Effect of Room Layout on Natural Gas Explosion in Kitchen

: In order to elucidate the overpressure and fire hazard effects of gas explosion in a congested room, the effects of gas concentration and room layout on a gas explosion in a kitchen were studied by CFD. The results showed that the flow field parameters in a kitchen exhibited an initial increase followed by a decrease as the gas concentration increased. The maximum gas flow rate recorded within the chamber was 390 m/s, while the corresponding maximum flame propagation rate and peak pressure reached 289.86 m/s and 30.95 kPa, respectively. The difference in the flow field induced by the concentration was further enhanced by the presence of congested materials. Additionally, the room layout influenced the gas congestion’s blowout effect due to variations in turbulence intensity and flammable gas volume caused by significant changes in the congestion within the room. Specifically, when the gas concentration was 10%, the order of indoor gas flow rate and flame combustion rate were II > U > L > I, while the turbulent kinetic energy and explosive overpressure followed the order I > II > L > U. The results are of great significance for the disaster assessment and accident prevention of natural gas explosion in civil kitchens.


Introduction
As a clean and environmentally friendly energy source, natural gas has been widely used in industrial production as well as in residential cooking and heating.With the rapid development of our country's urbanization and economy, the demand for natural gas has increased significantly.However, natural gas explosions occur frequently because of aging pipelines and inappropriate use of gas facilities, causing great loss of life and property to the people [1][2][3][4][5].For example, on 29 July 2023, a gas explosion occurred in a street shop at Jinse Yangguang Community, Hongqi Street, Anping County, Hengshui City, Hebei Province, resulting in two deaths and two serious injuries.According to the National Analysis Report on Gas Explosion Accidents, in the first half of 2023 alone, there were 294 gas accidents in China, resulting in 45 fatalities and 113 injuries.Among these, there were a total of 118 natural gas accidents, with the highest number of injuries occurring among residential users, reaching a total of 119 individuals.Therefore, in order to prevent the occurrence of such accidents, it is of great practical significance to study the gas explosion mechanism.
Natural gas explosion accidents usually occur in confined spaces such as kitchens and industrial plants.When the gas explosion occurs, the doors and windows of the kitchen, as well as the lightweight walls and roofs of industrial buildings, evolve into con-strained venting structures, leading to the phenomenon of constrained venting [6,7].Additionally, congested objects in kitchens, such as cabinets and countertops, as well as equipment and facilities in industrial plants, can promote indoor turbulence.The coupling between the venting surface and congested objects mutually influences the indoor explosion flow field and disaster development, increasing the complexity of the explosion venting [8][9][10][11][12].Moreover, this complexity poses challenges for on-site rescue operations and rapid postdisaster assessments of natural gas explosion disasters.
To this end, numerous scholars have conducted studies on the process of constrained release of natural gas.In the research on the mechanism of gas explosion accidents in buildings, various kinds of congestion such as equipment, facilities, furniture, articles, and pipelines are often taken as the main factors to be examined.Park et al. [13] investigated the effects of the shape of three continuous-obstacle cross-sectional shapes, namely, rectangular, triangular, and cylindrical shapes, on unconstrained venting in small-scale vessels with different diameter ratios.They found that rectangular obstacles corresponded to the highest venting pressure, while cylindrical obstacles corresponded to the lowest venting pressure.Lowesmith et al. [14] demonstrated the contribution of indoor equipment and pipelines to shock wave overpressure and flame propagation velocity by means of unconstrained explosion experiments with large-scale gas explosions.Based on Lowesmith's experimental results, Woolley et al. [15] further explored the quantitative distribution of overpressure and flame speed during methane/hydrogen explosion and unconstrained venting processes under the obstruction of multiple obstacles using computational fluid dynamics (CFD) techniques.Valeria et al. [16] and Gubba et al. [17] investigated the effect of characteristic parameters of single/continuous obstacles on indoor gas explosion and unconstrained venting processes using large eddy simulation techniques.Tomlin et al. [18], utilizing a large-scale explosion chamber measuring 9.0 m × 4.5 m × 4.5 m and an array of small-scale embedded pipes, investigated the impact of venting size and blockage rate on indoor natural gas explosion and unconstrained venting processes.This study indicated that despite the limited space in residential buildings, under the action of limited venting area and indoor congestion, explosion overpressure sufficient to damage the building structure can still be generated.Pang et al. [19] studied the effects of opening pressure, opening time, and outlet area on external explosion intensity, explosion location, and explosion duration through the AutoReaGas software (v3.1), summarizing and analyzing the distribution law of indoor hydrogen explosion disasters, which has important scientific value for future research on disaster assessment and prediction of hydrogen and other flammable gases.
Previous studies mainly focused on a large number of studies on internal obstacles, but the above studies still focus on small-scale explosive vessels, even large-scale cavities, because the obstacles are mainly arranged in tank and column structures; the size of the obstacles is also greatly different from actual activities.This is inconsistent with the fact that obstacles exist around the wall and are distributed up and down in actual buildings, and the experimental data obtained from this can hardly be used to guide actual production activities.Therefore, these research results cannot fully reveal the mechanism of gasrestrained detonation in the kitchen.In order to further study the influence of the size and layout structure of obstacles in the room layout on kitchen gas explosions, the indoor environment should be reasonably scaled and constructed according to the actual model size.In conclusion, the existing research results of natural gas explosions have greatly simplified the indoor obstacles, neglecting the coupled effects between the characteristics of congested objects within the room and the venting structure on the explosion process.These findings do not reflect the characteristics of gas explosion disasters in residential kitchens, thereby greatly limiting disaster rescue and accident investigation and the analysis of natural gas explosion accidents in residential buildings.In this paper, the CFD numerical method is used to establish different layouts of civil kitchens and to reveal the law of constrained gas blowout under the synergistic action of multiple constraints.

Numerical Model
AutoReaGas (v3.1) is a three-dimensional computational hydrodynamic analysis software designed to simulate gas explosions and their shock effects.The AutoReaGas (v3.1) numerical simulation results are in good agreement with the results of the famous BFETS test.Its reliability has been verified and is widely used around the world to simulate gas explosions and their after-effects in congested and confined environments [20][21][22].Therefore, this paper uses this software to carry out research on explosion venting in a confined space.
AutoReaGas (v3.1) uses finite volume methods to solve control equations and the conservation equations of mass, momentum, and energy on three-dimensional Descartes grids, as follows: The mass conservation equation is ∂ρ ∂t The momentum conservation equation is The energy conservation equation is The expression of E = C VT + m fu Hc, τ ij is where µt = Cµ ρk 2 /ε and Cµ = 0.09 m 2 /s.The mathematical expression of the conservation equation for the mass fraction of gaseous fuel is as follows: where R fu = C t ρ (s t 2 /Γ fu ); R min ; C t = 40.Turbulence is the key factor of gas explosions.The model is modeled by k-epsilon, which consists of the conservation equation of k-epsilon and the dissipation rate of turbulent kinetic energy.
where C 1 = 1.44 and C 2 = 1.79.The laminar combustion velocity Sb can be expressed as where F s = 0.15.
The Turbulent combustion velocity St can be expressed as In AutoReaGas(v3.1),Navier-Stokes and Euler partial differential equations are numerically solved by the finite volume method, and the pressure-velocity coupling and The above numerical model involves parameters and Greek letter symbols as shown in Table 1.

Verification of Grid Independence
In order to investigate the effect of grid size on the explosion simulation, the grid was divided into two sets of sizes: 0.1 m × 0.1 m × 0.1 m (M 1 ) and 0.05 m × 0.05 m × 0.05 m (M 2 ).The simulation is for rooms measuring 3.4 m × 2.1 m × 2.4 m, in which the roof, floor, and walls are set to solid boundaries and no congestion is set in the model rooms.A blowout surface of 0.8 m × 0.8 m is arranged on one wall with an opening pressure of 20 kPa and an opening time of 0 s.The room is filled with methane/air premixed gases with a gas concentration of 10%.The ignition source is located in the geometric center of the back wall and 0.05 m from the back wall.Table 2 shows the comparison of peak overpressure from numerical calculations for the two grid sizes at different locations.According to the table, the variation in overpressure between the lower and middle grid sizes is basically the same and the maximum relative error is less than 10%.In order to save time and improve computing efficiency, grid M1 was used to carry out this research.

Verification of Computational Domain
Taking into account the influence of the computational domain on numerical simulation, the computational domain was set to a solid wall surface at the bottom and a free-flow boundary at the top and around, and two different computational domain sizes were selected for validation: 20.4 m × 2.1 m × 2.4 m and 20.4 m × 3 m × 4 m.Calculations of maximum peak overpressure were obtained at different points in the chamber, as compared in Table 3.The results show that the maximum relative error in the two computational domains corresponds to 5.3% of the maximum explosive peak overpressure, indicating that both computational domains are suitable for this study.In order to save time and improve the efficiency of calculation, the size of 20.4 m × 2.1 m × 2.4 m is used for calculations in this paper.

Experimental Validation of the Model
In order to study the applicability of the selected numerical model to the problem in this paper, the numerical model is compared with Bauwens' large-scale gas vent experiment.Bauwens [23] carried out the constrained blowout study in an experimental device measuring 4.6 m (long) × 4.6 m (wide) × 3.0 m (high), with 8 medium size congested objects (0.4 m × 0.4 m × 3.0 m) and a volume obstruction rate of 6%.The comparison validation set the same parameter conditions as the experiment, in which the static opening pressure of the explosion venting was set at 0.5 kPa.The ignition source was located in the geometric center of the back wall and is 0.25 m from the back wall.
Figure 1 shows the time curves of the experiment and the numerical simulation of explosion overpressure, while Table 4 compares their peak overpressure and their arrival times.The relative error in peak overpressure and arrival time between the experiment and numerical simulation is less than 3.5%, and the two overpressure time curves show similar trends.Due to the uncertainty in the experimental equipment itself, the accuracy of the sensor, and the experimental process, the error may be caused by a combination of factors, but the calculation error of the numerical method is acceptable.In conclusion, we Figure 1 shows the time curves of the experiment and the numerical simulation of explosion overpressure, while Table 4 compares their peak overpressure and their arrival times.The relative error in peak overpressure and arrival time between the experiment and numerical simulation is less than 3.5%, and the two overpressure time curves show similar trends.Due to the uncertainty in the experimental equipment itself, the accuracy of the sensor, and the experimental process, the error may be caused by a combination of factors, but the calculation error of the numerical method is acceptable.In conclusion, we can solve the transient flow field problem of indoor gas explosion and confined outburst under the action of congestion by the numerical method mentioned above.

Research Methods
According to the Residential Design Specification (GB 50096) [24] typical civil kitchens are divided into four room layouts: type I, type II, type L, and type U. Based on previous research, it is found that the explosion overpressure is not only affected by the explosion surface and the characteristics of indoor congestion, but also may be affected by the layout of the congested objects.Therefore, on the basis of previous studies, it is necessary to further investigate the significant influence law of the congestion layout characteristics in the kitchen on the explosion flow field and overpressure disaster effect, so as to

Research Methods
According to the Residential Design Specification (GB 50096) [24] typical civil kitchens are divided into four room layouts: type I, type II, type L, and type U. Based on previous research, it is found that the explosion overpressure is not only affected by the explosion surface and the characteristics of indoor congestion, but also may be affected by the layout of the congested objects.Therefore, on the basis of previous studies, it is necessary to further investigate the significant influence law of the congestion layout characteristics in the kitchen on the explosion flow field and overpressure disaster effect, so as to determine whether to create indoor explosion overpressure disaster assessment methods for the four types of kitchen layout, so as to achieve a scientific and accurate assessment of indoor overpressure disasters.In this study, small kitchen units of "type I", "type II", "type L", and "type U" are selected as the research objects, as shown in Figure 2. The study found that the larger objects in the kitchen usually include counter cabinets, hanging cabinets, refrigerators, butters, and so on, and generally along the walls of the room throughout the layout, countertops, fridges, and so on, are usually placed along the walls of the floor, while hanging cabinets, butters, and so on, are usually suspended against the wall.Congestion in the kitchen is mainly in the shape of approximately cuboid objects, usually in the range of 0.3 m to 0.7 m in height and width, the maximum length can even be lengthways through the entire kitchen (such as gantry cabinets, ceiling cabinets, etc.); the room volume congestion rate is generally 10% to 35%.These parameters are used as the main basis for the selection of typical congestion characteristics in the numerical modeling.
Congestion in the kitchen is mainly in the shape of approximately cuboid objects, usually in the range of 0.3 m to 0.7 m in height and width, the maximum length can even be lengthways through the entire kitchen (such as gantry cabinets, ceiling cabinets, etc.); the room volume congestion rate is generally 10% to 35%.These parameters are used as the main basis for the selection of typical congestion characteristics in the numerical modeling.In order to study the influence of kitchen layout, and based on the general size characteristics of Chinese residential kitchens, we fixed the room size to 3.4 m (long) × 2.1 m (wide) × 2.4 m (high).The walls, roof, and floor around the kitchen and the congested surfaces were all set to rigid wall surfaces, equating the door closure in Figure 2 to the wall and the window to the blowout surface, set to a height of 1 m below the window in the center of the wall.The fixed blowout size was 0.8 m × 0.8 m, with an opening time of 0 s and an opening pressure of 20 kPa.The computational domain extends five times the length of the room along the direction of the vent to capture the effects of the external explosion and all the external computational domains are set to the free-flow boundary.The time step length is set to equal time step lengths of 1 × 10 −5 s.Although gas concentration also has a significant effect on indoor gas explosion overpressure, there is not enough systematic research on the effect of gas concentration on indoor gas explosion overpressure under different typical kitchen layouts.Therefore, we carry out numerical research in five conditions, with gas volume concentrations of 6%, 8%, 10%, 12%, and 14%, respectively, and use the back wall for ignition.The ignition source is set to be 0.05 m away from In order to study the influence of kitchen layout, and based on the general size characteristics of Chinese residential kitchens, we fixed the room size to 3.4 m (long) × 2.1 m (wide) × 2.4 m (high).The walls, roof, and floor around the kitchen and the congested surfaces were all set to rigid wall surfaces, equating the door closure in Figure 2 to the wall and the window to the blowout surface, set to a height of 1 m below the window in the center of the wall.The fixed blowout size was 0.8 m × 0.8 m, with an opening time of 0 s and an opening pressure of 20 kPa.The computational domain extends five times the length of the room along the direction of the vent to capture the effects of the external explosion and all the external computational domains are set to the free-flow boundary.The time step length is set to equal time step lengths of 1 × 10 −5 s.Although gas concentration also has a significant effect on indoor gas explosion overpressure, there is not enough systematic research on the effect of gas concentration on indoor gas explosion overpressure under different typical kitchen layouts.Therefore, we carry out numerical research in five conditions, with gas volume concentrations of 6%, 8%, 10%, 12%, and 14%, respectively, and use the back wall for ignition.The ignition source is set to be 0.05 m away from the geometric center of the rear wall.The initial ambient pressure and temperature in the computational domain were set at 1.0132 5 × 10 2 kPa and 300 K, respectively.
Based on the influence of kitchen layout on indoor overpressure disaster and the general characteristics of kitchens in Chinese residential buildings, a prediction model of constrained natural gas explosion peak overpressure in a typical kitchen layout was built through systematic research.

Distribution Characteristics of Indoor Flow Field
The effect of gas concentration on flow velocity was analyzed using a typical dualline layout (type II layout).Figure 3 shows the evolution of the gas flow velocity over time under different gas concentrations in the room.As can be seen from the figure, the gas flow velocity at the center of the room increases rapidly and a series of oscillations occur under the influence of congestion and venting surfaces, with a peak velocity of about 30 m/s.The gas flow velocity significantly increases near the explosion venting and exceeds 130 m/s, indicating that the explosion venting is more dominant than the congestion surface.Figure 4 shows the variation in peak flow velocity with distance under different gas concentrations.There is no significant difference in peak flow velocity at various positions away from the explosion venting within the room for different gas concentrations.However, it sharply increases near the explosion venting and reaches the maximum value near 10% concentration.When the gas concentration reaches 10%, the analysis believes that the gas has reached an equivalent concentration, and the combustion fully reacts with the air, resulting in an enhanced chemical reaction.As a result, the peak flow rate of the gas mixture near the exhaust port is significantly enhanced.It can be seen that the explosion venting and the gas concentration have a synergistic effect on the airflow velocity, and the latter has a more significant effect.

Distribution Characteristics of Indoor Flow Field
The effect of gas concentration on flow velocity was analyzed using a typical dualline layout (type II layout).Figure 3 shows the evolution of the gas flow velocity over time under different gas concentrations in the room.As can be seen from the figure, the gas flow velocity at the center of the room increases rapidly and a series of oscillations occur under the influence of congestion and venting surfaces, with a peak velocity of about 30 m/s.The gas flow velocity significantly increases near the explosion venting and exceeds 130 m/s, indicating that the explosion venting is more dominant than the congestion surface.Figure 4 shows the variation in peak flow velocity with distance under different gas concentrations.There is no significant difference in peak flow velocity at various positions away from the explosion venting within the room for different gas concentrations.However, it sharply increases near the explosion venting and reaches the maximum value near 10% concentration.When the gas concentration reaches 10%, the analysis believes that the gas has reached an equivalent concentration, and the combustion fully reacts with the air, resulting in an enhanced chemical reaction.As a result, the peak flow rate of the gas mixture near the exhaust port is significantly enhanced.It can be seen that the explosion venting and the gas concentration have a synergistic effect on the airflow velocity, and the latter has a more significant effect.Figure 5 shows the evolution of turbulent kinetic energy with time under different gas concentrations in the room.As can be seen from the figure, when the gas concentration increases from 6% to 14%, the turbulent kinetic energy tends to increase, and then, de-  Figure 5 shows the evolution of turbulent kinetic energy with time under different gas concentrations in the room.As can be seen from the figure, when the gas concentration increases from 6% to 14%, the turbulent kinetic energy tends to increase, and then, decrease, and reaches its maximum at a 10% concentration.Turbulent kinetic energy increases when the gas concentration is 10%.This is due to the greater reaction of the combustion explosion of a premixed gas cloud near the 10% concentration, which causes a higher pressure in the chamber, which increases the rate of gas release when the blowout surface opens.Figure 6 shows the variation curve of peak turbulent kinetic energy with distance under different gas concentrations in type II-layout rooms.As can be seen from the figure, the peak turbulent kinetic energy is maximized near the explosion surface.At the 10% concentration, the peak turbulent kinetic energy increases significantly in the range of 0.5 m to 1.5 m from the rear wall, so it is known that the concentration has a stronger effect on the turbulent kinetic energy than the opening of the blasting surface.Figure 5 shows the evolution of turbulent kinetic energy with time under different gas concentrations in the room.As can be seen from the figure, when the gas concentration increases from 6% to 14%, the turbulent kinetic energy tends to increase, and then, decrease, and reaches its maximum at a 10% concentration.Turbulent kinetic energy increases when the gas concentration is 10%.This is due to the greater reaction of the combustion explosion of a premixed gas cloud near the 10% concentration, which causes a higher pressure in the chamber, which increases the rate of gas release when the blowout surface opens.Figure 6 shows the variation curve of peak turbulent kinetic energy with distance under different gas concentrations in type II-layout rooms.As can be seen from the figure, the peak turbulent kinetic energy is maximized near the explosion surface.At the 10% concentration, the peak turbulent kinetic energy increases significantly in the range of 0.5 m to 1.5 m from the rear wall, so it is known that the concentration has a stronger effect on the turbulent kinetic energy than the opening of the blasting surface.

Characteristics of Indoor Flame Propagation
Figure 7 shows the variation curve of the peak combustion rate with distance under different gas concentrations in the room.According to the figure, under each working condition, the combustion rate increases from 1.5 m away from the rear wall, that is, the area where the congestion is complicated, to the maximum near the explosion venting.In addition, with the increase in indoor combustible gas concentration, the maximum com-

. Characteristics of Indoor Flame Propagation
Figure 7 shows the variation curve of the peak combustion rate with distance under different gas concentrations in the room.According to the figure, under each working condition, the combustion rate increases from 1.5 m away from the rear wall, that is, the area where the congestion is complicated, to the maximum near the explosion venting.In addition, with the increase in indoor combustible gas concentration, the maximum combustion rate tends to increase first, and then, decrease.The combustion rate is highest at a 10% gas concentration at 2.70 kg/s and lowest at a 14% gas concentration at 0.05 kg/s, with a maximum change of 98.15%.It can be seen that congestion, explosion venting, and concentration all have significant effects on the combustion rate, which together leads to a rapid increase.

Characteristics of Indoor Flame Propagation
Figure 7 shows the variation curve of the peak combustion rate with distance under different gas concentrations in the room.According to the figure, under each working condition, the combustion rate increases from 1.5 m away from the rear wall, that is, the area where the congestion is complicated, to the maximum near the explosion venting.In addition, with the increase in indoor combustible gas concentration, the maximum combustion rate tends to increase first, and then, decrease.The combustion rate is highest at a 10% gas concentration at 2.70 kg/s and lowest at a 14% gas concentration at 0.05 kg/s, with a maximum change of 98.15%.It can be seen that congestion, explosion venting, and concentration all have significant effects on the combustion rate, which together leads to a rapid increase.Figure 8 shows the flame propagation velocity with distance under different gas concentrations in the room.As can be seen from the figure, the flame propagation speed presents an obvious "double peak" structure, the first peak is related to the acceleration of the flame by complex congestion, and the closer the gas concentration is to the equivalent concentration, the more obvious the acceleration effect of congestion on the flame.The second peak is related to the intense turbulence induced by the opening of the explosion venting, which accelerates the flame front to propagate faster.With the increase in the concentration of combustible gas in the room, the maximum flame propagation speed increases first, and then, decreases.The study reveals that the obstacles, gas concentration, and the opening of the explosion venting have significant effects on the flame propagation.Specifically, at a concentration of 10%, the flame propagation velocity reaches its maximum at 289.86 m/s, while at a concentration of 14%, it reaches its minimum at 99.01 m/s, with an increase of 65.84%.
Figure 9 shows the combustion rate cloud of bursting flame under different gas concentrations in the room, reflecting the process of flame propagation.As shown in the figure, the process of flame propagation increases first, and then, decreases as the gas concentration increases, with the flame propagation fastest when the concentration is 10%.At gas concentrations of 6% and 14%, the flame structure is approximately finger-like between 0.2 and 0.31 s, and the combustion rate is relatively low.At gas concentrations of 8% and 10%, the flame propagation state is relatively similar, the flame is hindered and deflected when passing through the refrigerator, and it burns violently after the lampblack machine and expands outward, and finally, merges with the flame front.It can be seen that large volume congestion hinders flame propagation, and small-scale congestion promotes flame development.
concentration of combustible gas in the room, the maximum flame propagation speed increases first, and then, decreases.The study reveals that the obstacles, gas concentration, and the opening of the explosion venting have significant effects on the flame propagation.Specifically, at a concentration of 10%, the flame propagation velocity reaches its maximum at 289.86 m/s, while at a concentration of 14%, it reaches its minimum at 99.01 m/s, with an increase of 65.84%.Figure 9 shows the combustion rate cloud of bursting flame under different gas concentrations in the room, reflecting the process of flame propagation.As shown in the figure, the process of flame propagation increases first, and then, decreases as the gas concentration increases, with the flame propagation fastest when the concentration is 10%.At gas concentrations of 6% and 14%, the flame structure is approximately finger-like between 0.2 and 0.31 s, and the combustion rate is relatively low.At gas concentrations of 8% and 10%, the flame propagation state is relatively similar, the flame is hindered and deflected when passing through the refrigerator, and it burns violently after the lampblack machine and expands outward, and finally, merges with the flame front.It can be seen that large volume congestion hinders flame propagation, and small-scale congestion promotes flame development.

Characteristics of Indoor Overpressure Distribution
Figure 10 shows the variation curve of overpressure over time under different gas concentrations in the room.As shown in the figure, the trend in indoor overpressure is basically the same at all locations.The first peak is caused by the opening of the explosion venting, which is slightly greater than the opening pressure of the blasting surface (20 kPa).The second peak is caused by the maximum fire area in the chamber, and its value increases first, and then, decreases with the increase in gas concentration.The analysis shows that this is because the existence of obstacles increases the external turbulent kinetic energy in space, which leads to an increase in flame propagation speed and flame area, and significantly increases the combustion rate, resulting in the formation of a second pressure peak.And the maximum is achieved at a 10% concentration.It can be seen that the gas concentration has a significant effect on the second peak.

Characteristics of Indoor Overpressure Distribution
Figure 10 shows the variation curve of overpressure over time under different gas concentrations in the room.As shown in the figure, the trend in indoor overpressure is basically the same at all locations.The first peak is caused by the opening of the explosion venting, which is slightly greater than the opening pressure of the blasting surface (20 kPa).The second peak is caused by the maximum fire area in the chamber, and its value increases first, and then, decreases with the increase in gas concentration.The analysis shows that this is because the existence of obstacles increases the external turbulent kinetic energy in space, which leads to an increase in flame propagation speed and flame area, and significantly increases the combustion rate, resulting in the formation of a second pressure peak.And the maximum is achieved at a 10% concentration.It can be seen that the gas concentration has a significant effect on the second peak.Figure 11 shows the variation in maximum overpressure under different gas concentrations in the room.As can be seen from the figure, the maximum overpressure in the chamber tends to increase first, and then, decrease with the increase in gas concentration.The maximum indoor overpressure is 30.95 kPa when the gas concentration is 10%.The minimum indoor overpressure is 21.40 kPa and the maximum variation is 44.63% when the gas concentration is 12%.Figure 11 shows the variation in maximum overpressure under different gas concentrations in the room.As can be seen from the figure, the maximum overpressure in the chamber tends to increase first, and then, decrease with the increase in gas concentration.The maximum indoor overpressure is 30.95 kPa when the gas concentration is 10%.The minimum indoor overpressure is 21.40 kPa and the maximum variation is 44.63% when the gas concentration is 12%.Figure 11 shows the variation in maximum overpressure under different gas concentrations in the room.As can be seen from the figure, the maximum overpressure in the chamber tends to increase first, and then, decrease with the increase in gas concentration.The maximum indoor overpressure is 30.95 kPa when the gas concentration is 10%.The minimum indoor overpressure is 21.40 kPa and the maximum variation is 44.63% when the gas concentration is 12%.

Distribution Characteristics of Indoor Flow Field
Figure 12 shows the spatio-temporal evolution of unburned clouds under different room layouts at a 10% concentration (FUEL.M.FR).represents the mass fraction of a flammable gas).As can be seen from the figure, in the early stage of the explosion, the flame propagates forward rapidly in a finger-like manner.Due to the difference in the layout and scale of the indoor congestion, the shape of the combustion area changes significantly after passing through the refrigerator and the lampblack machine, that is, the large-scale congestion (refrigerator) acts as an obstacle to the flame, while the small-scale congestion (range hood) acts as a promotion effect, resulting in an asymmetric indoor flame.In addition, areas such as small gaps near the congestion are more likely to form small confined spaces and retain combustible gas, which is manifested as continuous burning flames.

Distribution Characteristics of Indoor Flow Field
Figure 12 shows the spatio-temporal evolution of unburned clouds under different room layouts at a 10% concentration (FUEL.M.FR).represents the mass fraction of a flammable gas).As can be seen from the figure, in the early stage of the explosion, the flame propagates forward rapidly in a finger-like manner.Due to the difference in the layout and scale of the indoor congestion, the shape of the combustion area changes significantly after passing through the refrigerator and the lampblack machine, that is, the large-scale congestion (refrigerator) acts as an obstacle to the flame, while the small-scale congestion (range hood) acts as a promotion effect, resulting in an asymmetric indoor flame.In addition, areas such as small gaps near the congestion are more likely to form small confined spaces and retain combustible gas, which is manifested as continuous burning flames.Figure 13 shows the variation in maximum gas flow velocity under different room layouts.Overall, in the type I, type II, type L, and type U layouts the maximum gas flow velocity within the room shows an increasing trend followed by a decrease with an increasing gas concentration.It reaches its maximum value at a gas concentration of 10%, measuring 22.06 m/s, 27.81 m/s, 22.81 m/s, and 23.06 m/s, respectively.The gas flow velocity of the four room layouts is in the order type II > U > L > I, indicating that there are differences in the gas flow velocity in these rooms, which may be related to the aspect ratio of the room.The larger the aspect ratio, the closer the flame development in the room is to spreading to the pipeline, and the more obvious the flame acceleration.Figure 13 shows the variation in maximum gas flow velocity under different room layouts.Overall, in the type I, type II, type L, and type U layouts the maximum gas flow velocity within the room shows an increasing trend followed by a decrease with an increasing gas concentration.It reaches its maximum value at a gas concentration of 10%, measuring 22.06 m/s, 27.81 m/s, 22.81 m/s, and 23.06 m/s, respectively.The gas flow velocity of the four room layouts is in the order type II > U > L > I, indicating that there are differences in the gas flow velocity in these rooms, which may be related to the aspect ratio of the room.The larger the aspect ratio, the closer the flame development in the room is to spreading to the pipeline, and the more obvious the flame acceleration.
Figure 14 shows the variation in maximum turbulent kinetic energy in different room layouts.Overall, the maximum turbulent kinetic energy in the four room configurations, type I, type II, type L, and type U, tends to increase, and then, decrease with increasing gas concentration, and reaches maxima of 8.03 m 2 /s 2 , 5.62 m 2 /s 2 , 7.74 m 2 /s 2 , and 7.19 m 2 / 2 , respectively.The turbulent kinetic energy order of the four room layouts is type I > type L > type U > type II, which is opposite to the gas flow velocity.We believe that this is because the gas concentration at the equivalent concentration will promote the chemical reaction, and as the gas concentration gradually exceeds the equivalent concentration, the chemical reaction will be inhibited, resulting in a gradual decline in turbulent kinetic energy.Figure 14 shows the variation in maximum turbulent kinetic energy in different room layouts.Overall, the maximum turbulent kinetic energy in the four room configurations, type I, type II, type L, and type U, tends to increase, and then, decrease with increasing gas concentration, and reaches maxima of 8.03 m 2 /s 2 , 5.62 m 2 /s 2 , 7.74 m 2 /s 2 , and 7.19 m 2 / 2 , respectively.The turbulent kinetic energy order of the four room layouts is type I > type L > type U > type II, which is opposite to the gas flow velocity.We believe that this is because the gas concentration at the equivalent concentration will promote the chemical reaction, and as the gas concentration gradually exceeds the equivalent concentration, the chemical reaction will be inhibited, resulting in a gradual decline in turbulent kinetic energy.Figure 14 shows the variation in maximum turbulent kinetic energy in different room layouts.Overall, the maximum turbulent kinetic energy in the four room configurations, type I, type II, type L, and type U, tends to increase, and then, decrease with increasing gas concentration, and reaches maxima of 8.03 m 2 /s 2 , 5.62 m 2 /s 2 , 7.74 m 2 /s 2 , and 7.19 m 2 / 2 , respectively.The turbulent kinetic energy order of the four room layouts is type I > type L > type U > type II, which is opposite to the gas flow velocity.We believe that this is because the gas concentration at the equivalent concentration will promote the chemical reaction, and as the gas concentration gradually exceeds the equivalent concentration, the chemical reaction will be inhibited, resulting in a gradual decline in turbulent kinetic energy.

Characteristics of Indoor Flame Propagation
Figure 15 shows the variation in maximum combustion rate under different room layouts.Overall, the maximum combustion rate in each room layout also increases first, and then, decreases with the increase in gas concentration, and reaches maxima of 0.25 kg/s, 0.51 kg/s, 0.26 kg/s, and 0.33 kg/s when the gas concentration is 10%.The order of the maximum peak combustion rate of the four room layouts is type II > type U > type L > type I, which is the same as the gas flow velocity.The results show that there is a positive feedback mechanism between gas flow velocity and combustion rate.
layouts.Overall, the maximum combustion rate in each room layout also increases first, and then, decreases with the increase in gas concentration, and reaches maxima of 0.25 kg/s, 0.51 kg/s, 0.26 kg/s, and 0.33 kg/s when the gas concentration is 10%.The order of the maximum peak combustion rate of the four room layouts is type II > type U > type L > type I, which is the same as the gas flow velocity.The results show that there is a positive feedback mechanism between gas flow velocity and combustion rate.Figure 16 shows the spatial and temporal evolution of indoor combustion rate clouds in different room configurations at a 10% gas concentration, which can reflect the propagation of indoor fire.It can be seen from the diagram, in the early stages of ignition, the flame front of room II has not yet reached the first of the dimensional mutations in the congestion, the flame spread is relatively slow, and the flame structure is finger-like.But in Figure 16a,c,d the flame passes through a large congestion (refrigerator) early in its development.Then, within 0.25 s to 0.29 s, the flame passes through a small size congestion (range hood).In the four room layouts, due to the obstruction and turbulence induced by the congestion, the circular flame front folds and deforms, the flame front dents, and the flame propagation speed increases gradually.At the same time, the flow field distribution in the type II and U-shaped rooms is more complex because of the double-row congestion layout, and the flame front deforms under the action of the complex flow field, and the air is sucked into the flame under the Helmholtz effect, forming continuous turbulent combustion.Therefore, the flame in the room with a double-row congestion layout develops more rapidly.Figure 16 shows the spatial and temporal evolution of indoor combustion rate clouds in different room configurations at a 10% gas concentration, which can reflect the propagation of indoor fire.It can be seen from the diagram, in the early stages of ignition, the flame front of room II has not yet reached the first of the dimensional mutations in the congestion, the flame spread is relatively slow, and the flame structure is finger-like.But in Figure 16a,c,d the flame passes through a large congestion (refrigerator) early in its development.Then, within 0.25 s to 0.29 s, the flame passes through a small size congestion (range hood).In the four room layouts, due to the obstruction and turbulence induced by the congestion, the circular flame front folds and deforms, the flame front dents, and the flame propagation speed increases gradually.At the same time, the flow field distribution in the type II and U-shaped rooms is more complex because of the double-row congestion layout, and the flame front deforms under the action of the complex flow field, and the air is sucked into the flame under the Helmholtz effect, forming continuous turbulent combustion.Therefore, the flame in the room with a double-row congestion layout develops more rapidly.Figure 17 shows the variation in maximum indoor overpressure under different room layouts.Overall, the maximum indoor overpressure in each room layout increases, and then, decrease with gas concentration, and the maximum values are obtained at gas con-

Characteristics of Indoor Overpressure Distribution
Figure 17 shows the variation in maximum indoor overpressure under different room layouts.Overall, the maximum indoor overpressure in each room layout increases, and then, decrease with gas concentration, and the maximum values are obtained at gas concentrations of 10% with 29.03 kPa, 28.06 kPa, 27.84 kPa, and 24.11 kPa, respectively.The maximum peak overpressure of the four room layouts is type I > type II > type L > type U.This shows that the simpler the room type, the higher the pressure produced by the gas explosion; there is a discrepancy between this and the general understanding of explosions, that is, the more obstacles, the greater the explosion overpressure.This may be related to a number of influencing factors of explosion overpressure in confined spaces.When there are fewer obstacles in a room of the same size, the volume of combustible clouds in the space increases, which will induce the release of more energy in the explosion process and form a better overpressure.Therefore, when choosing room design or conducting a gas leak risk assessment and emergency rescue, the impact of layout on the aftermath of an explosion needs to be considered.

Conclusions
This study scrutinizes four conventional civil kitchen layouts, using gas concentration as a factor to assess the impact of different room configurations on natural gas congestion and explosion ventilation.The main research conclusions are as follows: (1) The distribution of flow fields in a standard kitchen becomes intricate under varying concentrations of gas clouds.Near the venting surface, the highest levels of gas flow speed, turbulent kinetic energy, and combustion rate are observed, whereas the peak combustion rate and overpressure time graphs display a two-peaked pattern, peaking around room center obstructions.Within the quartet of room configurations, the indoor flowfield's parameters escalate as the gas concentration rises, diminish, and reach their peak at a 10% gas concentration.The limited release of natural gas in these confined areas is intimately linked to the gas concentration, and the dimensional nature of overcrowded kitchen items amplifies the disparity in explosion flow fields, particularly the fluctuation in turbulence intensity due to these congested items with considerable size differences, like refrigerators and range hoods.
(2) Varied arrangements of overcrowded items in the kitchen lead to distinct explosion flow patterns and markedly alter the risk of gas explosions indoors.In the scenario of four distinct kitchen designs, the indoor flow-field's parameters escalate as the gas concentration rises, then diminish, attaining their peak at a 10% gas concentration.At a 10% gas concentration, the order of gas flow speed and flame burning rate in the room is as follows: type II, type U, type L, type I, whereas the rankings for turbulent kinetic energy and explosion overpressure are as follows: type I, type II, type L, type U.

Conclusions
This study scrutinizes four conventional civil kitchen layouts, using gas concentration as a factor to assess the impact of different room configurations on natural gas congestion and explosion ventilation.The main research conclusions are as follows: (1) The distribution of flow fields in a standard kitchen becomes intricate under varying concentrations of gas clouds.Near the venting surface, the highest levels of gas flow speed, turbulent kinetic energy, and combustion rate are observed, whereas the peak combustion rate and overpressure time graphs display a two-peaked pattern, peaking around room center obstructions.Within the quartet of room configurations, the indoor flow-field's parameters escalate as the gas concentration rises, diminish, and reach their peak at a 10% gas concentration.The limited release of natural gas in these confined areas is intimately linked to the gas concentration, and the dimensional nature of overcrowded kitchen items amplifies the disparity in explosion flow fields, particularly the fluctuation in turbulence intensity due to these congested items with considerable size differences, like refrigerators and range hoods.
(2) Varied arrangements of overcrowded items in the kitchen lead to distinct explosion flow patterns and markedly alter the risk of gas explosions indoors.In the scenario of four distinct kitchen designs, the indoor flow-field's parameters escalate as the gas

Fire
transient flow fields are solved by an improved simple algorithm satisfying the Courant-Freidrich-Lewy stability criteria:

Fire
transient flow field problem of indoor gas explosion and confined outburst under the action of congestion by the numerical method mentioned above.

Figure 1 .
Figure 1.Comparison between experimental and numerical simulation explosion overpressure time curve (rear wall ignition and vent area is 5.43 m 2 ).

Figure 1 .
Figure 1.Comparison between experimental and numerical simulation explosion overpressure time curve (rear wall ignition and vent area is 5.43 m 2 ).

Figure 2 .
Figure 2. Four types of kitchen layouts.

Figure 2 .
Figure 2. Four types of kitchen layouts.

Figure 3 .
Figure 3. Variation curve of gas flow velocity with time under different gas concentrations.Figure 3. Variation curve of gas flow velocity with time under different gas concentrations.

Figure 3 . 19 Figure 4 .
Figure 3. Variation curve of gas flow velocity with time under different gas concentrations.Figure 3. Variation curve of gas flow velocity with time under different gas concentrations.Fire 2024, 7, x FOR PEER REVIEW 9 of 19

Figure 4 .
Figure 4. Variation curve of peak gas flow velocity with distance under different gas concentrations.

Figure 4 .
Figure 4. Variation curve of peak gas flow velocity with distance under different gas concentrations.

Figure 5 . 19 Figure 5 .
Figure 5. Variation curve of combustion rate with time under different gas concentrations.

Figure 6 .
Figure 6.Variation curve of peak turbulent kinetic energy with distance under different gas concentrations.

Figure 6 .
Figure 6.Variation curve of peak turbulent kinetic energy with distance under different gas concentrations.

Figure 6 .
Figure 6.Variation curve of peak turbulent kinetic energy with distance under different gas concentrations.

Figure 7 .
Figure 7. Variation curve of peak combustion rate with distance under different gas concentrations.

Figure 7 .
Figure 7. Variation curve of peak combustion rate with distance under different gas concentrations.

Figure 8 .
Figure 8. Variation curve of flame speed with distance under different gas concentrations.

Figure 8 .
Figure 8. Variation curve of flame speed with distance under different gas concentrations.

Fire 2024, 7 , 19 Figure 9 .
Figure 9. Cloud map of indoor combustion rate distribution under different gas concentrations.

Figure 9 .
Figure 9. Cloud map of indoor combustion rate distribution under different gas concentrations.

FireFigure 10 .
Figure 10.Variation curve of overpressure with time under different gas concentrations.

Figure 11 .
Figure 11.Variation curve of maximum overpressure under different gas concentrations.

Figure 10 .
Figure 10.Variation curve of overpressure with time under different gas concentrations.

Fire 2024, 7 ,Figure 10 .
Figure 10.Variation curve of overpressure with time under different gas concentrations.

Figure 11 .
Figure 11.Variation curve of maximum overpressure under different gas concentrations.Figure 11.Variation curve of maximum overpressure under different gas concentrations.

Figure 11 .
Figure 11.Variation curve of maximum overpressure under different gas concentrations.Figure 11.Variation curve of maximum overpressure under different gas concentrations.

Figure 12 .
Figure 12.Time evolution of unburned gas cloud under different room layouts.

Figure 12 .
Figure 12.Time evolution of unburned gas cloud under different room layouts.

Figure 13 .
Figure 13.Variation curve of maximum gas flow velocity under different room layouts.

Figure 14 .
Figure 14.Variation curve of maximum turbulent kinetic energy under different room layouts.

Figure 13 .
Figure 13.Variation curve of maximum gas flow velocity under different room layouts.

Figure 13 .
Figure 13.Variation curve of maximum gas flow velocity under different room layouts.

Figure 14 .
Figure 14.Variation curve of maximum turbulent kinetic energy under different room layouts.Figure 14.Variation curve of maximum turbulent kinetic energy under different room layouts.

Figure 14 .
Figure 14.Variation curve of maximum turbulent kinetic energy under different room layouts.Figure 14.Variation curve of maximum turbulent kinetic energy under different room layouts.

Figure 15 .
Figure 15.Variation curve of maximum combustion rate under different room layouts.

Figure 15 .
Figure 15.Variation curve of maximum combustion rate under different room layouts.

Figure 16 .
Figure 16.Time evolution of combustion rate cloud under different room layouts.

Figure 17 .
Figure 17.Variation curve of maximum overpressure under different room layouts.

Table 1 .
Parameters and Greek letter symbols.

Table 2 .
Explosion peak overpressure for the two different grids.

Table 3 .
Comparison of peak overpressure under different calculation domain sizes.

Table 4 .
Comparison of the numerical and experimental results.

Table 4 .
Comparison of the numerical and experimental results.