Search

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. © 2019 Elsevier Ltd

Modelling the view factor, F, of a ‘grain-like’ observer near a tilted pool fire assumed to be cylindrical in shape, requires the observer (on the same ground level as the pool fire) to be as close as possible to the flame surface so that the flame surface is viewed as a plane. The orientation of the observer is to receive a maximum view, with the view factor integrated over the flame area seen by the observer. The derived expression for F was expressed in terms of standard functions and sensitive parameters: α, defined to ensure the observer receives a maximum view of the flame surface and β, the angle of inclination of the differential observer plane. Results from a planar approximation to F are compared with those of PHAST 7.2 simulated results in MATLAB. The values of the planar approximation to F was found to increase with increasing β. This suggests that the larger the value of β, the more the radiation received by a near observer. For β = 48 . 961 ∘ , horizontal distance, X = 30 m, α = 0.01 , flame length, L = 12.14 m, tilt angle, θ = 55 . 17 ∘ , and pool diameter, D = 5 m: planar and PHAST 7.2 approximations to F were found to coincide up to 6 significant digits and differ by 3.2 × 10 − 8 . The close agreement between both approximations depend heavily on the choice of flame properties and sensitive parameters. Interestingly, this result may depend sensitively on the choice of parameters, leading to prediction of planar approximation to F much higher or lower than those of PHAST 7.2 approximations. The advantage of this approach is that a knowledge of the plane where the largest concentration of radiation is located will help minimise loss of lives to fire hazards and improve the efficiency of risk safety assessment/management.

International audience; The paper presents experimental results indicating the influence of the fire heat release rate (HRR) evolution on the fire-induced pressure variation in air-tight compartments, which increasingly appear in modern buildings. The fire evolution is stipulated according to a power law with variations in fire growth rate coefficient, maximum steady state HRR, fire growth exponent and fire duration. It is illustrated that the higher the fire growth rate coefficient, the higher the over-pressure/under-pressure peaks become. The peak values approach a maximum level when the fire growth rate coefficient exceeds a certain level (0.0926 kW/s2 in the present cases). Besides, the over-pressure and under-pressure peak values increase monotonously with the increase of the maximum steady state HRR. For cases with fire growth exponent below 1, the pressure reaches peak values before the HRR reaches a steady state (maximum value or zero) because the net heat gained in the gas phase already reaches its peak values then. In cases with fire growth exponent above 1, the pressure reaches peak values at the moment when the HRR reaches the steady state. Finally, in one test low-frequency oscillatory behavior is illustrated for the first time with gaseous fuel. The present work indicates the significance of the fire-induced pressure risk and provides data for model validation.

A water system, consisting of several water mist nozzles, has been installed in a reduced-scale tunnel. Its effectiveness in blocking fire-induced smoke and heat is tested, with and without longitudinal ventilation. A total of 14 fire tests have been carried out, with 250 ml methanol in an iron tray (25 cm × 20 cm) as fuel. Temperatures have been measured by 30 thermocouples, located upstream and downstream of the fire location. The aim is to assess the effectiveness of the water system in preventing smoke spread and in reducing the temperature in the tunnel. Interaction of the water with the fire is avoided. The impact of water pressure, ventilation velocity and nozzle arrangement on the effectiveness in smoke blocking and temperature reduction is discussed. The result confirms that the water system effectively reduces the temperatures and prevents smoke spreading in the absence of longitudinal ventilation. However, strong longitudinal ventilation (0.8 m/s ventilation velocity in the reduced-scale tunnel, corresponding to critical velocity in full-scale (1:10) tunnel) reduces the effectiveness in blocking the smoke spreading by the water system, although the temperature reduction downstream the water system remains in place. Higher water pressure makes the cooling effect stronger, because more and smaller water droplets are injected into the tunnel. For a given level of water pressure level, the impact of the nozzle row configuration is small in the tests.

The effects of convective flows in n-octane pool fires in an ice cavity were investigated and it was found that a new set of parameters to the classical problem of bounded pool fires arises under these unique conditions. To systematically understand these parameters, two sets of experiments were performed by burning n-octane in cylindrically shaped ice cavities of 5.7 cm diameter. The first set of experiments was intended to provide a clear understanding of the geometry change of the cavity and displacement of the fuel layer. The results of these experiments showed that the rate of melting of the ice walls were higher in areas where the fuel layer was in contact with ice than in places where the flame was present. Due to the melting of the ice walls, a ring-shaped void was formed around the perimeter of the cavity. In the second set of experiments, the change in the temperature of the fuel layer was measured by use of multiple thermocouples at different locations inside the ice cavity. The results of the temperature analysis showed that the lateral temperature gradient of the fuel layer was an increasing function of time, whereas the vertical temperature gradient was a decreasing function of time. Using these experimental results, two dimensionless numbers (Marangoni and Rayleigh) were calculated. The Marangoni number represents the surface tension driven flows in the fuel layer and the Rayleigh number represents the buoyancy driven flows in the fuel layer. The results of this study showed two major convective phases; in the first half of the burning time, the buoyancy driven flows (Rayleigh) were dominant, while Marangoni convection was dominant in the second half of the burning time. The role of these mechanisms in affecting the flow and melting the ice is discussed. (C) 2015 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

The multi-component fuel based eddy dissipation concept(EDC) model, PaSR based soot model and grey gas based FvDOM radiation model etc in ExfireFoam is further extended to simulate a 30 cm × 30 cm ethanol-gasoline square pool fire. The predictions achieved good agreement with the measurements for flame height and the radial temperature profiles at different heights, demonstrating the good potential of the ExfireFoam for predicting the combustion process of pool fires from multi-component fuels.

In order to study fire extinguishment additives which are no pollution and few damage on object protected, establish a fixed water mist fire extinguish experiment platform, explore the temperature variation characteristics and flam shape by adding surfactant on water mist for the pool fire suppression, though the comparison with the flame temperature and extinguish time of gasoline and diesel pool fire, get their fire performance. The results show that adding surfactant to the water mist can greatly improve the water mist fire performance. The process of additives-containing water mist system inhibit the pool fire can be divided into four stages, which are flame early growth stage, preliminary inhibition stage, flame again growth stage and again inhibition stage. The mechanism of containing a surfactant water mist inhibit the pool fire of gasoline and diesel, which can capture and annihilate the free radical of chain reaction, decrease surface tension, not simply cooling effect. The surfactant molecules reduce the energy of surface and interface, which is good for the descent of water mist droplet size, the emulsification of surface gasoline with water and the formation and stabilization of foam, improve the effect on the cooling and isolation of fame and gasoline with water mist.

Liquid free burning pool fires usually could maintain at a steady burning rate under a certain environment. Due to that the burning rate and the heat feedback could reach a steady self-equilibrium state. Based on mathematical analysis of the equilibrium process, a new modeling approach of liquid pool fires was developed and validated by liquid n-heptane pool fire experimental data. The difference between actual heat feedback and sustainable heat feedback clearly reveal the positive feedback process towards the equilibrium mass burning rate, which is given when the difference is zero. From the modeling results, the heat feedback difference increases from zero at zero burning rate in the positive direction, then drops in the negative direction above a certain burning rate. The equilibrium mass burning rates were close to the experimentally measured values for the n-heptane pool fires under two different environmental pressures. Other flame characteristics, such as axial temperature, absorption coefficient and radiative heat flux were generally in good agreement with the available experimental measurements. The mathematical analysis and numerical modeling of the n- heptane pool fires improve the understanding of the interaction mechanism between liquid fuel evaporation and burning behaviors for free burning pool fires.

This paper studies suppression of n-Heptane pool fire with water mist under different longitudinal ventilation velocities in a long and narrow space. Through the suppressing experiments with two different spraying positions and seven ventilation speeds, it is found that the extinguishment of fires has a close relationship with the fuel temperature. Moreover, the fuel temperature, when the fire is extinguished by spraying at different positions, is almost the same within the measurement ranges, and it tends to decrease as the wind speed rises. Mechanism studies show that the coupling effect of dilution and replacement of oxygen and fuel surface cooling is the main mechanism in n-Heptane fire extinguishment of water mist under longitudinal ventilation. In addition, the ability to efficiently cool the fuel plays a determinant role in the n-Heptane pool fire suppression.

The radiant heat of the oil tank liquid combustion flame is the main reason of the fire spread between the oil tanks. In this paper,the thermal radiation characteristics of 10 million cubic meters oil tanks in Zhoushan Crude Oil Reserve Base in the pool fire environment are simulated using Fluent software.The results show that, the pool fire radiation intensity close to the ground was smaller than that in the parts over the tank mouth. The incident radiation intensity is over 200 kw/m 2 within the scope of radius 200m from the flame center. The incident radiation intensity decreased to 100 kw/m 2 in the place of 250m from the flame center radius. The radiant heat intensity was nearly 400-1000 kw/m 2 in the place of 72-152m from the flame center radius. The thermal radiation flux will lead to disastrous fire or explosion of the adjacent oil tanks.

Small-scaled ethanol circular pool fire experiments were conducted in a self-designed low pressure experiment tank to investigate the influence of ambient pressure on the combustion characteristics of the pool fire. The difference of combustion characteristics under different pressure environment were investigated, combustion characteristics of ethanol pool fire, such as fuel's mass burning rate, flame shape, flame centre-line temperature and plume velocity in different ambient pressure conditions were measured. The experiment results show that the mass burning rate decrease when the ambient pressure decrease, and the variation could be well fitted by a power function. Plume centre-line velocity also decrease with the ambient pressure. Generally, flame height would increase with the decreasing of the ambient pressure, until reaching the turning point, and then decrease. Temperature rise of plume centre-line take a power function of height and heat release rate.

In order to study the influence of top opening on oil tank fire characters, small scale oil tank fire experiments with 1, 1/2, 1/4,1/8 opening proportion factors (opening area / tank top area) and different opening eccentricities (opening centroid position to the center of a circle) were carried out in a still environment. The mass loss rate, flame radiation and images, tank-shell temperature were recorded. The results suggested that, the oil tank fire characteristics varied with the opening proportion factor of tank top. Comparisons between the combustion behaviors of partly open oil tank fire and fully open pool fire with the same equivalent diameters showed that, the mass loss rates and flame frequencies of tank fires were almost consistent with pool fire. But flame shape varied with the opening proportion factor and eccentricity, for partly opened oil tank fire, the flame shape is unsymmetrical but L-shaped, and flame height showed difference with pool fire as flame height gets lower when opening proportion factor approaches 0. We compared the flame radiation and tank-shell temperature of open end side and opposite side, found the difference and inferred the significant causes leading to the difference were opening proportion and opening eccentricity.

Intensification of pool fire will occur under the application of water mist. The phenomenon was discovered in previous studies; however, there has been no unified conclusion for the reason of this phenomenon. Studying mechanisms of the intensification has a very important significance in the engineering application of water mist. According to previous studies, the intensification can be divided into two categories, one is the fire intensification in gas flame; the other is the interaction of water mist with the heated fuel, which lead to the increase of mass loss rate. In this paper, Mechanisms of the second type of fire intensification are studied through theoretical analysis and experiment. Research shows that the primary mechanism of fire intensification by water mist is boil over.

The experimental observation of pool fires under dynamic pressure is of significant meaning to understand the fire behaviors under different low pressure environments. Previous studies on low-pressure fires mainly focused on fire behaviors at static low pressure, where the fire tests were usually conducted at several discrete altitudes. An altitude chamber of size 3 m× 2 m× 2 m has been built for this study, where different dynamic pressure descent rates were configured to simulate pool fire behaviors under different low pressures. In each test, the chamber pressure varies from standard atmospheric pressure 101.3 KPa to 34 KPa. The dynamic pressure descent rate is controlled by replenishing air at different rates: 0, 15 and 20 Nm3/h. The study tested two different sizes of pool fires, where the pan heights are all 2 cm and the pan diameters are 6 cm and 10 cm respectively. N-Heptane with industrial purity of 99% was used as the testing fuel. Parameters were measured along the whole burning process, including axial flame temperature, burning rate, radiative heat flux, chamber pressure as well as fire video recording. Some specific phenomena were observed for the fires under dynamic depressurization, e.g. the flame transmitted from turbulent to laminar and its base turned blue as pressure drops, then in the final burning stage a polyhedral flame appeared and swirled in the pan until extinguishment. The measurement data, i.e. mass burning rate, flame temperature and radiative heat flux, were analyzed to reveal the pressure effect in influencing fire behaviors.

The self-extinction behavior of N-heptane pool fire located in a ceiling-vented compartment is investigated experimentally. The compartment is a rectangular chamber of 1.00 m (L) × 1.00 m (W) × 0.75 m (H) with only a ceiling vent with sizes between 0 and 0.490 m × 0.490 m. Three sizes of pans with the diameters of 0.10 m, 0.141 m and 0.20 m were at the center of the compartment with similar initial fuel thickness. The impact of vent area on fuel consumption and flame extinction time is mainly examined as the result, meanwhile the local oxygen concentration near the fire source and the mass loss rate of fuel are also reported. The result shows that the pool fire burning behavior can be classified into four types according to the ceiling vent size. The extinction time t e of first type of fire nearly equals to the closed condition, while it grows sharply with vent size for the second type of fire. When the fire belongs to the third type, t e decreases with vent size due to the growth of fuel burning rate, and for the fourth type, t e keeps constant again with the well-ventilated/condition. Based on species concentration, the ceiling vent size can be normalized into (ρ ∞g 1/2 A v 5/4 )/(m F ”A F ) . An empirical exponential relationship between self-extinction time ratio and the dimensionless ceiling vent size was proposed, and the results well ntegrate all the profiles of self-extinction time of different fire sizes.

The self-extinction time of pool fires in closed compartments was studied. Experiments were conducted in two bench-scale compartments with volumes of 0.75 m 3 and 17.55 m 3 . It was found that the fire self-extinction time was proportional to the compartment volume but inversed to the pool area for n -heptane. The fuel mass loss rate in closed compartments was lower than that in the open space. The fire self-extinction occurred when the local oxygen mole fraction in the flame vicinity descended to a level of 10.7-15.3%. The mean remaining oxygen mole fraction at the self-extinction moment was about 14.1%. Based on the mass conservation of the oxygen, a model for predicting the self-extinction time of pool fires in closed compartments was developed. By defining a concept of the dimensionless fire volume, the dimensionless self-extinction time was proposed. The dimensionless self-extinction time is proportional to the difference between the initial and remaining oxygen mass fraction, fuel properties, such as heat of combustion and stoichiometric ratio etc., but inverses to the dimensionless fire volume and the integrated combustion coefficient. The predicted results showed a good agreement with the experimental results. The model also provides a good prediction for the results of NRL's t sts.