Although hydrogen is a clean and renewable fuel, there is still a need to understand and evaluate the potential risks posed in the event of an accidental explosion. This paper presents large eddy simulation (LES) numerical analysis for lean hydrogen premixed flames propagating inside a small laboratory combustion chamber with built in solid obstructions. The small-scale chamber is 0.625 litres in volume with three removable turbulence generating baffles and a square solid obstacle. A lean equivalence ratio of 0.7 is selected in this study. The LES model is utilised to investigate the influence of obstruction configuration and area blockage ratio on the flame characteristics and the generated combustion over-pressure. The LES turbulence technique is used with an in-house computational fluid dynamics (CFD) model for compressible flows. The numerical simulations are carried out with various arrangements of the baffle plates and a solid obstacle to examine the effects of the area blockage ratio and generated turbulence on the flame structure and generated over-pressure. Two different area blockage ratios of 0.24 and 0.5 are studied. Four configurations with different baffle arrangements are studied to examine the resulting turbulence effects on the generated over-pressure, flame position-time traces and flame transient speed following ignition. Direct comparisons are made between the different baffle/flow configurations to identify the various effects of an increased area blockage ratio. Numerical results showing the flame structure at various time windows after ignition are presented and compared with published experimental images. High speed laser induced fluorescence (LIF-OH) images of the reaction zones obtained from the experiments at a rate of 5 kHz provide the flame position data and convey the impact of the turbulence generated by the baffles and solid obstacle on the propagating flame structure [1]. The pressure is recorded at a rate of 25 kHz using a piezo-electric pressure transducer in the base plate of the chamber [2]. The rise in over-pressure as a result of increased turbulence due to additional baffles and an increased area blockage ratio is found to be consistent with experimental data. This is also found to be consistent for the flame position-time and speed-time traces across all configurations. Main points of interest such as the peak over-pressure, maximum rate of pressure rise and the flame propagation trends are also observed along with variations in flame speed as the flame interacts with the baffles and obstacles. Validation of the numerical results against available published experimental data conveys good agreement confirming the ability of the numerical model to predict numerical results for an increased area blockage ratio. Further numerical simulations are also carried out for flame/flow parameters where experimental data is unavailable due to physical limitations. Satisfactory agreement between numerical results and experimental data endorses further predictions for computational models in studying vented hydrogen explosions where there is an increased risk or limited experimental data.

In order to study the optimization modes of smoke control in the case of fire in metro transfer channel, a full-scale numerical station model with a single transfer channel is established. In case of fire, the influencing of ceiling beam, vent size on smoke control effect are analyzed. The ceiling temperature profile, CO concentration, visibility and smoke layer height under different conditions were studied to evaluate the smoke control effect of transfer channel and station hall. Based on this, the optimization suggestions of ventilation and smoke exhaust system design and smoke control modes for transfer channel are put forward. The results show that, the ceiling beams can effectively delay the speed of smoke diffusion and gain time for personnel escape and emergency rescue. The vent size has a great influence on the ventilation and smoke exhaust effect under the condition of meeting the designed ventilation capacity. Among the three common vent sizes listed in this research, the smaller the size is, the better the smoke control effect is. The length and width of vent with best smoke control effect are 1.2 m and 0.8m respectively.

The estimation of heat release rate (HRR) in a building fire is a meaningful yet challenging task to improve first emergency response. Inspired by the capability of deep learning method that mines patterns from raw data, the inverse model based on Gated Recurrent Unit (GRU) is presented for the inversion of HRR. First, a series of fire scenarios is simulated to form the dataset by forward fire modeling. Second, GRU is applied to learn representations from windows of time-series sensor data. Output layer is used to map the learned representations to targets. Third, the HRR is estimated by the trained GRU network with observed data from the fire. Finally, the accuracy and efficiency of the inverse model are evaluated in a multi-compartment configuration. Preliminary results indicate that GRU network can be applied to the inversion of fire HRR with higher accuracy and efficiency.

Firebrands are an important mechanism of fire spread and one of the primary ways in which wildland fires ignite structures. Inverse heat transfer using thin steel plates has been shown to be an effective method for measuring heat transfer distributions from firebrands. To fully understand the dynamic process of heat transfer from firebrands to surfaces; however, it is necessary to view the underside of the firebrands, which is not possible through a steel plate. This work develops a method of inverse heat transfer using a visually transparent quartz plate and a long-wave (7.5–14.0 μm) infrared camera to facilitate visual access to the firebrands from all angles. The heat flux measurements using a quartz plate were compared with heat flux measurements using a steel plate and finite element heat transfer simulations for radiation-dominant tests using heater panels. Additionally, heat transfer measurements using cuboidal oak firebrands were conducted using both the quartz and steel plates. A corrective factor was developed based on the ratio of the effective emissivity of the quartz and stainless-steel plates at typical firebrand temperatures. The measured heat fluxes were within 1–6% after correcting for radiant energy transmitted through the quartz which was absorbed by the stainless-steel plate.

Sizing aerosolized nanoparticles through time-resolved laser-induced incandescence (TiRe-LII) requires an accurate model of the heat conduction from the laser-energized particle to the surrounding gas. Under transition regime conditions this is often done using Fuchs’ boundary-sphere method, which requires the analyst to specify the thickness of a collisionless layer surrounding the particle, representing the Knudsen layer. Traditionally the boundary layer thickness is set to the mean free path of the gas at the boundary temperature, but recently some TiRe-LII practitioners have adopted a more complex treatment that accounts for particle curvature and directional distribution of gas molecules. This paper presents a critical reassessment of this approach; while this modification is more representative of the true Knudsen layer thickness, it does not improve the accuracy of heat conduction rates estimated using Fuchs’ boundary sphere methods under conditions prevailing in most TiRe-LII experiments.

The evaporation and combustion of nanometric aluminum particles with an oxidizer MoO 3 is analyzed. The analysis was performed to correlate individual Al particle gasification rates to macroscopic flame propagation rates observed in flame tube experiments. Examination of various characteristic times relevant to propagation of a deflagration reveals that particles below about 1.7 nm in diameter evaporate before appreciable chemical reactions occur. Experimental studies use Al particles greater than 1.7 nm in diameter such that a diffusion flame model was developed to better understand the combustion dynamics of multiphase Al particles. The results showed that it is unlikely that droplets will fully evaporate before reacting in the gas phase. A droplet evaporation and combustion model was further applied to quantify single droplet reaction velocities in comparison to the bulk flame propagation measurements observed in the literature. The diffusion flame model predicted orders of magnitude slower propagation rates than experimentally observed. These results imply that another reaction mechanism is responsible for promoting reaction propagation or modes other than diffusion play a more dominant role in flame propagation.

The aim of this paper is to present a simple model, based on a time and space integral method, for prediction of preheating and conversion time of a charring solid particle exposed to a non-oxidative hot environment. The main assumptions are 1) thermo-physical properties remain constant throughout the process; 2) temperature profile within the particle is assumed to obey a quadratic function with respect to the space coordinate; 3) pyrolysis initiates when the surface temperature reaches a characteristic pyrolysis temperature; 4) decomposition of virgin material occurs at an infinitesimal thin layer dividing the particle into char and virgin material regions; 5) the volume of the particle remains unaltered; 6) volatiles escape through the pores immediately after formation. Employing assumption (2) allows one to convert the energy conservation equation of the particle, which is basically described in the form of a partial differential equation (PDE), into an ordinary differential equation (ODE) by performing space integration. Next, by applying approximate time integration the ODE is transformed into an algebraic equation. Applying this approach to the preheating and pyrolysis stages of a thermally thin charring solid particle leads to a set of algebraic equations which provides reactor designers with a convenient means for computation of the heating up time, mass loss history and total conversion of particle. The accuracy of the simple model is assessed by comparing its prediction with that of a one-dimensional detailed pyrolysis model. Overall, good agreement is achieved indicating that this new model can be used for engineering and design purposes.

A thermodynamic model and parametric analysis of a natural gas fired power plant with carbon dioxide (CO 2 ) capture using multi-stage chemical looping combustion (CLC) are presented. CLC is an innovative concept and an attractive option to capture CO 2 with a significantly lower energy penalty than other carbon-capture technologies. The principal idea behind CLC is to split the combustion process into two separate steps (redox reactions) carried out in two separate reactors: an oxidation reaction and a reduction reaction, by introducing suitable metal oxide which acts as an oxygen-carrier that circulates between the two reactors. In this study, an Aspen Plus model was developed by employing the conservation of mass and energy for all the components of the CLC system. In the analysis, equilibrium based thermodynamic reactions with no oxygen-carrier deactivation were considered. The model was employed to investigate the effect of various key operating parameters such as air, fuel and oxygen carrier (OC) mass flow rates, operating pressure, and waste heat recovery on the performance of a natural gas fired power plant with multi-stage CLC. Results of these parameters on the plant efficiency are presented. The analysis shows efficiency gain of more than 6% over that of conventional power plant with CO 2 capture technologies when CLC is integrated with the power plant.

A cable box is a metal enclosure that provides connections to the main electric power machineries such as switchgear or transformer. Cable terminals housed in cable boxes have a high fire risk because very few cable boxes can withstand the pressure generated by a low impedance arc. And the cable boxes usually suffer from significant damage causing breakage of bushings between the cable box and the electric equipment. In case of transformer, oil is then spilled from the damaged bushing and the released oil generates the fire ignition. In this study numerical prediction for the pressure rise and propagation during the arc inside the cable box has been performed considering the effects of the shock waves, the deformation of the tank and the fluid-structure interactions between the pressure wave and the cable box. Based on the Euler-Lagrange coupling method, flow field and structure analysis has been conducted to predict pressure rise value of the cable box and the stresses of the wall. Two arcing positions such as line to line and line to ground has been chosen in order to consider how the initial arc point affects the pressure wave propagation and the shapes of pressure wave pattern due to the interaction with near wall.

Ablative materials are employed as thermal insulators in hypersonic space vehicles and for fire proofing in commercial applications. However, the chemical reactions involved in the transformation of ablative materials to char during pyrolysis are not well understood. Reactive molecular dynamics simulations were performed to study the initial stage of the pyrolysis of crosslinked aromatic hydrocarbon polymers. The products formed were characterized and acetylene was found to be the primary product. The value of the activation energy for acetylene formation was estimated. The acetylene formation mechanisms were analyzed at different degrees of crosslinking and the graphitic precursor formation mechanism was studied.

Secondary atomization is one of the most attractive and misunderstood effects in the combustion of microemulsified fuel blends. The occurrence of secondary atomization has been studied to determine its effects on improved combustion efficiency especially when low vapor pressure fuels are used. Several methods to detect microexplosion as alternative to secondary atomization have been considered including acoustic signal processing. As part of the physical characterization of an emulsified vegetable oil-methanol blend, microexplosion behavior of fuel blend droplets has been observed to take place under certain environmental conditions. Droplets microexplode as methanol surrounded by vegetable oil molecules flashes or microexplodes under intense temperature and intense droplet pressure. The droplets of emulsified methanol-in-oil break up forming tiny droplets with greater surface-to-volume ratio in the process. To understand the effects of emulsification on microexplosion, characterization of secondary atomization has been performed using a temperature probe, a high-speed camera and an acoustic sound signal processor. Experiments have been conducted at temperatures similar to those encountered in liquid fuel boilers. The acoustic signal data were analyzed using Fast Fourier Transform (FFT) to define and understand the overall microexplosion process. Also, the effect of temperature, droplet sizes and the percentage of methanol in the vegetable oil blend have been studied to understand what leads to a higher probability of microexplosion occurrence. A correlation between the analyzed acoustic signal data and high speed images were used to differentiate between the different microexplosion events. The results of the study can be useful in predicting the occurrence of microxplosion in liquid fuel boiler which should result in more complete combustion processes, reducing contaminant levels significantly.

The concern for the reduction of the environmental impact caused by greenhouse emissions (CO 2 ) from fossil fuel combustion processes is growing around the world. This has increased research on new energy technologies to produce clean fuels. One of them is the use of biomass as feedstock in gasification processes. The rice agriculture industry around the world produces a great amount of rice husk wastes (RHW) which show the potential for water, soil, and air pollution (including global warming by way of potent greenhouse emissions such as CH 4 ) since waste handling system and structures for storage and treatment frequently are not appropriate. However, the concentration of the rice husk in industrial units makes this low Btu feedstock a viable source for locally based thermal gasification. The current paper presents results on both HRW adiabatic gasification modeling using air-steam blends for partial oxidation and pyrolysis kinetic model to determine, by thermogravimetric analysis (TGA), the RHW activation energy (E). The Chemical Equilibrium with Applications program (CEA), developed by NASA, was used to estimate the effect of both the equivalence ratio (ER) and the steam to fuel ratio (S:F) on adiabatic temperature, gas quality (gas composition and energy density), and energy recovery of an unlimited number of species (∼150). The thermogravimetric analysis (TGA) was carried out using N 2 as carrier gas and under different heating rates (β: 10, 20, 40, and 50 °C/min). Furthermore, the activation energy (E) was estimated based in the results from TGA and using the isoconversional method (i.e., free-model). In general, for the range of parameters studied (0.2 < S:F < 0.8 and 1.5 < Φ < 6), the results from equilibrium adiabatic modeling (CEA) showed that increased ER and (S:F) ratios increase the production of H 2 and CO 2 but decrease the production of CO. Equilibrium temperature decreases with increased ER until ER = 3.0 whereas at ER > 3.0, the effect of ER on equilibrium temperature is negligible. Also, the activation energy average value, estimated from the kinetics model, results to be 233 kJ/kmol.

A quasi-steady-state ablation model is used to investigate the behavior of thermochemically ablating systems in equilibrium and nonequilibrium surface thermochemistry regimes. The model is simplified to allow extraction of relevant nondimensional parameters and comparison with existing experimental data on solid carbon combustion. Good agreement is found between model predictions and experimental data, and the data and model are collapsed in terms of the B number and surface Damkohler number. A new formulation for the surface Damkohler number is proposed, and a relationship between the B number and this Damkohler number is derived for the surface equilibrium and nonequilibrium regimes. The Damkohler formulation is applied to the reentry scenario, and the behavior of the B number in this context is explored. Nondimensional parameters governing behavior in the nonequilibrium regime are determined for graphite oxidation, and the results are extrapolated to more complex surface thermochemistry conditions.

Material processing furnaces are the key component of the manufacturing industries. The burners used in these furnaces require precise control over the flame structure such as flame shape, height, and width. This study mainly focused on the simulation of the flame structure with Computational Fluid Dynamics (CFD) approach. ANSYS Fluent 13.0 was used to predict the flame characteristics in an enclosed cylinder. Non-premixed combustion model was applied to this combustion phenomenon. To control the flame structure, a micro jet at the centre of the burner is introduced. The effect on flame parameters with varying flow rates of micro jet, fuel jet and co-flow jet is examined. This study confirms the experimental study by Sinha et al. [1], which concluded that an air micro jet at the center of a non-premixed flame can control the flame height and luminosity. Moreover, this paper visualizes the thermo chemistry and transport phenomenon of non-premixed combustion process. Emissions from the combustion are monitored for different boundary conditions. This study shows that innovative strategies can be developed for the precise control over the different types of flames with the help of numerical modeling.

Theoretical and experimental investigations of the working processes in the reverse-vortex plasma assisted combustion system have been conducted. The system operates as a multi-mode, multi-purpose reactor in a wide range of plasma feedstock gases and turn down ratios, convenient and simultaneous feeding of several reagents into the discharge zone. The selected concept of the portable plasma assisted system can provide higher performance, wider turn down ratios, more efficient propellants utilization, demonstrate potential fuel flexibility, and satisfy major gravimetric and volumetric density requirements. The results and recommendations obtained can be used for the modeling of reverse-vortex plasma assisted combustion system operation modes, geometry optimization, prospective propulsion and power generation units design and engineering.

Two tiny-oil ignition combustion burners with oxygen-enriched were studied in this paper. One is adding oxygen directly into primary air, and the other is adding oxygen by extra adding-oxygen tube. The bituminous coal, the lignite, and the meager coal, are simulated in these burners. By comparing the two different combustion characteristics, the simulation results show that, for bituminous coal, the high temperature zone congregates in the first directly added oxygen burner, while, for the meager coal, it is difficult to burn in the first directly added oxygen burner. As to the lignite, because of its characteristics of high volatility and low heating value, the first directly added oxygen burner provide oxygen timely for its high volatile, its flame combust more quickly and directly.

A detailed mathematical model is developed for simulation of heat and mass transfer processes during the pyrolysis and combustion of a single biomass particle. The kinetic scheme of Shafizadeh and Chin is employed to describe the pyrolysis process. The light gases formed during the biomass pyrolysis is assumed to consist of methane, carbon dioxide, carbon monoxide, hydrogen and water vapor with given mass fractions relevant to those found in the experiments of high heating conditions. The combustion model takes into account the reactions of oxygen with methane, hydrogen, carbon monoxide, tar and char as well as gasification of char with water vapor and carbon dioxide. Appropriate correlations taken from past studies are used for computation of the rate of these reactions. The model allows calculation of time and space evolution of various parameters including biomass and char densities, gaseous species and temperature. Different experimental data reported in the literature are employed to validate the pyrolysis and combustion models. The reasonable agreement obtained between the predictions and measured data reveals that the presented model is capable of successfully capturing various experiments of wood particle undergoing a pyrolysis or combustion process. In particular, the role of gas phase reactions within and adjacent to particle on the combustion process is examined. The results indicate that for the case of small particles in the order of millimeter size and less, one may neglect any effects of gas phase reactions. However, for larger particles, a combustion model may need to include hydrogen oxidation and even carbon monoxide combustion reactions.

Computational Fluid Dynamics (CFD) is a popular design tool in many projects for ensuring fire safety through performance-based design. However, there are always challenges on the quality and uncertainties of the CFD simulated results. Two points raised are on the grid size and free boundary conditions. A simple corridor fire with a small design fire is taken as an example to address these two points in this paper. The CFD model Fire Dynamics Simulator (FDS) version 5 was taken as the simulation tool Two-dimensional and three-dimensional simulations are compared. The geometry is proposed to outside for better description on minimizing opening boundary.

The M-1 radiation model is a thermal radiation transport model that is derived from a maximum entropy approximation to the radiative transport equation. It involves the solution of four hyperbolic equations for conservation of radiative energy. The M-1 model has similarities to the classical diffusion approximations (like P-1), but is able to better predict directed flux. Consequently, shadowing and long-range transport can be well resolved for a fraction of the cost of methods with exponentially increasing accuracy costs like the method of discrete ordinates and Monte Carlo ray-tracing. The M-1 method is mostly used historically in astronomical radiation transport, but has recently been shown to work for combustion applications of smaller scale. Past work has shown it to give good comparisons to fire problems with length scales of interest. Because of the potential for the model to predict radiation transport more cost-effectively, it is being examined for implementation as an option in our fire codes. We present the theory behind the model. The Eddington factor is used to partition directed and diffuse radiation. It is normally modeled since it is derived from a transcendental functional relationship. We analyze Eddington factor models presented in previous work, and present a new model that we show to be superior in most ways to all the previously presented models. Some 1-dimensional calculations are also shown that illustrate the potential accuracy and challenges with implementing the M-1 model. Such challenges include the specification of boundary conditions and the development of robust solver methods.

Computational Fluid Dynamics (CFD) has been widely used to evaluate the effects of fire on people and buildings by predicting temperature, radiation heat flux, smoke visibility and combustion products. In this paper, CFD was used to study plume temperature, radiation heat flux, visibility and carbon monoxide concentration in a building fire. The CFD software Fire Dynamics Simulator (FDS) developed by the National Institute of Standards and Technology of USA is generally adopted to model the various fire scenarios in buildings. For FDS modeling, the type of fuel is considered an important input parameter which has a significant effect on the simulation results of different types of combustibles, hence the fire hazard evaluation. In the combustion model of FDS, the fuel and oxygen are burnt to give fire products such as carbon dioxide, water, carbon monoxide and soot. Fuels including propane, wood, polyurethane and polyvinyl-chloride were used in the simulation. By using the same fire size and fuel bed for different combustibles under study, a peak fire size of 3 MW was adopted. Simulations were then carried out with different fuel and respective reaction parameters of these combustibles. Simulation results, the output quantities of plume temperature, visibility, carbon monoxide concentration and radiation heat flux for different scenarios obtained were analyzed and evaluated. It was observed that there were reasonable differences in the results from the fuel packages of different combustibles. These differences were evaluated and their significance was discussed.