrogen dispersion than the conventional notional nozzle models without significantly increasing the computational expense. However, the model was only validated for predicting the concentration distribution and has not been adequately validated for predicting the velocity distributions. In the present study, particle imaging velocimetry (PIV) was used to measure the velocity field of an underexpanded hydrogen jet released at 10 bar from a 1.5 mm diameter orifice. The two-layer model was the used to calculate the inlet conditions for a two-dimensional axisymmetric CFD model to simulate the hydrogen jet downstream of the Mach disk. The predicted velocity spreading and centerline decay rates agreed well with the PIV measurements. The predicted concentration distribution was consistent with data from previous planar Rayleigh scattering measurements used to verify the concentration distribution predictions in an earlier study. The jet spreading was also simulated using several widely used notional nozzle models combined with the integral plume model for comparison. These results show that the velocity and concentration distributions are both better predicted by the two-layer model than the notional nozzle models to complement previous studies verifying only the predicted concentration profiles. Thus, this study shows that the two-layer model can accurately predict the jet velocity distributions as well as the concentration distributions as verified earlier. Though more validation studies are needed to improve confidence in the model and increase the range of validity, the present work indicates that the two-layer model is a promising tool for fast, accurate predictions of the flow fields of underexpanded hydrogen jets.

The detection and localization of hydrogen releases are of vital importance in large confined spaces to enable safe commercial penetration of hydrogen energy into society. In the present study, a hydrogen leak localization system was developed based on measured concentration data and a machine learning system. A scale parking garage model was used to experimentally model fuel cell vehicle leaks in a confined space with helium used as a surrogate of hydrogen for safety reasons. Twelve gas sensors were placed along the ceiling to measure the helium concentrations. The leak position was changed in each test to gather a range of concentration data. Then, the measured helium concentrations were used as input data to train the machine learning models. Two machine learning algorithms were used to locate the leak position, an artificial neural network (ANN) and the K-DTW algorithm. The helium concentrations recorded by the twelve sensors at a specific time point were used as the input data for the ANN model, while time-series concentration data was used as the input for the K-DTW algorithm. The results show that the ANN model can predict leak locations that were not included in the training datasets and that the K-DTW algorithm can classify the unknown leak location to the given location label. The ANN prediction accuracy was 78.4%. The K-DTW algorithm identified the leak location with 87.5% accuracy. The localization technology developed in the present work can provide safe monitoring of large parking garages and the model accuracy can be improved by feeding more training data to the machine learning models.

Hydrogen is likely to be stored and transported as a liquid with high densities to have sufficient hydrogen on-site and for efficient distribution. Thus, cryogenic hydrogen jets and dispersion characteristics need to be studied to quantify the risks of accidental releases. The present study modeled cryogenic hydrogen jets with stagnation temperatures around 50 K released from round nozzles at stagnation pressures of 2–10 bar. The instantaneous flow fields were modeled using the Large Eddy Simulation (LES) turbulence model to accurately and efficiently capture the flow characteristics. The numerical models were validated with experimental data. The concentration and velocity distributions show that the centerline mass fraction and velocity decay at similar rates to the decays for room temperature jets. The radial distributions of the mass fraction, temperature and velocity are Gaussian and self-similar. Additionally, the numerical model can reproduce the shock structures near the nozzle for these underexpanded cryogenic hydrogen jets. The location of the first normal shock was predicted to be closer to the nozzle than for room temperature jets. The present work can be used to develop and revise safety codes and standards for liquid hydrogen facilities.

Most previous studies of hydrogen fuel cell vehicle (HFCV) releases in garages have focused on residential garages. However, hydrogen releases in large underground parking garages can be completely different with more concerns about public safety. In the present work, unintended hydrogen releases were studied in a scaled-down underground parking garage model to fill the knowledge gap concerning hydrogen dispersion in a large confined space. Helium was used in place of hydrogen for safety reasons. Gas sensors were used to measure the helium concentrations along the ground and the ceiling. The nearfield concentration distributions were studied to show that the leak jet was strongly affected by the buoyancy, so the gas concentration was very low along the floor with a flammable gas mixture only appearing right under the leaking car. The gas plume driven by the buoyancy reached the ceiling and spread in all directions resulting in a flammable layer along the ceiling. The effects of neighboring cars and walls were also investigated to show that the obstacles affected the nearfield gas dispersion but did not significantly affect the farfield concentration distributions. The concentration increased linearly with the release flow rate for the flow rates in this study. The gas dispersion during the release showed self-similar characteristics, which could be used for leak detection. This database will be useful for model validation for modeling HFCV releases in large underground parking garages and also provides guidelines for gas sensor placement in large confined spaces.

Photocatalytic hydrogen evolution from water is a valid method to face the challenge of energy shortage. TiO2, as the most well-known and investigated semiconductor photocatalyst, has been extensively applied in this field. However, the hydrogen evolution performance of TiO2 suffer from low efficiency due to the high combination of photo-induced carries and the narrow light utilization range resulting from wide band gap. In this work, we use the commercial TiO2 (P25) and fullerene (C60) as precursors to construct an efficient photocatalyst via a mild mixed method. The introduction of C60 not only effectively broaden light absorption, but also suppressed the recombination of photo-generated carriers on TiO2. The photocatalytic water splitting property of the optimum sample achieve 597.2 μmol g−1 h−1, which is 31 times higher than pure P25.

The development and revision of safety codes and standards for hydrogen infrastructure requires a solid scientific basis, including studies of unignited releases from high pressure systems for various scenarios. Most hydrogen releases are modeled as axisymmetric jets, but real leaks are more likely to be non-axisymmetric jets issuing from high aspect ratio cracks or slots. In the present study, underexpanded hydrogen jets from square and rectangular nozzles with aspect ratios of 1–16 were numerically modeled for stagnation pressures up to 20 MPa. The near and far flow fields were modeled separately using two sequential computational domains to accurately and efficiently capture the flow characteristics. The numerical models were first validated with experimental data from a previous experimental study and literature data. The mass fraction and velocity distributions show that the centerline decay rates increase as the nozzle aspect ratio increases, but this increase is dependent on the pressure. This means that the canonical decay law of round turbulent jets and plumes no longer applies to the slot nozzle jets for high pressures. The radial profiles collapse onto a Gaussian curve in the major axis plane, but neither collapse, nor are they Gaussian in the minor axis plane with peaks away from the jet centerline. Different shock patterns were identified along the major and minor axes and the axis switching phenomenon seen in the literature was also reproduced. The axis switching resulted in significantly wider flattened concentration distributions compared with the axisymmetric jet which may require consideration during safety analyses for non-circular nozzles. A scaling factor taking both the nozzle shape and pressure effects into account was then developed to better scale the centerline decay rates for jets from both the square and rectangular nozzles. The present study demonstrates that the nozzle shape effects on the jet spreading should not be overlooked and proper scaling factors are required to collapse the data and calculate decay rates.

Various reduced-order models have been developed to quickly model high pressure underexpanded jets. One example is the two-layer partitioning model which was developed to model underexpanded jets, but it has not been evaluated for high pressure jets with obstacles in the jet flow region. This research describes an improved two-layer partitioning model based on the Abel-Noble equation of state that is applied here to model horizontal jet flows impacting a vertical obstacle with validations against high pressure gas experiments, full CFD simulations and a revised notional nozzle model based on the Abel-Noble equation of state. The improved two-layer partitioning model accurately predicts the gas concentrations on the obstacle for a 15 MPa underexpanded jet while consuming much less computational resources and time compared with the full CFD simulation.

Previously studies of hydrogen releases have focused on hydrogen jets through round nozzles, but leaks are more likely to result in jets through slots. In this study, the concentration decays along the jet centerline were measured for underexpanded jets from rectangular slot nozzles with aspect ratios (AR) of 1–12.6 for storage pressures up to 3.7 MPa. The mass fractions along the jet centerlines decay inversely with distance from the nozzle as with axisymmetric jets but with a faster decay rate. All the data collapses onto a single curve when normalized by the proper factor of the downstream location, except for the jets from square nozzles (AR1). The decay rates for jets from rectangular nozzles with AR greater than unity are significantly larger than for the jets from square nozzles. The jets from square nozzles behave somewhat like the circular jets from circular nozzles due to the symmetric geometry, so the decay rate is closer to that of circular jets relative to the jets from the other rectangular nozzles. This database will be useful for model validations for the modeling of jets from rectangular nozzles.

A two-layer, reduced order model of high pressure hydrogen jets was developed which includes partitioning of the flow between the central core jet region leading to the Mach disk and the supersonic slip region around the core. The flow after the Mach disk is subsonic while the flow around the Mach disk is supersonic with a significant amount of entrained air. This flow structure significantly affects the hydrogen concentration profiles downstream. The predictions of this model are compared to previous experimental data for high pressure hydrogen jets up to 20 MPa and to notional nozzle models and CFD models for pressures up to 35 MPa using ideal gas properties. The results show that this reduced order model gives better predictions of the mole fraction distributions than previous models for highly underexpanded jets. The predicted locations of the 4% lower flammability limit also show that the two-layer model much more accurately predicts the measured locations than the notional nozzle models. The comparisons also show that the CFD model always underpredicts the measured mole fraction concentrations.

Much effort has been made to model hydrogen releases from leaks during potential failures of hydrogen storage systems. A reduced-order jet model can be used to quickly characterize these flows, with low computational cost. Notional nozzle models are often used to avoid modeling the complex shock structures produced by the underexpanded jets by determining an “effective” source to produce the observed downstream trends. In this work, the mean hydrogen concentration fields were measured in a series of subsonic and underexpanded jets using a planar laser Rayleigh scattering system. The experimental data was compared to a reduced order jet model for subsonic flows and a notional nozzle model coupled to the jet model for underexpanded jets. The values of some key model parameters were determined by comparisons with the experimental data. The coupled model was also validated against hydrogen concentrations measurements for 100 and 200 bar hydrogen jets with the predictions agreeing well with data in the literature.

The effect of blower ventilation to disperse hydrogen leaking from a hydrogen fueled vehicle in an emergency was studied numerically with comparison with experimental data to investigate the practicability of rapid hydrogen dispersion to mitigate dangerously high hydrogen concentrations. The hydrogen gas was assumed to leak at 2000 NL/min at the center of the vehicle chassis while the blower was placed on the left side of the vehicle. Steady-state calculations showed that the blower significantly reduced the hydrogen concentrations around the vehicle with the hydrogen concentrations in most of the area below the lower flammability limit. Transient calculations showed that the hydrogen concentration just downstream of the vehicle had a high concentration peak shortly after the ventilation began with the hydrogen then being rapidly dispersed with additional ventilation. More cases with various air capacities and blower shapes were analyzed to find a better blower geometry with a higher blower efficiency. The results show that both a larger air capacity and a smaller blower with a higher air velocity improved the hydrogen dispersion, with the smaller blower shape being the more efficient method because the greater air entrainment into the jet for higher jet velocities improved the hydrogen dispersion, especially when the leak position was known.