Experimental Study on the Inerting Effect of Premixed Inert Gas of CO₂ and N₂ in Goaf

Abstract

As the conventional inert gas, it is used for the prevention and control of coal’s spontaneous combustion, mainly N₂ and CO₂. However, there is limited research focusing on the inerting effect of composite inert gas. This paper studied the impact of using a premixed inert gas (N₂ accounted for 50%, 60%, 70%, and 80%) instead of CO₂ on the inerting effect of nearly horizontal and gently inclined goaf by building a physical similarity simulation experiment platform. The experimental results showed that the inerting effect of premixed inert gas was better than that of CO₂. For instance, in a nearly horizontal goaf, the inerting effect of the premixed inert gas was optimal when the N₂ accounted for 70%. The average O₂ concentration in the monitored area decreased from 9.7% with CO₂ to 6.4%. In addition, in the gently inclined goaf, the premixed inert gas exhibited an accumulation state similar to CO₂, primarily occurring in the lower part region adjacent to the working face. Furthermore, the accumulation state of premixed inert gas was inversely proportional to its inerting effect. This study has important reference significance for applying inert gas fire prevention and extinguishing technology in mines.

1. Introduction

Coal is China’s most crucial energy source, accounting for over 50% of the total energy consumption [1]. Coal pillars and specific coal resources are left in the goaf during coal mining. The residual coal exposed to the air is prone to oxidation, generating heat. Accumulation of heat leads to coal spontaneous combustion, ultimately resulting in fires within the goaf [2,3,4]. According to statistics, over 56% of the coal mines in China have recorded fire incidents, of which 90–94% are caused by coal spontaneous combustion [5]. Therefore, the coal spontaneous combustion in goaf is a pivotal challenge constraining safe mining production [6,7,8], and the prevention and control of coal spontaneous combustion is a task of great practical significance.

Among existing technologies, the injection of inert gas (N₂ or CO₂) has emerged as one of the most crucial measures to prevent coal spontaneous combustion, owing to its wide coverage, effective extinguishing capabilities, and being non-polluting towards the working face [9,10]. Among them, N₂, characterized by its low cost and ready availability, has been widely studied by scholars. Zhou et al. and Shi et al. proposed a new liquid-nitrogen method to control mine fires. Subsequently, a new system called a ground-fixed liquid-nitrogen station had been developed [11,12]. Cao et al. identified the optimal nitrogen-injection position and flow rate by analyzing field data on the range of three spontaneous combustion zones before and after nitrogen injection [13]. Researchers have recently studied the interaction between nitrogen-injection position, nitrogen-injection volume, and air-supply volume through the response surface method and Box–Behnken combination experiment. Based on experimental findings, they had optimized the process parameters for continuous and precise nitrogen injection in goaf areas [14].

With the development of inert gas fire prevention and extinguishing technology, more experimental observations suggest that CO₂ exhibits superior inerting effects on residual coal in goaf compared to N₂. Wu et al. found through experimental studies that residual coal in the goaf tended to adsorb CO₂ more readily compared to N₂ [15]. Therefore, when CO₂ was injected, more adsorption sites would be occupied on the coal surface, and an isolation layer would be formed quickly to prevent the coal from contacting oxygen. Consequently, the research hotspot gradually focuses on CO₂ fire prevention and extinguishing technology. In recent years, researchers have conducted a study to investigate the relationship between CO₂ injection and safe production by establishing a three-dimensional physical model of the goaf. They ultimately identified optimal injection parameters to balance the inerting effect of CO₂ with production safety [16]. Other studies utilized a self-developed CO₂ adsorption experimental device to investigate the influencing factors of CO₂ adsorption by coal. Eventually, it theoretically improved the existing gas-injection equation, offering a new perspective for CO₂ fire prevention and control in a goaf [17]. Liu et al. proposed a double-pipe liquid CO₂ injection technology through theoretical analysis, field experiments, and numerical simulations [18]. This study can provide guidance and reference for fire prevention and extinguishing in goaves under the condition of narrow coal pillar-utilized deep mining.

With the expanding scope of coal mining, the demand for prevention and control of coal spontaneous combustion in goaves continues to rise. Solely relying on CO₂ or N₂ may require help that simultaneously addresses fire prevention, cost, and safety concerns. The composite inert gas of CO₂ and N₂ may be a feasible solution. Taking the 5306 working face of the Tangkou Coal Mine as an example, Tang et al. tested the effect of composite inert-gas fire prevention and extinguishing. The results showed that the coal spontaneous combustion in a goaf was effectively prevented and controlled after using a composite inert gas, and its cost was also significantly reduced [19,20].

Through the analysis of the research status of inert gas fire prevention and extinguishing technology, it can be seen that the use of composite inert gas to prevent coal spontaneous combustion in goaves is gradually becoming a hot spot. However, research explicitly targeting composite inert gases is relatively limited. Therefore, there is an urgent need for further testing of the inerting effect of composite inert gases to promote the development of this technology. Compared with previous studies, this paper constructed a physical similarity simulation experiment platform to study the effect of replacing CO₂ with premixed inert gas of CO₂ and N₂ on the inerting effect. Moreover, the formation of the inerting blind area and the accumulation state of premixed inert gases during the inerting process were analyzed. This study can guide mine fire prevention and extinguishing work, holding significant practical implications for preventing and controlling coal spontaneous combustion in goaves.

2. Materials and Methods

2.1. Experimental Platform

The design of the physical similarity simulation experiment platform needs to be based on specific similarity parameters. This study adopted the Froude similarity criterion to construct a small-scale similarity simulation platform at a 1:60 scale based on conventional goaf conditions. The fluid-flow velocity must be scaled according to the Froude similarity criterion to ensure that the actual conditions and the similar model meet the corresponding dynamic similarity conditions. The Froude scale criterion is shown in Table 1.

The finalized parameters of the experimental platform are depicted in Table 2, as presented.

In addition, through numerical simulation, the porosity range of the goaf is 0.13–0.38, and the permeability range is 3.63 × 10⁻⁷ m²–1.78 × 10⁻⁵ m².

The experimental platform’s schematic diagram, illustrated in Figure 1, encompassed the goaf module (a), negative pressure ventilation module (b), data-acquisition module (c), and inert-gas injection module (d).

The goaf module served as the main component of the experimental system, as depicted in Figure 2. The simulated goaf boundary layer was constructed using a 50 mm angle steel framework adhered to organic glass acrylic panels. The bottom plate was assembled from four 2 mm thick stainless-steel plates. Additionally, the platform allowed for arbitrary rotation between −90° and 90° to simulate different inclinations of the goaf.

After coal mining, the overlying strata of the goaf undergo deformation and failure due to the effect of formation stress. In the vertical direction, due to the difference in the failure range and failure strength of rock strata at different heights, it is divided into three different regions: caved zone, fractured zone, and bent subsidence zone. In the horizontal direction, based on the theory of overlying strata movement in the goaf and by analyzing the fragmentation and expansion characteristics of the roof and caved rock, the goaf can be divided into three different zones: natural accumulation zone, load-influenced zone, and compacted stable zone. The model established in this study simplified the goaf. During the design, the strata stress load was simplified, and no loading stress was added. In the vertical direction, the model only considered the caved zone and the fractured zone. In the horizontal direction, the model omitted the three areas inside the goaf and instead made assumptions based on the ‘O-ring’ theory.

The material in the goaf is highly destroyed into irregular shapes of varying sizes, resulting in a high porosity. Therefore, the goaf can be considered an entirely fragmented porous media [21]. To facilitate the alteration of various conditions simulating the goaf and realize the various movements of the gas in the real goaf, foam blocks were employed to fill the simulated goaf instead of residual coal and rubble [22]. In the simulated goaf, 50–100 mm irregularly shaped foam blocks were used to fill the caved zone, and 100–300 mm irregularly shaped foam blocks were used to fill the fractured zone [23]. Furthermore, in accordance with the ‘O-ring’ theory, the permeability is highest around the periphery of the goaf, decreasing progressively towards the central region [24]. Therefore, the foam blocks in the center of the simulated goaf were compacted to a certain extent to make the permeability lower than that around the perimeter of the simulated goaf [25].

The negative pressure ventilation module comprised flowmeters and a vacuum pump. The flowmeter monitored the gas flow within the pipeline, while the vacuum pump, connected to the return airway via an exhaust-pipe interface, achieved negative pressure ventilation.

The data-acquisition module included oxygen concentration sensors, a receiver, and a computer. The oxygen concentration sensor utilized an electrochemical sensor with a measurement range of 0–25%vol and a response time of less than 15 s. The layout of the sensors is shown in Figure 3. The plane layout of the sensor measuring points was 12, and the longitudinal distribution was 5 layers.

The inert-gas injection system comprised premixed inert-gas cylinders, pressure-reducing valves, flowmeters, and related components. The premixed inert-gas cylinder simulated the source of inert gas injected into the goaf, and the pressure-reducing valve was connected to the cylinder to ensure the continuous and smooth outflow of the premixed inert gas.

2.2. Experimental Method

Utilizing ambient-temperature N₂ to perform the process of isochoric heating gasification on liquid CO₂ [26], a premixed inert gas with N₂ ratios of 50%, 60%, 70%, and 80% was produced. Additionally, to facilitate a more direct comparison of the enhancement in inerting effectiveness, experiments injecting CO₂ as a single gas were conducted. The specific experimental parameters are shown in Table 3, where the symbols ‘+’ and ‘−’ denote upward and downward ventilation, respectively.

Prior to the commencement of the experiment, the sealing integrity of the simulated goaf must be verified. Subsequently, the inclination angle and ventilation mode of the goaf module were adjusted according to the set experimental parameters. Open the data-monitoring system to check whether the oxygen concentration in the simulated goaf was 20.95%. After that, the negative pressure ventilation system was opened to adjust the negative pressure and air extraction flow to ensure that the air speed of the working face was 0.25 m/s. At the same time, it is necessary to ensure that the negative pressure ventilation system provides continuous and stable ventilation to simulate the typical ventilation environment of the working face. Then, open the valve of the inert-gas cylinder and adjust the pressure-reducing valve to ensure the continuous and stable injection of premixed inert gas with a flow rate of 37 L/min into the simulated goaf. The data acquisition system monitored and recorded the oxygen concentration within the simulated goaf in real time. Upon reaching the predetermined inerting duration, the experiment was stopped. Simultaneously, the oxygen concentration values at each measurement point were recorded for further analysis.

3. Results

Due to its higher density compared to air, CO₂ injected into the goaf will diffuse at the bottom of the goaf. Similarly, the premixed inert gas composed of CO₂ and N₂, with densities higher than that of air, will predominantly influence the lower regions of the goaf, akin to CO₂. Additionally, most residual coal within the goaf is situated at its bottom. Hence, inerting focused on the low level of the goaf, corresponding to layer 1 of the simulated goaf. Therefore, the analysis in this study focused on the data obtained from 12 oxygen concentration sensors located within layer 1.

3.1. Inerting Effect of Premixed Inert Gas in Nearly Horizontal Goaf

To investigate the temporal variation of the oxygen concentration when different proportions of premixed inert gas were injected under the condition of a nearly horizontal goaf, the oxygen concentration at monitoring points 1, 4, 7, and 10 on layer 1 were taken as examples for analysis. As shown in Figure 4, the higher the N₂ proportion of the pre-mixed inert gas, the shorter the time to reach dynamic equilibrium. Taking measuring point 4 as an example, it took about 20 min to reach equilibrium when N₂ accounts for 60% and about 5 min when N₂ accounts for 80%. At N₂ proportions between 50% and 70%, the O₂ concentration of measuring point 4 was the lowest, indicating that the flow toward the working surface dominated the flow of premixed inert gas on the air-intake side.

To analyze the inerting effect of different ratio gases, the O₂ concentration diagram of the different measuring points was drawn according to the data of layer 1, as shown in Figure 5. Furthermore, according to the ‘Coal Mine Safety Regulation’, the oxygen concentration of the mine air must not be less than 20% to ensure workers’ safety. Therefore, in this study, this concentration was regarded as the normal oxygen concentration in the model and was annotated in Figure 5 for subsequent analysis.

The inerting effect had been significantly improved after using premixed inert gas instead of CO₂. When the proportion of N₂ was 70%, the inerting effect was optimal. The average O₂ concentration in the monitored area decreased from 9.7% when CO₂ was used to 6.4%. Furthermore, after the inert gas was injected into the air-intake side, the inerting effect gradually diminished from the air-intake side to the middle and then to the air-return side.

In the upper part, when N₂ accounts for 50% to 70%, the main motion path was opposite to the direction of the working face. Furthermore, according to Figure 5, as the proportion of N₂ in the premixed inert gas increased, the O₂ concentration at measuring point 4 decreased, reaching 0% when the proportion reached 70%. When the proportion of N₂ was 80%, the inert gas was more likely to flow to the deep part of the goaf, and the O₂ concentration at measuring point 10 decreased to 0%. The experimental results indicated that, within a specific range of N₂ proportions, premixed inert gases with smaller densities tended to flow back toward the working face. However, inert gases outside this range, regardless of their density, tended to flow into the deeper regions of the goaf after flowing out of the injection port.

The inerting effects of different proportions of inert gases were relatively close in the middle area and lower part of the goaf. At N₂ proportions of 50%, 60%, 70%, and 80%, the average O₂ concentrations in the middle area were 7.9%, 7.4%, 6.9%, and 7.6%, respectively, while those in the lower part were 11.2%, 11.3%, 10.7%, and 12.3%, respectively. According to the experimental data, when N₂ accounts for 50%, the difference in the O₂ concentration between measuring points 2 and 11 was 2.0%, while the difference between measuring points 3 and 12 reached 5.3%. This suggested that, after the inert gas flowed to the lower part, it rapidly converged towards the air return of the goaf.

3.2. Inerting Effect of Premixed Inert Gas in Gently Inclined Goaf with Downward Ventilation

To investigate the inerting effect of premixed inert gases in a gently inclined goaf when using U-shaped downward ventilation, the simulated goaf was rotated to 20°, and the inerting experiment of premixed inert gas was conducted. During the experiment, the variation of the O₂ concentration with time is shown in Figure 6.

In a gently inclined goaf with downward ventilation, the higher the proportion of N₂ in the premixed inert gas, the longer it took to reach dynamic equilibrium. Taking measuring point 4 as an example, it took approximately 10 min to reach equilibrium when N₂ was a proportion of 60%, while at N₂ proportions of 70% and 80%, it took approximately 20 min. Compared with the nearly horizontal goaf, the premixed inert gases with higher densities reached dynamic equilibrium faster, while those with lower densities took longer to achieve dynamic equilibrium.

In addition, compared to the nearly horizontal goaf, under downward ventilation conditions in a gently inclined goaf, the primary influence position of inert gas in the upper part shifted from measuring point 4 to measuring points 7 and 10. After the inert gas flowed out from the injection nozzle, its primary flow direction changed from contraflow towards the working face to along the inclination (in the direction of gravity component force) or the strike direction (to the deep part of the goaf). As shown in Figure 6, when CO₂ was injected, the influence on measuring point 10 was weaker compared to measuring point 7. As the proportion of N₂ increased, the influence on measuring points 7 and 10 tended to align. When N₂ accounts for 80%, measuring point 10 experienced a more substantial impact than measuring point 7. The experimental phenomena demonstrated that, when injecting premixed inert gases with higher densities, its primary motion was along the inclination, while with lower densities, it was towards the deeper part of the goaf. Moreover, with the increase in the proportion of N₂ in the premixed inert gas, its primary motion gradually transitioned from along the inclination to the strike direction.

To analyze the inerting effect of premixed inert gas with different proportions under downward ventilation, the simulated goaf layer 1 was taken as the primary research object, and the O₂ concentration curves for various measuring points were drawn, as shown in Figure 7. Additionally, Figure 7 was processed in the same way as Figure 5.

According to Figure 7, the premixed inert gas’s inerting effect was superior to that of CO₂ when downward ventilation was applied in a gently inclined goaf. Optimal inerting effectiveness was achieved when the N₂ proportion was 70%, and the average O₂ concentration decreased from 15.1% when CO₂ was used to 11.2%.

At the upper part of the gently inclined goaf, the gravity field significantly affected the flow of inert gas, shortening the residence time of inert gas at the upper part and ultimately leading to a substantial decrease in inerting effectiveness. Under the influence of different proportions of inert gas, the average O₂ concentration at the upper part increased from 5.1%, 3.0%, 2.2%, 1.5%, and 1.5% in the nearly horizontal goaf to 15.3%, 13.8%, 12.1%, 10.9%, and 10.9%. The inclined floor caused the collapsed coal and rock to concentrate in the lower part, resulting in a looser structure and increased porosity at the upper part relative to the lower part. The flow resistance decreased in the region in front of the inert-gas injection pipe. In contrast, the lower-density inert gases were less affected by gravity in the inclination direction, leading to similar inert-gas flows in both the inclination and the strike directions. The experimental data showed that, when N₂ accounts for 70%, the difference in the O₂ concentration between measuring point 7 and measuring point 10 was only 0.1%. However, when the N₂ proportion increased to 80%, the O₂ concentration of measuring point 10 decreased to 5.9%, which was lower than that of measuring point 7. This indicated that the inert-gas flow in the strike direction exceeded that in the inclination.

For the middle area and lower part of the goaf, gravity along the inclination continued to take effect, so the primary flow direction of inert gas was still the inclination. When the proportion of N₂ was 70%, the O₂ concentrations of measuring points 8 and 9 were 9.4% and 10.9%, respectively, which were lower than those of other measuring points in the middle area and lower part.

3.3. Inerting Effect of Premixed Inert Gas in Gently Inclined Goaf with Upward Ventilation

To investigate the inerting effect of premixed inert gases in a gently inclined goaf when using U-shaped upward ventilation, the simulated goaf was rotated to 20°, and the inerting experiment of premixed inert gas was conducted. During the experiment, the variation of the O₂ concentration with time is shown in Figure 8.

Different from the above two scenarios, the dynamic equilibrium state did not occur within the simulated goaf within the same time frame. After the sudden mutation in O₂ concentration at each measurement point, it uniformly and steadily decreased at a certain rate. Similar to downward ventilation, under the effect of upward ventilation, when the N₂ proportion in the premixed inert gas was 50% and 60%, the inert gas still moved mainly along the inclination in the upper part. When the N₂ proportion was 70%, the inert gas flow in the strike and inclination directions was relatively balanced. When the N₂ proportion increased to 80%, the inert gas moved primarily along the strike direction.

In the upward ventilation state, the air-intake side of the platform is located at the lower part, while the air-return side is situated in the upper part. When the inert gas flows towards the lower part, in addition to the flow resistance, it is also necessary to overcome the influence of air leakage on the inert gas. To analyze the inerting effect of premixed inert gas with different proportions under upward ventilation, the simulated goaf layer 1 was taken as the primary research object, and the O₂ concentration curves for various measuring points were drawn, as shown in Figure 9. Additionally, Figure 9 was processed in the same way as Figure 5.

When employing upward ventilation, the inerting effect of premixed inert gas remained superior to that of CO₂. Furthermore, the inerting effect was optimal when the N₂ proportion was 80%. The average O₂ concentration in the monitored area decreased from 12.1% when using CO₂ to 8.4%. In the upper part of the gently inclined goaf, upward ventilation was similar to downward ventilation, where gravity and pore distribution still predominantly influenced the direction of inert-gas movement. It could be seen from the experimental data that, when the N₂ proportion was 50%, the O₂ concentrations of measuring points 7 and 10 were 6.9% and 9.8%, respectively. When the N₂ proportion increased to 80%, the O₂ concentrations of these two measuring points became 4.0% and 3.7%, respectively. This indicated that, with the increase of the proportion of N₂ in the premixed gas, the gas transitioned from predominantly inclination movement to strike movement predominantly, and the inerting effect continued to enhance.

In the middle area of the gently inclined goaf, the main influence area of premixed inert gas with N₂ proportions of 60% and 70% was the same as that of downward ventilation, which was located at measuring point 8 and measuring point 11. When the N₂ proportion in the premixed inert gas was 60%, the O₂ concentrations at measuring points 8 and 11 were 8.6% and 9.3%, respectively. That is, in the gently inclined goaf, when employing either upward or downward ventilation without generating flow around near the compaction area, the inert gas in the middle area was more prone to flow towards the deep part of the goaf, in addition to its inclination movement towards the lower part along the floor. In the lower part of the gently inclined goaf, similar to the case with downward ventilation, the O₂ concentration at various measuring points remained relatively uniform under different proportions of premixed inert gas. This showed that the pore distribution significantly affected the migration of premixed inert gas in the lower part.

4. Discussion

When the inert gas entered the goaf, it would be affected by the pore flow resistance and gravity, resulting in the premixed inert gas being bypassed near the compaction zone and flowing to the lower part, thus forming an inerting blind zone [27]. The so-called inerting blind zone referred to the fact that the oxygen concentration in the region did not change significantly after the injection of inert gas. That is, the oxygen concentration was still within the normal oxygen concentration range.

Figure 5 and Figure 9 showed that, in the nearly horizontal goaf and gently inclined goaf with upward ventilation, injecting premixed inert gas did not result in the formation of an inerting blind zone. Notably, in the nearly horizontal goaf, using premixed inert gas instead of CO₂ eliminated the inerting blind zone formed when CO₂ was used. According to Figure 7, in the gently inclined goaf with downward ventilation, there was no inerting blind zone only when the proportions of N₂ were 60% and 70%. Furthermore, the inerting blind zone was mainly distributed at measuring point 11 in the middle area of the goaf.

In the gently inclined goaf, the flow rate of premixed inert gas in the floor was accelerated due to the influence of gravity and porosity, resulting in a shortened time to flow into the lower part. Meanwhile, as the porosity decreased in the lower part, the flow resistance of the inert gas increased, causing it to flow out of the goaf slowly. Ultimately, this led to the accumulation of premixed inert gas in the lower part of the goaf. With the accumulation of inert gas in the lower part exceeding that in the middle area and upper part, the O₂ concentration in this area was reduced to a minimum.

As illustrated in Figure 10, in the gently inclined goaf with downward ventilation, when premixed inert gas was injected, the O₂ concentration of measuring point 3 and measuring point 6 near the outlet of the air-return side in the lower part was lower than that of measuring point 2 and measuring point 5 in the middle area.

When the proportion of N₂ was 50%, the difference in O₂ concentration between measuring point 3 and measuring point 2 was 4.4%, and the difference in O₂ concentration between measuring point 6 and measuring point 5 was 0.6%. At an N₂ proportion of 70%, the O₂ concentration differences were 3.2% and 0.3%, respectively. When the N₂ proportion was 80%, the O₂ concentration differences were 3.6% and 0.2%, respectively. This showed that the accumulation area was mainly located near measuring point 3 near the outlet of the air-return side in the lower part. Furthermore, the premixed inert gas with a better inerting effect exhibited a weaker accumulation state.

When using upward ventilation, the accumulation state of premixed inert gas still persisted in the lower part of the gently inclined goaf. The accumulated gas was premixed inert gas with an N₂ proportion of 50–70%, and the main accumulation area was still located near measuring points 3 and 6. As shown in Figure 11, when the proportion of N₂ in the premixed inert gas was 70%, the difference in O₂ concentration between measuring points 3 and 2 was 1.6%, and the difference in O₂ concentration between measuring points 6 and 5 was 0.2%, which was lower than 3.2% and 0.3% in downward ventilation. This phenomenon showed that the accumulation state was weakened. When the N₂ proportion in the premixed inert gas was 80%, the O₂ concentrations of measuring points 5 and 6 were 8.8% and 9.9%, respectively. In contrast, measuring points 2 and 3 were 10.8% and 10.2%, respectively, indicating that the accumulation state was partially eliminated.

In the nearly horizontal goaf, the inerting effect of premixed inert gas was significantly better than that of CO₂. However, there were several issues with the gently inclined goaf. The retention time of the premixed gas in the upper part was short, resulting in a poor inerting effect, an inerting blind zone in the middle part, and a noticeable accumulation state occurred in the lower part. Therefore, future research can be based on the shortcomings of a single injection method in the gently inclined goaf and further study the inerting effect of a synergistic injection of premixed inert gas and N₂ in a gently inclined goaf. The so-called synergistic inerting refers to the collaborative injection of premixed inert gas at the upper part of the goaf and N₂ at the lower part. Specifically, it can be categorized into two scenarios: one involves the simultaneous injection of premixed inert gas at the upper part of the goaf and N₂ at the lower part, while the other entails the initial injection of premixed inert gas at the upper part, followed by a duration of time before coordinating with an injection of N₂ at the lower part of the goaf. Subsequent studies may be conducted based on the aforementioned methods, providing guidance for mine fire prevention and control.

It is worth noting that the model established in this study simplified the goaf. In the vertical direction, the model only considered the caved zone and the fractured zone. In the horizontal direction, the model omitted the three areas inside the goaf and instead made assumptions based on the ‘O-ring’ theory. Therefore, the physical similarity simulation experiment platform constructed in this study had limitations in reproducing the conditions of an actual goaf. Thus, it could not fully reflect the nature of the goaf in the real mine. The conclusions obtained in this study were all obtained in the laboratory environment, and may not always be accurate in the real goaf. Therefore, this study has reference significance for the prevention and control of coal spontaneous combustion in mines.

5. Conclusions

This paper studied the effect of using a premixed inert gas (N₂ accounted for 50%, 60%, 70%, and 80%) instead of CO₂ on the inerting effect of nearly horizontal and gently inclined goaves by building a physical similarity simulation experiment platform. Based on the findings, the main conclusions are as follows.

• In the nearly horizontal goaf, the inerting effect was significantly improved after replacing CO₂ with premixed inert gas, and the inerting blind zone was eliminated. The optimal inerting effect was achieved when the N₂ proportion in the premixed inert gas was 70%, and the average O₂ concentration in the monitoring area decreased from 9.7% when CO₂ was used to 6.4%. In the upper part of the goaf, the higher the proportion of N₂ in the premixed inert gas, the shorter the time required to achieve dynamic equilibrium. In the middle and lower parts of the simulated goaf, the inerting effect of premixed inert gas diminished with the increase of the distance from the working face. However, the variation magnitude between the measuring points in the middle was relatively small, while the variation magnitude between the lower part measuring points was relatively large;
• In the gently inclined goaf with downward ventilation, the inerting effect of premixed inert gas was superior to that of CO₂. However, there was no inerting blind zone, only when the proportion of N₂ was 60% and 70%. When the proportion of N₂ in the premixed inert gas was 70%, the inerting effect was the best, and the average O₂ concentration in the monitoring area decreased from 15.1% when CO₂ was used to 11.2%. Compared with the nearly horizontal goaf, the premixed inert gases with higher densities reached dynamic equilibrium faster, while those with lower densities took longer to achieve dynamic equilibrium. The premixed inert gas formed an accumulation state near the outlet of the air-return side in the lower part. Still, the premixed inert gas with a better inerting effect exhibited a weaker accumulation state;
• In the gently inclined goaf with upward ventilation, the inerting effect of premixed inert gas remained superior to that of CO₂, and there was no inerting blind zone. The optimal inerting effect was achieved when the N₂ proportion in the premixed inert gas was 80%, and the average O₂ concentration in the monitoring area decreased from 12.1% when CO₂ was used to 8.4%. However, the goaf had no dynamic equilibrium within the same time frame. The accumulation state of premixed inert gas was weaker than in downward ventilation, with the accumulation area primarily located at the air intake. The stronger the inerting effect of the premixed inert gas, the weaker the accumulation state. When the proportion of N₂ reached 80%, the accumulation state was partially eliminated.

The physical similarity simulation experiment platform constructed in this study had limitations in reproducing the conditions of an actual goaf. Thus, it could not fully reflect the nature of the goaf in the real mine. The model built in this study simplifies the goaf, including no formation stress, no consideration of convection, no heat and moisture supply inside the goaf, and no consideration of the sorption of CO₂ and N₂ on coal. The conclusions obtained in this study were all obtained in the laboratory environment and may not always be accurate in the real goaf. Therefore, this study has reference significance for the prevention and control of coal spontaneous combustion in mines.