No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Experimental Study of LNG Pool Fire on Land in the Field
Abstract
The objective of this experimental research work is to determine several key parameters of an LNG pool fire on land through a field test. The pool fire was conducted in the largest pit at Brayton Fire Training Field in College Station, Texas. The mass burning rate was 0.186 kg/(s m2) using the thermocouple method. Through the analysis of images from a high-speed camera, a puffing frequency of 0.53 Hz and 0.69 Hz was determined for two consecutive oscillations of the flame; the flame tilt was found to be 58 degrees and the flame length was estimated as 25.4 m. The flame velocity field was first studied for an LNG field test with a high-speed camera and the maximum velocity was approximately 6 m/s at the center of the flame. The solid flame model provided a good prediction of thermal radiation at both downwind and crosswind directions, using Federal Energy Regulatory Commission’s recommendation for Surface Emissive Power. In addition, new correlations were developed for mass burning rate, flame length and tilt for LNG pool fires based on the experimental results summarized in this work.
... As it is possible to note, the majority of data have been collected for Dp > 5 m (i.e., larger than ), whereas a dearth of studies can be observed for Dp < 1 m. Generally speaking, even if large-scale experimental scenarios have already been tested [26] and modelled [27], quite unexpectedly, small-and medium-scale experiments are rare, as recently observed also by Zhang et al. (2018) [28]. To this aim, data on the mass evolution over time can be conveniently integrated by qualitative (e.g., images) and quantitative (e.g., thermocouples and pyrometers) analyses on the thermal aspects in the proximity of a pool fire [29]. ...
... As it is possible to note, the majority of data have been collected for Dp > 5 m (i.e., larger than ), whereas a dearth of studies can be observed for Dp < 1 m. Generally speaking, even if large-scale experimental scenarios have already been tested [26] and modelled [27], quite unexpectedly, small-and medium-scale experiments are rare, as recently observed also by Zhang et al. (2018) [28]. To this aim, data on the mass evolution over time can be conveniently integrated by qualitative (e.g., images) and quantitative (e.g., thermocouples and pyrometers) analyses on the thermal aspects in the proximity of a pool fire [29]. ...
... As it is possible to note, the majority of data have been collected for D p > 5 m (i.e., larger than D ∞ ), whereas a dearth of studies can be observed for D p < 1 m. Generally speaking, even if large-scale experimental scenarios have already been tested [26] and modelled [27], quite unexpectedly, small-and medium-scale experiments are rare, as recently observed also by Zhang et al. (2018) [28]. To this aim, data on the mass evolution over time can be conveniently integrated by qualitative (e.g., images) and quantitative (e.g., thermocouples and pyrometers) analyses on the thermal aspects in the proximity of a pool fire [29]. ...
The need for sustainable energy sources has recently promoted the use of liquefied natural gas (LNG) as a low-carbon fuel. Although economic evaluations indicate the transportation of LNG as a convenient solution for long distances between markets and reservoirs, several concerns are still present regarding its safe use and transportation. The preliminary evaluations performed in this work indicate that credible releases deriving from real bunkering operations result in pools having a diameter smaller than 1 m, which has been poorly investigated so far. Hence, an experimental campaign devoted to the characterization of a medium-scale release of LNG was carried out either in the presence or absence of an ignition source. An evaporation rate of 0.005 kg s −1 m −2 was collected for the non-reactive scenario, whereas the measured burning rate was 0.100 kg s −1 m −2. The reduction factor of 20 demonstrates the inaccuracy in the commonly adopted assumption of equality between these values for the LNG pool. Flame morphology was characterized quantitatively and qualitatively, showing a maximum ratio between flame height and flame diameter equal to 2.5 and temperatures up to 1100 K in the proximity of the flame.
... Several LNG pool fire experiments have been carried out to study the characteristics (mass burning rate, flame geometry, flame thermal radiation, and flame velocity field) on different scales of land pool fires (May and McQueen, 1973;AGA, 1974;Raj and Atallah, 1975;Mizner and Eyre, 1982;Nedelka et al., 1989;Burgess and Zabetakis, 1962;Zhang et al., 2018) (Suardin, 2008;Herrera Gomez, 2011). However, due to the low-temperature and flammability of LNG, large-scale tests are difficult and costly to perform. ...
... Considering the factors that may significantly influence the mass burning rate for the application on process safety, the effects of wind velocity and dike depth were also studied. Model outputs were validated using the results of a set of LNG pool fire experiments carried out by Mary Kay O'Connor Process Safety Center (MKOPSC) (Zhang et al., 2018;Herrera Gomez, 2011;Suardin, 2008). A flame geometry analysis software was also developed to support the validation. ...
... A set of LNG experiments carried out by MKOPSC (Zhang et al., 2018;Herrera Gomez, 2011;Suardin, 2008) in consistent experimental and boundary conditions were selected to support model construction, grid sensitivity analysis and model validation. The experimental conditions used in MKCPSC experiments are summarized in Table 2. ...
Liquefied natural gas (LNG) is widely used, because it provides an easy and economic solution to the transport and storage of natural gas (NG), especially on long distances or when transport by pipeline is not viable. The LNG pool fire is one major process safety accident at the LNG facilities according to the report of the U.S. Government Accountability Office. Moreover, due to the high surface emissive power of LNG compared to other hydrocarbon fuels, LNG pool fires have a high potential in causing domino effects and cascading events in process industry. Previous studies developed Computational Fluid Dynamics (CFD) models of LNG pool fires, but the mass burning rate was fixed manually as a model input, not considering that the mass burning rate is determined by the fuel’s physical properties and heat input vaporizing the fuel. In this study, a model of LNG pool fire controlled by material physical properties was developed using fire dynamics simulator (FDS), and was validated against LNG pool fire experiments carried out by Mary Kay O’Connor Process Safety Center (MKOPSC). Statistical performance measurement shows that the model is superior to the semi-empirical approaches in predicting the mass burning rate of LNG pool fire under different pool sizes. A flame geometry analysis software was developed comparing different algorithms, and the centroid method was selected. The results show that the fire model (200 kW/m3 as flame contour) can accurately predict the flame geometry observed in the experimental runs. The influence of wind velocity and dike height on the mass burning rate was also investigated. The results show that the height of the concrete dike is negatively correlated with the mass burning rate of LNG pool fire. The effect of wind velocity on LNG mass burning rate is twofold. The forced convective boundary layer at lower wind velocity promotes LNG combustion and increases the mass burning rate, while higher wind velocities reduce the mass burning rate due to the reduction of the thermal radiation feedback. The findings in this study will contribute to an accurate risk assessment of LNG pool fire accidents in the process industry.
... Quite surprisingly, small and medium-scale experiments are rare, as recently observed by Zhang et al. (2018) [16]. Alternatively, collected data are partially inconsistent [17]. ...
... Quite surprisingly, small and medium-scale experiments are rare, as recently observed by Zhang et al. (2018) [16]. Alternatively, collected data are partially inconsistent [17]. ...
... where the subscript ∞ stands for the mass burning rate ( " ) obtained for an "infinite" diameter of the pool, Af is the flame area, D is flame or the pool diameter, k is the absorption-extinction coefficient, which is approximately 3.0 m -1 for LNG [24] and is the correction coefficient for the beam length. These two parameters are commonly aggregated for the sake of simplicity, as reported by Zhang et al. (2018) [16] for small and medium scale pool fire of LNG. Most of the reported values range between 0.14 m -1 and 0.46 m -1 . ...
Liquefied natural gas (LNG) has been largely indicated as a promising alternative solution for the transportation and storage of natural gas. In the case of accidental release on the ground, a pool fire scenario may occur. Despite the relevance of this accident, due to its likelihood and potential to trigger domino effects, accurate analyses addressing the characterization of pool fires of LNG are still missing.
In this work, the fire dynamic simulator (FDS) has been adopted for the evaluation of the effects of the released amount of fuel and its composition (methane, ethane, and propane), on the thermal and chemical properties of small-scale LNG pool fire. More specifically, the heat release rate, the burning rate, the flame height, and thermal radiation, at different initial conditions, have been evaluated for pool having diameter smaller than 10 m. Safety distances have been calculated for all the investigated conditions, as well.
Results have also been compared with data and correlations retrieved from the current literature. The equation of Thomas seems to work properly for the definition of the height over diameter ratio of the LNG pool fire for all the mixture and the investigated diameters.
The addition of ethane and propane significantly affects the obtained results, especially in terms of radiative thermal radiation peaks, thus indicating the inadequacy of the commonly adopted assumption of pure methane as single, surrogate species for the LNG mixture.
... The effects of a pipeline rupture fire must be considered unique from wildland fire because the fire physics are fundamentally different than a wildland fire and subsequently facilitate a novel disturbance (Bragin and Molkov 2011;Twidwell et al. 2013;Zhang et al. 2018). In a wildland fire, a primary driver of fire effects on vegetation recovery is fire intensity. ...
Pipelines are critical for energy distribution, but incidents causing rupture fires are hazardous. While wildland fires are a natural disturbance, rupture fires are a potential risk and novel disturbance given the greater heat yield constants for fossil fuels, fuel volume, and flaming concentration and duration. We quantified vegetation response to a 2018 rupture fire case study in the montane cordillera of Canada. Plant species, functional groups, ground cover, and live vegetation height were sampled in 2018, 2019, 2020, and 2021 [0, 1, 2, and 3 years since fire (YSF)] in permanent plots stratified by burn severity and compared to the unburned reference plots sampled in 2019. Woody plant species and forb cover in burned plots recovered to levels similar to unburned plots. Litter and bare soil changes relative to YSF suggest trajectories to return to levels similar to unburned plots within 3 to 5 years post-rupture. Plant species richness, evenness, and diversity had also recovered to levels statistically similar to unburned comparisons by the final year of sampling in this study. Plots closest to the rupture epicenter that experienced ‘extreme’ burn had greater botanical dissimilarity from other burn severities or unburned comparisons. Vegetation structure showed significant ( p < 0.0001) recovery with additional growth expected as the overstory re-establishes. The multiple metrics of ecological recovery on 3–5 year trajectories are comparable to published responses to wildland fire in the literature for this ecosystem’s response to fire. The recovery of conifers and soil microbiota should be assessed in the next decade.
... When low-temperature LNG leakage out of the storage tank, a large temperature difference (about 180°C) causes intense heat exchange between LNG and the environment, as a result, the LNG boils off violently [14][15][16]. In the initial stage of the leakage, some of the LNG droplets will flash rapidly into the air, while the others will drop to the ground to form a pool. ...
High expansion foam (Hi-Ex) is recommended to suppress the leakage and diffusion of cryogenic liquid due to its light weight and large volume. However, the disadvantages of low stability and high break rate under environmental conditions are all limited the further application of the Hi-Ex foam in vapor mitigation and fire extinguishing. So that, this paper focus on the effect and mechanism of nanoparticles in stabilizing Hi-Ex foam. Three kinds of nanoparticles with different concentration were selected to evaluate the effect of foam half-life and the mechanism of solid particles on improving the foam stability. The results indicated that different particle concentrations can improve the foam stability to a specific extent, and the maximum improving of half-life can increase by 95.4% in presence of the hydrophilic SiO 2 at 0.5 wt%. Meanwhile, the hydrophilicity, size and morphology of the particles have a specific impact on the foam stability. From the microscopic point of view, it was observed that the bubble size gradually increases with time by two processes of ripening and coalescence and satisfied in a logarithmic distribution. While, the liquid film thickness remarkably decreases in a large decline rate of 77.1% due to foam drainage without particles and the adsorption and accumulation of nanoparticles on foam lamella can provide a spatial barrier for the film thinning and the inter bubble diffusion. So that, the microscopic interaction mechanism of improving the foam stability between the nanoparticles and bubble have been further explored and revealed in these two aspects.
Fires spreading at limited distances one to the other can cause changes in burning rates and flame shapes due to synergistic effects. These may result in more severe consequences and in the generation of specific and more intense escalation vectors, causing a higher probability of domino scenarios. In this study, a specific approach was developed to consider synergistic effects of double pool fires in a quantitative risk assessment (QRA) framework aiming at the assessment of domino accident scenarios. The double pool fire synergy (DPFS) model developed was validated against experimental results obtained using different fuels. A comprehensive framework was developed to address the quantitative risk assessment of escalation leading to domino scenarios, combining the DPFS model, escalation probability estimation for domino accident, and Bayesian network analysis. Three case studies were carried out to analyze the impact of synergistic effects on escalation probability. The results show that synergistic effect of simultaneous multiple pool fires significantly increase the intensity of the escalation vector, and the escalation probability increases up to three orders of magnitude. Furthermore, the inclusion of synergistic effect in the analysis of posterior probability improves the accuracy of identifying the critical units in domino scenarios, contributing to a better selection of the emergency response targets.
Expansion foam is a type of aqueous foam that is widely used for mitigating process safety incident, i.e., suppressing vapor dispersion, controlling fire and decontaminating chemical release. High expansion foam has a high volumetric ratio of air to liquid, from 200 up to around 1,000, making it effective in controlling fires in confined spaces and mitigating the hazards of cryogenic releases. Previous studies have investigated foam expansion ratio and foam stability using a research foam generator; however, the effects of air flow rate and mesh hole diameter were not studied, and there is a lack of mathematical models to predict expansion ratio and production rate. This work will design and build an improved foam generator, allowing the control of air flow rate and mesh hole diameter. The improved foam generator will be used to study the key properties of expansion foam for mitigating process safety incident, and the foam generation mechanism. In addition, predictive models of expansion ratio and production rate will be developed based on the experimental results. The findings of this study will provide a scientific basis for the design of foam generation system and the guideline of hazard mitigation operation for a process safety incident.
This study investigated the influence of different rear bottom wall lengths on the blow-out limit, flame shape, flame mean length, and flame tilt angle of jet diffusion flames under cross-wind conditions. The experimental results reveal that the bottom wall had a significant effect on the flame blow-out limit. The flame blow-out velocity was maximum only when the front bottom wall existed, and all jet diffusion flames had similar blow-out velocity under the influence of the rear bottom wall. Once the length of the rear bottom exceeded a certain value, the jet diffusion flames were prone to lift-off. Moreover, the jet diffusion flames adhered to the bottom wall, which is attributed to the Coanda effect. The flame length significantly increased under the influence of the bottom wall. Additionally, under the influence of the front bottom wall, the flame length was maximum when the cross-wind velocity was higher. Under the influence of different rear bottom wall lengths, the flame tilt angle was divided into two variation regions. Among them, the flame tilt angle affected only by the front bottom wall was minimum. The flame tilt angle was maximum when the rear bottom wall length was 20 cm. Finally, the empirical formula of the jet diffusion flame tilt angle was obtained by analogy analysis and exhibited good correlation with the experimental data.
Hydrocarbon fuels are involved in most fire accidents occurring in industrial facilities, which may lead to catastrophic consequences in terms of life and property losses. The prediction of the related-fire effects can contribute to provide effective protection measures, hence mitigating the consequences of hydrocarbon flames in processing environments. In this regard, there is an essential need for developing state-of-the-art correlations able to accurately estimate the flame shapes and heat fluxes for a wide range of possible hydrocarbon fires. The present paper collects a vast data bank of vertical hydrocarbon fire experiments involving multiple fuels, burner diameters, mass flow rates and wind speeds. Correlations are developed, based on the data collected, to describe the flame-geometry descriptors and the heat fluxes of vertical hydrocarbon flames under different flow regimes.
Evaluating potential hazards caused by accidental LNG release from underwater pipelines or vessels is a significant consideration in marine transportation safety. The aim of this study was to capture the dynamic behavior of LNG jet released under water and to analyze its vapor dispersion characteristics and combustion characteristics on the water surface during different release scenarios. Controlled experiments were conducted where LNG was jet released from a cryogenic storage tank. The dynamic process of LNG being jet released from orifices of different sizes and shapes, as well as the rising plume structure, were captured by a high-speed camera. The leakage flow rate and pipeline pressure were recorded by a flow meter and pressure gauge, respectively. The concentration distribution that emanated from the water surface was measured utilizing methane sensors in different positions with various wind speeds. The flame combustion characteristics of LNG vapor clouds, which immediately ignited upon the enclosed water tank, were also recorded. Additionally, the mass burning rate of the flame on the water surface was evaluated, and a new correlation between the ratio of flame length and width was established. The results indicated a large dimensionless heat release rate (Q*) and a continuous release flow rate in a limited burning area. This study could provide greater understanding of the mechanisms of LNG release and combustion behavior under water.
Liquefied natural gas (LNG) trade has increased globally; therefore, assessments of the hazards of its accidental release and associated consequences must be conducted to ensure LNG security during marine transportation. This study aims to understand the dynamic behavior occurring when an LNG jet is released underwater and the combustion behavior of a flammable vapor cloud on the water’s surface with airflows from 0 to 4 m/s. A series of controlled LNG vertical jet release experiments were conducted using a cryogenic storage tank with three orifices. Various instruments were employed to measure the flow rate and pressure in pipelines during different leakage scenarios, and the emanating LNG vapor clouds were immediately ignited with the mass loss rate of 0.016, 0.037, and 0.049 kg/s in three orifices, respectively. The flame behavior was recorded by a video camera. With respect to flame length rapidly decreased with crosswinds of 0−2 m/s and then gradually decreased to a constant value with the velocities increase to 4 m/s. A dimensionless number of Ri was employed to analyze the relative magnitude between the buoyancy force and transverse flow. The flame tilt angle was found in accordance trend with flame length for first increased and further reach to become constant at Ri⁻¹ increase to 2. As existing correlations provide an overestimation, a new correlation was established to describe the flame length as a function of crosswind speed, and a good prediction was found for measured tilt angles with the correlated values.
Digital particle image velocimetry (DPIV) is a non-intrusive analysis technique that is very popular for mapping flows quantitatively. To get accurate results, in particular in complex flow fields, a number of challenges have to be faced and solved: The quality of the flow measurements is affected by computational details such as image pre-conditioning, sub-pixel peak estimators, data validation procedures, interpolation algorithms and smoothing methods. The accuracy of several algorithms was determined and the best performing methods were implemented in a user-friendly open-source tool for performing DPIV flow analysis in Matlab.
A review of literature and published full-scale measurements has been undertaken in order to assess the current status of the modelling of thermal radiation from hydrocarbon pool fires. Based on the review, a semi-empirical model was developed, in which the pool fire is assumed to radiate in two layers; a high emissive power, clean burning zone, at the base, with a smoky, obscured layer above. The choice and development of model correlations was made through comparison against a wide range of field-trial data, which was drawn together to form a validation database. The review also enabled a property database to be produced, containing burning rate and surface emissive power data for a broad range of liquid hydrocarbon fuels. The uncertainty in the application of semi-empirical pool fire modelling is discussed with regard to its use in assessing thermal radiation effects and estimating flame dimensions.
Data for estimating the burning rate and heat output of large pool fires (diameter ≳ 0.2 m) are compiled and computational
equations presented. Since a large scatter in the reported data is noted, attention is also focused on areas where further
research is most needed in order to improve predictability.
Previous research suggests that high expansion foam with an expansion ratio of 500 to 1 is one of the best options for controlling liquefied natural gas (LNG) pool fire on land. However, its effectiveness heavily depends on the foam application rate, foam generator location, and the design of LNG spill containment dike. Examination of these factors is necessary to achieve the maximum benefit for applying HEX on LNG pool fires. While theoretical study of the effects of foam on LNG fires is important, the complicated phenomena involved in LNG pool fire and foam application increase the need for LNG field experimentation. Therefore, five LNG experiments were conducted at Texas A&M University's Brayton Fire Training Field. ANGUS FIRE provided Expandol solution to form 500 to 1 high expansion foam (HEX) and its latest LNG Turbex Fixed High Expansion Foam Generators. In this paper, data collected during five experiments are presented and analyzed. The effectiveness of high expansion foam for controlling LNG pool fires with various application rates at two different types of containment pits is discussed. LNG fire behaviors and the effects of dike wall height are also presented and discussed.
The cylindrical tanks are widely used to store liquids in the industry, among which LNG poses a unique hazard to personnel, assets and surrounding community due to its flammability and cryogenic nature. There are regulations, such as NFPA 59A, to require a dike for mitigating the risk in case of a catastrophic failure of LNG tanks. The dike helps to control pool spreading, limit vaporization and mitigate vapor cloud, but this effect would be undermined if liquids overtop from dikes. Previous studies involved small-scale tests and a medium-scale quadrant setup. In this study, the overtopping study was further enhanced by investigating dike and tank configuration, dike materials, liquid types through both lab-scale and field tests. A CFD model was developed to understand the physics, which was compared with experimental results. The findings can help to provide science-based guidelines for overtopping risk assessment and design of storage tanks and dikes.
This paper reviews the physics and correlations for the burning behaviour of pool fires in wind, discussing also challenges for future research on this topic. In the past decades, the burning behaviour of pool fires in still air, which is solely buoyancy driven, has been extensively studied. These studies are primarily focused on scale, radiation, soot, pressure and gravity effects. However, these phenomena and physics change significantly with much more complexity in the presence of wind, with regard to heat feedback and burning rate; flame geometrical characteristics; flame turbulence, soot and radiation emission. Remarkable progress has been made in understanding the behaviour of the heat feedback and burning rate, flame tilt, flame length and flame base drag of wind-blown pool fires. Several semi-empirical correlations have been proposed for these quantities, based on experimental data and the physically dimensional analysis. However, for wind-blown pool fires, the flame soot and radiation emission coupling with complex flow turbulence scales due to the interaction of buoyancy with wind still require more basic research. All these processes are more challenging especially for wind-blown large scale pool fires, which require knowledge and understanding of the physics, especially for establishing evaluation methodologies of their hazard and adverse impact.
This paper reports a comprehensive experimental study on evolutions of burning rate and flame tilting of medium-to-large size pool fires with cross flows. Square pool fires with length of 25, 35, 45, 60 and 70 cm were employed with heptane as fuel, whose burning behaviors are in radiation-controlled optical-thin regime in still air. The evolutions of their burning rates and flame tilt angles were measured with cross flow air speed from 0 to 4.5 m/s, in which range the data is still lacking in the literature which mostly concerned relative smaller pools (up to 25-30 cm) in laboratory or relative larger pools (>1.0 m) by outdoor burning where the wind is hard to maintain. It was found that there were several transitions for the burning rate with increase in cross flow air speed in this range for different size pool fires. For relative smaller pools (25 cm and 35 cm), the burning rate increased and then decreased; meanwhile for the medium pool (45 cm), the burning rate first increased then decreased, and further increased then finally decreased. However, for the relative larger pools (60 cm and 70 cm), the burning rate first increased then decreased and finally increased with flow speed up to 4.5 m/s. These transitions were discussed based on physically the change of the controlling mechanisms due to heat feedback or the aerodynamic extinction at the flame leading edge (Damköhler number effect). The corresponding cross flow air speeds for these transitions correlated well with buoyancy induced velocity scaled by [(gD)^0.5] based on Froude number. The flame tilt angles measured were compared with previous correlations to validate their applicability in this range. This work provided a comprehensive data base revealing the transitions of burning behaviors of medium-to-large pool fires with increase in cross flows air speed.
The flammable vapor cloud is the primary hazard caused by a liquefied natural gas (LNG) spill on land. If it is not properly mitigated, an ignition of the vapor cloud will result in a fire or explosion hazard. High expansion foam is recommended by NFPA 11 and NFPA 471 for LNG spill hazards mitigation. This work studied the physical interaction of the LNG and expansion foam system using a foam generator and a foam test apparatus that are built in-house. The performance of the foam generator was characterized in terms of the foam expansion ratio and generation rate. The temperature profile in the foam zone quantitatively confirmed the warming effect of foam on LNG vapor. The foam breaking rate was determined at the initial stage and steady state. The boil-off effect was studied quantitatively with more details using this novel foam generator. The vapor channel formation and vapor concentration were also investigated for the mitigation effect.
Jet fires are one of the major hazards in the oil and gas industry, and half of the reported jet fires caused a domino effect that enlarged the incident. Previous work with propane focused on vertical jet fires. In this work, both horizontal and vertical fires were studied through 21 tests conducted at the Brayton Fire Training Field. The fires were sustained by either vapor or liquid propane. The visualization of flames using both video and infrared cameras identified that cylindrical shape was a good description of vertical and horizontal jet fires. For the vertical fires, a correlation was developed for radiation against distance from the flame axis. For horizontal fires, the solid flame model was used to predict radiation; the selected parameters were evaluated by statistical performance measures. The effects of vapor flow rate, surface emissive power, and the fraction of radiation were also studied.
Jet fires are among the least severe fires in terms of direct effects, but are very important in terms of risk assessment due to the potential escalation of the incident by impingement or engulfment of the jet fire on the surrounding vessel, pipework or other components. This paper focuses on the determination of the main geometrical features (flame shape, length, and width) of medium-scale horizontal jet fires in air. The study is based on the experimental results of LPG jet fire released from a horizontal pipe of 1.9 cm diameter at different flow rates, with vapor flows, reaching a flame size of 1 to 10 m long. For each test, visible and infrared visualizations were recorded. First, the two visualization techniques are compared with each other, and with the different methods of flame shape determination available in the literature. The flame detected with each technique, both instantaneous and averaged, is then compared with basic flame shapes.
Numerical simulations have been conducted for a range of liquefied natural gas (LNG) pool fires on land and water using FireFOAM, the large eddy simulation (LES)–based fire simulation code within the framework of open source computational fluid dynamics (CFD) toolbox OpenFOAM®. The studied pool diameters range from 14–400 m with cross winds from 1.6–9.6 m/s. The code uses the extended eddy dissipation concept (EDC) and a newly developed soot model based on the laminar smoke point concept. For the low-temperature (111.65–200 K) thermodynamic data of natural gas, the nine-coefficient correlations in the NASA thermodynamic database are used. Comparison between the predictions and measurements were carried out for the first four cases where full-scale test data are available. For all cases, the variations of flame length, tilt angle, and surface emissive power with LNG pool diameters are analyzed. New nonlinear correlations for predicting length-to-diameter ratio and tilt angle are also proposed.
With increasing consumption of natural gas, the safety of liquefied natural gas (LNG) utilization has become an issue that requires a comprehensive study on the risk of LNG spillage in facilities with mitigation measures. The immediate hazard associated with an LNG spill is the vapor hazard, i.e., a flammable vapor cloud at the ground level, due to rapid vaporization and dense gas behavior. It was believed that high expansion foam mitigated LNG vapor hazard through warming effect (raising vapor buoyancy), but the boil-off effect increased vaporization rate due to the heat from water drainage of foam. This work reveals the existence of blocking effect (blocking convection and radiation to the pool) to reduce vaporization rate. The blanketing effect on source term (vaporization rate) is a combination of boil-off and blocking effect, which was quantitatively studied through seven tests conducted in a wind tunnel with liquid nitrogen. Since the blocking effect reduces more heat to the pool than the boil-off effect adds, the blanketing effect contributes to the net reduction of heat convection and radiation to the pool by 70%. Water drainage rate of high expansion foam is essential to determine the effectiveness of blanketing effect, since water provides the boil-off effect.
Boilover is defined as a violent ejection of fuel due to the vaporization of a water sub-layer, resulting in an enormous fire enlargement and formation of fireball and ground fire. This paper focuses on the physical principles behind the thin-layer boilover phenomenon, and on the improvement of the thin-layer boilover modeling. Small scale field and laboratory experiments, with a mixture of diesel and oil, and reservoirs ranging from 0.08 to 0.3 m, have been performed. High-speed visualizations and image processing, in parallel to temperature and mass loss measurements allows to better understand the water boiling, and the consequent flame enlargement. The modeling of the boilover period is done through the calculation of the pre-boilover burnt mass ratio and the boilover intensity based on the mass loss and on the flame enlargement.
Geometric and thermal data, obtained from a series of large outdoor jet fire experiments, were used to estimate the thermal radiation intensity from the flames towards targets located at diverse distances. Vertical turbulent sonic and subsonic exit velocity propane jet fires, up to 10.3 m in length released in still air, were studied. The temperatures of the flame surface and the surface emissive power of the flame were also analysed by processing infrared images. Thermal radiation intensity was estimated by applying the solid flame model in both one-zone and multiple-zone configurations (taking into account the variation of surface emissive power), considering the flame as a cylinder defined by the 800 K isotherm. Experimental and predicted thermal radiation intensity values were compared.
This work aims to estimate the effects of thin-layer boilover on flame geometry and dynamics. A series of large scale experiments (in pools ranging from 1.5 to 6 m in diameter) were performed using gasoline and diesel as fuel. As expected, only diesel showed evidence of this phenomenon. This article presents a summary of the results obtained for flame height, tilt and pulsation. Flame height increases during water ebullition, though the increase is no longer detectable when wind speed exceeds certain values. Correlations previously presented in the literature to predict flame length and tilt were modified in order to fit the results obtained during thin-layer boilover. However, the influence on flame tilt is not as great and the equations for the stationary period seem suitable for the entire fire. Results of flame pulsation during the stationary period fill the gap in the literature for fires between 1.5 and 6 m and fit previous correlations. On the other hand, during ebullition, the flame pulsates faster, as air entrainment is greater and, as one would expect, this effect decreases with pool size. A new equation for estimating pulsation frequency during boilover is proposed.
Uncontrolled fires produce flames where the initial momentum of the fuel is low compared with the momentum produced by buoyancy. The heights of such flames with wood as the fuel are examined and discussed in terms of both a dimensional analysis and the entrainment of air into the turbulent flame. They are then compared with other experiments on the flow of hot gases.
Some recent experiments on the effects of wind on such flames are also reported.
Caracteristiques de combustion d'une nappe d'huile: vitesse de combustion, diametre de la nappe, hauteur de la flamme. Calcul du rayonnement thermique, evaluation de l'absorption atmospherique, presentation des facteurs de risques
The radiation intensity at a given distance depends mainly on the radiative power and the flame's size and shape. Considerable literature describing both experimental and theoretical studies of thermal radiation from flames is available. Even so, predicting the radiant power of large flames is still subject to considerable uncertainty, because some parameters associated with large turbulent diffusion flames cannot be determined accurately for a given fire. A series of outdoor large pool-fire experiments were performed using gasoline and diesel fuels lying above a layer of water. Five concentric circular pools made of reinforced concrete (1.5, 3, 4, 5, and 6 m in diameter) were used. The experiments were filmed with at least two video cameras registering visible light (VHS) and a thermographic camera (IR). In this study, thermographic images were used to determine the flames' distribution of emissive power, the mean emissive power, and the flame's irradiance. The contribution of each part of the flame to the total radiated energy was analyzed. A method is presented combining the IR images and the visible images; it offers further insight into the relationship between the heat emitted by the luminous part and the obscured, nonluminous, part of the flame.
An experimental technique is described for accurately measuring the steady-state fuel consumption rates in small-scale pool fires, less than 7 cm diameter. The technique is applied to studying ethanol fires burning in cylindrical vessels constructed from various materials. The results indicate that the distance between the top of a vessel and the fuel surface profoundly influences the properties of liquid pool fires, including their structure and their burning rates. For combustion in glass cylinders, the burning rates decrease exponentially with increasing freeboard until a critical ullage is attained. At this ullage, the fuel begins to burn on the inside of the vessel, and the burning rate tends to increase slightly. With a further increase of the lip height, flame instabilities develop leading ultimately to flame self-extinction. The exponential decline in fuel consumption with the lip height depends strongly on the vessel material of construction. For fires in cylinders constructed from better conducting materials (copper and mild steel), the ethanol starts to boil beyond a certain ullage. The appearance of this phenomenon redefines the fuel consumption curve. Finally, free convection leads to non-negligible heat losses, especially from the more conducting copper and steel vessels, with the burning rates becoming dependent on the outside surface area of the cylinders.
In flames produced by freely burning fuel, buoyancy may play an important role in determining the speed of the gases in the flame zone and hence the flame height. Measurements have been made of the height of flames from burning cribs of wood on a square horizontal base and a few for two other arrangements. The results are consistent with a dimensionless analysis, leading, for one particular fuel system, to the functional equation L/D = f(Q2/gD5), where L is the flame height, D the linear dimension of the fire or orifice, Q the volumetric flow rate of gaseous fuel at ambient temperature and g the acceleration due to gravity. In turbulent fuel jets L/D is a constant for a given fuel which is shown theoretically to be a limiting case of this relation.
The unique properties of expansion foam in blanketing the surface of most hydrocarbon fuels have made it possible to be used as a mitigation measure against a boiling and evaporating pool of flammable gases and subsequent pool fires. Because of this fire suppression characteristic, the liquefied natural gas (LNG) industry has identified expansion foam as one of its safety provisions for pool fire suppression. However, the effectiveness and key parameters of foam in controlling LNG fires have not been thoroughly investigated from previous field tests. In this paper, we investigated the effects of foam application on LNG pool fires through outdoor spill experiments at the Brayton Fire Training Field. The primary objectives of this study are to identify the foam effectiveness in suppressing LNG pool fires and to determine the thermal exclusion zone, by investigating temperature changes of foam and fire, profiles of radiant heat flux, and fire height changes with foam. Additionally, a schematic model of a LNG−foam system with fire for theoretical modeling was also developed. Results showed that expansion foam has positive effects on reducing flame height and radiant heat flux by decreasing heat release and radiant heat feedback from the LNG pool fire, ultimately reducing the safe separation distance. Through extensive data analysis, we also identified several key parameters, such as the minimum effective foam depth and the mass-burning rate of LNG with applied foam. Results from this study can be used to design an effective expansion foam system as well as to develop defensive measures and emergency response plans for mitigating the consequences of LNG releases.
Current models for evaluating the exclusion (hazard) zones around liquefied natural gas (LNG) fires in both U.S. Regulations and NFPA-59A standard are prescriptive and require the consideration of large LNG releases. These models do not consider the effects of the combustion dynamics associated with large-size pool burning. Oxygen starvation in the core of LNG fires of diameters ≳ 35 m leads to the formation of nonluminous, cold soot (smoke) resulting in a reduction of thermal energy radiated by the fire to the surroundings. The net effect is smaller (calculated) thermal hazard distances for exposure to people (by factors of 2 or 3 compared to results ignoring this phenomenon). Available large-scale LNG fire test information is reviewed to quantify the effect of this phenomenon. This paper also discusses the common mistakes made in calculating the thermal radiation hazard distances around large fires by using, for the energy radiated from the fire, a constant percentage of energy generated by combustion. The criteria for setting thermal radiation hazard zones around large hydrocarbon fires are also reviewed. © 2005 American Institute of Chemical Engineers Process Saf Prog, 2005
A convenient, widely accessible radiation model has been developed to simplify calculations of radiant heat transfer from pool fires. The non-homogeneous, non-isothermal fires are described by an equivalent homogeneous, isothermal, spectrally gray volume of flame gases defined by a composite flame shape. Input parameters for this model are obtained with conventional camers and radiometers. Those parameters are 1) the axisymmetric flame shape for particular fuel scale, and configuration derived from photographs; 2) an averaged absorption-emission coefficient, , from flame transmittance measurements and 3) an averaged flame temperature, , from measurements with a radiometer viewing the flames through a horizontal slit. The averaging of radiative properties accounts for most flame inhomogeneities.Simple analytic expressions are provided for calculations of radiative feedback to the fuel surface and radiant transfer to targets away from the fire. Results of those analytic calculations are in good agreement with exact numerical computations and with experiments.The simplified model is verified with measurements on 381-mm and 730-mm diameter PMMA pool fires. A two-fold variation in average pyrolysis rates is induced at the 381-mm scale by systematic variation of the distance between the fuel surface and the container lip.Composite photographic flame shapes were obtained for seven 381-mm diameter pool fires with lip sizes between 0 and 76 mm, and for the 730 mm pool with a 13-mm lip. Identifical averaged radiation property parameters , were used with those respective flame shapes. Calculated radiant feedback and radiation transfer to target locations are in good agreement with measurement for the range of pool fires investigated.
Simultaneous temporally and spatially resolved, 2-D velocity fields are obtained using Particle Image Velocimetry (PIV) in a one-meter diameter methane fire. The flow rate of methane is 0.066 kg/m2-s, comparable to fuel burning rates in a large JP8 pool fire. Raw PIV images are recorded with 35 mm cinematography at 200 images/s. They are digitized and post-processed to obtain velocity data for a region ∼0.8 m high by 1 m wide centered on the centerline of the flame and extending from just above the surface of the burner to include the fuel core, near-field combusting zones, and surrounding air. The data cover 11 puff cycles of the fire. Instantaneous, phase-, and time-averaged 2-D velocity plots (103 × 82 vectors) are obtained for each of 1331 time-planes (121 time-planes per puff cycle) spaced 5 ms apart. Each vector represents a statistical estimate of the velocity in 2.1 cm by 2.1 cm by 0.8 cm volumes, which are overlapped by 50% in the vector plots. Time-averaged turbulent statistics (, , & ) are also presented. Boundary conditions have been carefully measured and the results are intended for validation of numerical simulations of the fire behavior. The results clearly show the dominant effect of puffing, measured at 1.65 cycles/s for this fire, on the temporal and spatial development of the velocity field.
Natural gas diffusion flames stabilized on 0.10, 0.19 and 0.50 m. diameter porous bed burners have been studied for heat release rates ranging from 10 to 200 kW. Flame heights were measured from video tape recordings and by eye averaged techniques. The dependence of flame height on a dimensionless heat addition parameter shows a transition for values of the parameter around unity. For flames taller than three burner diameters, the initial diameter of the fire does not affect the length of the flame whereas for short flames, initial geometry becomes important. Another prominent feature of these flames is the presence of large scale axisymmetric structures which are formed close to the burner surface with more or less regular frequency and which rise through the flame region. These structures are responsible for the fluctuations of the flame top and strongly influence the geometry of the flame.
The NFPA 59A Standard and the Federal Regulation, 49 CFR Part 193, stipulate a level of 5 kW/m2 as the criterion for determining the hazard distance to people exposure from a LNG fire. Another regulation (24CFR, Section 51.204) while stipulating a lower exposure limit of 1.42 kW/m2 provides administrative relief from the regulation if mitigation measures are provided. Several countries in Europe and the Far East have adopted both a specified heat flux value (generally, 5 kW/m2) as well as modified dose criteria for human exposure hazard calculation in risk assessments. In some cases, the regulations in Europe require the use of lower values for children and physically challenged persons.This paper reviews the available literature on the phenomenon of skin burn caused by radiant heat exposure. The associated thermal and spectral properties of human skin are reviewed. The basis for regulatory setting, of 5 kW/m2 and other exposure criteria (as a part of hazard and risk calculations) for evaluating distances to hazards from the exposure of people to radiant heat effects of large fires, is evaluated. An example calculation is provided to show the extent of reduction in the hazard distance to specified radiant heat flux from a fire when the spectral reflection and absorption properties of skin are considered with and without the inclusion of the mitigating effects of clothing. The results indicate that hazard distances calculated including the reflective and band absorptive properties (in IR wavelength) of skin results in a reduction of between 30 and 50% in the hazard distances obtained with current methodology, which ignores these effects. Unfortunately, there are no test results, from full-scale human-exposure-to-IR radiation, with which these predictions can be compared.
Liquefied Natural Gas (LNG) hazards include LNG flammable vapor dispersion and LNG pool fire thermal radiation. A large LNG pool fire emits high thermal radiation thus preventing fire fighters from approaching and extinguishing the fire. One of the strategies used in the LNG industry and recommended by federal regulation National Fire Protection Association (NFPA) 59A is to use expansion foam to suppress LNG vapors and to control LNG fire by reducing the fire size. In its application, expansion foam effectiveness heavily depends on application rate, generator location, and LNG containment pit design. Complicated phenomena involved and previous studies have not completely filled the gaps increases the needs for LNG field experiments involving expansion foam. In addition, alternative LNG vapor dispersion and pool fire suppression methodology, Foamglas® pool fire suppression (PFS), is investigated as well. This dissertation details the research and experiment development. Results regarding important phenomena are presented and discussed. Foamglas® PFS effectiveness is described. Recommendations for advancing current guidelines in LNG vapor dispersion and pool fire suppression methods are developed. The gaps are presented as the future work and recommendation on how to do the experiment better in the future. This will benefit LNG industries to enhance its safety system and to make LNG facilities safer.
Spectral description of thermal emission from fires provides a fundamental basis on which the fire thermal radiation hazard assessment models can be developed. Several field experiments were conducted during the 1970s and 1980s to measure the thermal radiation field surrounding LNG fires. Most of these tests involved the measurement of fire thermal radiation to objects outside the fire envelope using either narrow-angle or wide-angle radiometers. Extrapolating the wide-angle radiometer data without understanding the nature of fire emission is prone to errors. Spectral emissions from LNG fires have been recorded in four test series conducted with LNG fires on different substrates and of different diameters. These include the AGA test series of LNG fires on land of diameters 1.8 and 6 m, 35 m diameter fire on an insulated concrete dike in the Montoir tests conducted by Gaz de France, a 1976 test with 13 m diameter and the 1980 tests with 10 m diameter LNG fire on water carried out at China Lake, CA. The spectral data from the Montoir test series have not been published in technical journals; only recently has some data from this series have become available. This paper presents the details of the LNG fire spectral data from, primarily, the China Lake test series, their analysis and results. Available data from other test series are also discussed.
A number of experimental investigations of LNG fires (of sizes 35 m diameter and smaller) were undertaken, world wide, during the 1970s and 1980s to study their physical and radiative characteristics. This paper reviews the published data from several of these tests including from the largest test to date, the 35 m, Montoir tests. Also reviewed in this paper is the state of the art in modeling LNG pool and vapor fires, including thermal radiation hazard modeling. The review is limited to considering the integral and semi-empirical models (solid flame and point source); CFD models are not reviewed. Several aspects of modeling LNG fires are reviewed including, the physical characteristics, such as the (visible) fire size and shape, tilt and drag in windy conditions, smoke production, radiant thermal output, etc., and the consideration of experimental data in the models. Comparisons of model results with experimental data are indicated and current deficiencies in modeling are discussed. The requirements in the US and European regulations related to LNG fire hazard assessment are reviewed, in brief, in the light of model inaccuracies, criteria for hazards to people and structures, and the effects of mitigating circumstances. The paper identifies: (i) critical parameters for which there exist no data, (ii) uncertainties and unknowns in modeling and (iii) deficiencies and gaps in current regulatory recipes for predicting hazards.
2013. (4) Raj, P. K. A. Review of the Criteria for People Exposure to Radiant Heat Flux from Fires
- B Zhang
- Y Liu
- W Zhu
- N Gopalaswami
- M S Mannan
Liquefied Natural Gas Facilities; Federal Safety Standards, 1980.
(3) National Fire Protection Association. NFPA 59A: Standard for
the Production, Storage, and Handling of Liquefied Natural Gas (LNG).
2013.
(4) Raj, P. K. A. Review of the Criteria for People Exposure to
Radiant Heat Flux from Fires. J. Hazard. Mater. 2008, 159 (1), 61.
(5) Zhang, B.; Liu, Y.; Zhu, W.; Gopalaswami, N.; Mannan, M. S.
Experimental Study of Bund Overtopping Caused by a Catastrophic
Failure of Tanks. Ind. Eng. Chem. Res. 2017, 56, 12227.
(6) Zhang, B.; Liu, Y.; Olewski, T.; Vechot, L.; Mannan, M. S.
1380. (31) Thielicke, W.; Stamhuis, E. J. PIVlab − Towards User-Friendly, Affordable and Accurate Digital Particle Image Velocimetry in MATLAB
- I E Chem
- R C Gonzalez
- R E Woods
- S L Eddins
- D Laboureur
- L Aprin
- A Osmont
- J M Buchlin
- P Rambaud
I. Chem. E. Symposium Series 1982, 71, 147−163.
(26) Gomez, C. Experiments for the Measurement of LNG Mass
Burning Rates; Texas A&M University: College Station, TX, 2011.
(27) American Gas Association. LNG Safety Program: Interim Report
on Phase II Work; Project IS-3-1; American Gas Association:
Columbus, Ohio, 1974.
(28) Gonzalez, R. C.; Woods, R. E.; Eddins, S. L. Digital Image
Processing Using Matlab; Education. 2004; p 609.
(29) Otsu, N. A Threshold Selection Method from Gray-Level
Histograms. IEEE Trans. Syst. Man Cyber 1979, 9, 62−66.
(30) Laboureur, D.; Aprin, L.; Osmont, A.; Buchlin, J. M.; Rambaud,
P. Small Scale Thin-Layer Boilover Experiments: Physical Understanding and Modeling of the Water Sub-Layer Boiling and the Flame
Enlargement. J. Loss Prev. Process Ind. 2013, 26 (6), 1380.
(31) Thielicke, W.; Stamhuis, E. J. PIVlab − Towards User-Friendly,
Affordable and Accurate Digital Particle Image Velocimetry in
MATLAB. J. Open Res. Softw. 2014, 2 (1), No. e30, DOI: 10.5334/
jors.bl.
(32) Thielicke, W. The Flapping Flight of Birds − Analysis and
Application; University of Groningen: Groningen, Netherlands, 2014.
(33) Horvath, I. A. Extreme PIV Applications: Simultaneous and
Instantaneous Velocity and Concentration Measurements on Model and
Real Scale Car Park Fire Scenarios;
Experimental Tests on Large Hydrocarbon Pool Fires
- G Mangialavori
- F Rubino
Mangialavori, G.; Rubino, F. Experimental Tests on Large
Hydrocarbon Pool Fires. In 7th International Symposium on Loss
Prevention and Safety Promotion in the Process Industries; Italy, 1992.
Experiments for the Measurement of LNG Mass Burning Rates
- C Gomez
Gomez, C. Experiments for the Measurement of LNG Mass
Burning Rates;
- R C Gonzalez
- R E Woods
- S L Eddins
Gonzalez, R. C.; Woods, R. E.; Eddins, S. L. Digital Image
Processing Using Matlab; Education. 2004; p 609.
33) Horvath, I. A. Extreme PIV Applications: Simultaneous and Instantaneous Velocity and Concentration Measurements on Model and Real Scale Car Park Fire Scenarios
- W Thielicke
Thielicke, W. The Flapping Flight of Birds − Analysis and
Application; University of Groningen: Groningen, Netherlands, 2014.
(33) Horvath, I. A. Extreme PIV Applications: Simultaneous and
Instantaneous Velocity and Concentration Measurements on Model and
Real Scale Car Park Fire Scenarios;
Liquefied Natural Gas Hazards Mitigation with High Expansion Foam
- B Zhang
Zhang, B. Liquefied Natural Gas Hazards Mitigation with High
Expansion Foam;
- E E Zukoski
- B M Cetegen
- T Kubota
- P H Thomas
- C T Webster
- M M Raftery
Zukoski, E. E.; Cetegen, B. M.; Kubota, T. Visible Structure of
Buoyant Diffusion Flames. Symp. Combust., [Proc.] 1985, 20 (1), 361.
(46) Thomas, P. H.; Webster, C. T.; Raftery, M. M. Some
Experiments on Buoyant Diffusion Flames. Combust. Flame 1961, 5,
359.
Experimental Characterization and Modeling of Hazards : BLEVE and Boilover, The von Karman Institute for Fluid Dynamics
- D Laboureur
Laboureur, D. Experimental Characterization and Modeling of
Hazards : BLEVE and Boilover, The von Karman Institute for Fluid
Dynamics; Sint-Genesius-Rode: Belgium, 2012.