Experimental and Mechanistic Study of Geometric Asymmetry Effects on Gas–Coal Dust Coupling Explosions in Turning Pipelines

Abstract

The geometric symmetry of the pipeline constitutes a critical determinant in regulating the energy propagation dynamics during the explosion process. In the present study, a transparent plexiglass pipe experimental system incorporating a range of angles (30° to 150°) was meticulously constructed. Leveraging high-frequency pressure sensors in conjunction with high-speed camera technology, this investigation examines the influence of the pipe angle, which disrupts geometric symmetry, on the coupling explosion of gas and coal dust. The experimental findings illustrate that an increase in the pipeline turning angle significantly enhances the velocity of the explosion flame front (with the maximum velocity escalating from 97.92 m/s to 361.28 m/s) and concurrently reduces the total propagation time (from 71 ms to 56.5 ms). Moreover, there is a notable reduction in the duration of the explosion flame, decreasing from 240.5 ms to 64.17 ms at the coal dust deposition point. The peak overpressure of the shock wave exhibits a significant increase with the augmentation of the turning angle (rising from 7.07 kPa at 30° to 88.40 kPa at 150°). Furthermore, the overpressure in the fore section of the turning is amplified, attributable to the superimposition of reflected waves and turbulent effects. This study elucidates critical mechanisms including turbulence-enhanced combustion, secondary dust generation from coal dust, and energy dissipation resulting from abrupt alterations in pipeline geometry, thereby offering a theoretical framework for the prevention and effective emergency management of coal mine explosion disasters.

1. Introduction

Gas–coal dust coupling explosions constitute a highly destructive disaster phenomenon prevalent in industrial settings, such as coal mines, characterized by intricate propagation dynamics that present significant threats to both personnel and infrastructure [1,2,3]. Over the past two decades, a total of 223 explosion incidents have been recorded in coal mines throughout China, culminating in 1475 fatalities. The occurrence of gas–coal dust coupling explosions entails a composite reaction between gas and coal dust, during which the resultant shock wave overpressure and high-temperature flames emerge as the principal contributors to destruction. Consequently, an exhaustive investigation into the transmission characteristics of shock wave overpressure and flame propagation in gas–coal dust coupling explosions is imperative for accurately delineating the spatial extent and severity of explosion hazards. Additionally, it lays a scientific foundation for formulating effective coal mine disaster prevention strategies and emergency protocols.

In recent decades, considerable scholarly efforts have been devoted to systematically investigating the dynamics of shock wave overpressure and flame propagation within the context of gaseous explosions. The research on gas–coal dust coupling explosions encompasses a multitude of dimensions, primarily concentrating on the fundamental mechanisms underlying ignition sources and the intrinsic properties governing coal dust combustibility.

Researchers have demonstrated that coal dust can release substantial quantities of combustible gas, which, in most instances, renders gas–coal dust coupling explosions significantly more powerful than standalone gas explosions [4,5,6,7,8]. Ajrash et al. [9] performed methane/coal dust explosion experiments within large cylindrical pipelines, discovering that the incorporation of 30 g/m³ of coal dust into a 6% volume concentration of methane/air resulted in nearly a doubling of the explosion pressure. Wang et al. [10] investigated the characteristics and influencing factors of methane and coal dust gas-solid mixture explosions through experimental methods, revealing that the physical parameters and internal components of coal dust significantly affect the reaction rate of gas/dust mixture explosions. Song et al. [11] conducted a series of investigations within a 20 L spherical explosion container, revealing that the concentrations of methane and coal dust, along with the initial pressure, exert significant effects on mixed explosions. Liu et al. [12] examined the explosion characteristics of coal dust and gas mixtures with varying particle sizes, discovering that coal particle size significantly influences the dynamics of explosion propagation, with smaller particles generating faster flame speeds. Taviau et al. [13] carried out a series of studies on chemical igniters, revealing that the elevated temperatures produced during the ignition process induce a preheating effect on the gas and dust in the surrounding unburned regions, thereby affecting the overall explosion characteristics of the gas and dust mixture. Cloney et al. [14] investigated the impact of ignition heads on dust distribution, emphasizing that these ignition heads facilitate dust movement, thereby altering the overall distribution of dust particles. Kundu et al. [15] conducted experiments examining the propagation of gas and coal dust overpressure within spherical pipeline containers, exploring the effects of coal dust addition and ignition energy on explosion pressure. Conversely, researchers have undertaken extensive investigations into suppression and prevention techniques associated with gas and coal dust explosions. Si, Cai, Qu, and colleagues [16,17,18,19,20] performed extensive research on gas and coal dust explosions using large-scale simulation experimental systems, providing a detailed analysis of the various explosion characteristics inherent to gas and coal dust. Pei et al. [21] investigated the explosion suppression efficacy of CO₂ ultrafine water mist in gas/coal dust composite systems, analyzing its effects across three dimensions: overpressure, flame propagation speed, and flame structure. Li, Wei, Wang, Song, and their colleagues [22,23,24,25,26,27] performed a series of studies focused on the sedimentary coal dust explosions induced by gas explosions. Li et al. [28] established a small-scale experimental platform aimed at analyzing the effectiveness of fine water mist in mitigating pipeline mixture explosions. Li [29] conducted a comparative study examining the explosion suppression effects of a single explosion suppressant alongside CO₂ ultrafine water mist, while also exploring the underlying suppression mechanisms. Huang et al. [30] conducted performance tests on explosion-proof powder cloud curtains within large-scale cross-section tunnels, exploring various prevention and control methods for gas and coal dust explosions. Lastly, researchers have undertaken investigations regarding the influence of numerical simulations and experimental parameters on explosion characteristics. Wei et al. [31,32] examined the explosion characteristics of methane coal dust utilizing numerical simulation methodologies, emphasizing the advantages provided by numerical simulations in mitigating the limitations associated with experimental research.

In conclusion, while numerous studies have been conducted on the overpressure and flame propagation characteristics of gas–coal dust coupling explosions, the majority of these investigations have predominantly focused on straight pipes or open spaces. There remains a notable deficiency in the research addressing the impact of geometric asymmetry in pipe corners on gas–coal dust coupling explosions. The influence mechanisms underlying the pipeline turning angle on critical parameters—such as explosion shock wave overpressure, flame front velocity, and duration—require further elucidation. Building upon the previous research findings, this study systematically investigates the pressure and flame propagation characteristics of gas–coal dust coupling explosions within turning pipelines, utilizing transparent organic glass pipeline test systems designed with varying turning angles. The experiments are conducted under composite explosion conditions involving methane–air premixed gas (gas concentration of 7.5%) and coal dust (concentration of 100 g/m³, particle size of 48–75 μm). This study integrates high-frequency pressure sensors and high-speed camera technology to provide a detailed analysis of the influence of pipeline turning angles (30° to 150°) on explosion shock wave overpressure, flame front velocity, and duration. The primary objective of this study is to elucidate the propagation mechanisms of gas–coal dust coupling explosions within turning pipelines, thereby offering a theoretical foundation and technical support for the prevention and control of explosion incidents in industrial environments, such as coal mines.

2. Experimental System and Plan

2.1. Test System

The experimental system included experimental pipelines, gas distribution systems, ignition systems, pressure data acquisition systems, high-speed camera systems, and synchronous control systems.

The experimental system was composed of six core components: the test pipeline, gas distribution system, ignition system, pressure data acquisition system, high-speed imaging system, and a synchronous control system. The pipeline itself was constructed from transparent organic glass and features various turning angles to simulate geometric transitions. Structurally, it consisted of two segments: a horizontal section measuring 850 mm and an inclined section of 750 mm in length. The pipeline possessed a square cross-section of 80 mm × 80 mm, a wall thickness of 20 mm, and a compressive strength of up to 2 MPa. Although underground coal mine tunnels typically exhibit diverse cross-sectional geometries—such as arched or rectangular profiles—this study employed square-section transparent pipelines. The use of high-transparency organic glass allowed for the direct observation of flame propagation through high-speed imaging, thus enabling effective visualization of turbulence intensification and flame–wall interactions induced by geometric discontinuities. Due to experimental constraints, the current setup does not fully replicate long-distance flame acceleration or pressure wave reflections as observed in full-scale tunnels. Nonetheless, it provides critical mechanistic insights into early-stage explosion phenomena, such as turbulence enhancement. One end of the pipeline remained hermetically sealed, while the opposite end was sealed with a PVC membrane, serving as a pressure release boundary. The gas delivery system consisted of a methane cylinder, an air compressor, and dual mass flow controllers (ALICAT, Tucson, AZ, USA) to ensure precise methane–air premixing at the desired 7.5% volumetric concentration. Ignition was achieved using a high-energy igniter and paired electrodes (6 kV, 2.5 J), placed 100 mm from the sealed terminal. Pressure evolution was captured in real time via an MD-HF high-frequency pressure sensor (20 kHz sampling rate) (MEOKON, Shanghai, China), interfaced with a USB-1608FS acquisition module (NI MCC, Austin, TX, USA). Flame development and propagation were recorded through a high-speed camera (Revealer, Hefei, China) operating at 2000 frames per second with a resolution of 1024 × 1024 pixels, enabling the precise tracking of flame front displacement and structural morphology. All subsystems—including ignition, pressure measurement, and high-speed imaging—were synchronized using a centralized control unit (OMRON, Kyoto, Japan) that executed predefined time sequences to ensure coordinated data acquisition. Figure 1 presents the schematic layout of the experimental setup.

The interconnections of the experimental setup are illustrated in Figure 1, where all system components are assembled according to the schematic configuration. A PVC membrane barrier was affixed at the designated interface prior to each trial to seal one end of the pipeline. A pre-determined mass of coal dust was evenly distributed at the central section of the horizontal pipe, precisely 200 mm from the ignition point located at the sealed left end of the pipeline. Upon completing dust placement and pipeline sealing, intake and exhaust valves were simultaneously opened to initiate ventilation. Methane and air inflow rates were regulated via dual mass flow controllers, enabling accurate premixing at the target concentration. The premixed methane–air gas was introduced into the pipeline, displacing residual gases, which were vented through the outlet at the opposite end. To ensure the complete purging of the original gas within the pipeline, the total volume of injected mixture was set to four times the internal volume of the pipeline. Following inflation, all valves—including those on the gas cylinder, compressor, and inlet/exhaust—were closed, and the flow controllers were deactivated, completing the pre-test preparation phase. To guarantee full synchronization and complete data acquisition during the explosion event, the pressure data acquisition system and the high-speed imaging system were both activated prior to ignition. Upon pressing the start button on the synchronous controller, the pressure and imaging systems were immediately triggered (t = 0 ms), while the ignition system was activated with a 10 ms delay (t = 10 ms). Upon completion of the experiment, all experimental data were organized and saved, the pipeline was thoroughly cleaned, and preparations for the subsequent trial were carried out.

2.2. Experimental Plan Design

In this study, the composition of coal dust varied significantly, leading to diverse test results. The coal samples utilized during the experiments were acquired from the No. 8 Mine of Pingdingshan Tian’an Coal Mining Co., Ltd. (Pingdingshan, China), classified as bituminous coal. To ascertain the precise composition of the coal dust, key parameters—including ash content, moisture, volatile matter, and fixed carbon—were measured in accordance with the GB/T 212-2008 [33] Industrial Analysis Method for Coal. The analytical results are summarized in

Table 1. Industrial analysis of coal dust samples.

Coal Dust Composition	Percentage Content/%
Ash content (Aad)	13.1
Moisture (Mad)	1.25
Volatile matter (Vad)	27.34
Fixed carbon (Fcad)	58.31

. The coal dust concentration was consistently maintained at 100 g/m³, with particle sizes ranging from 48 to 75 μm. Methane gas, with a purity of 99.99%, served as the fuel for this investigation, achieving a target gas concentration of 7.5%. The experimental conditions included an ambient temperature of 20 to 25 °C and a relative humidity of approximately 50%.

In this investigation, seven distinct turning angles—30°, 45°, 60°, 90°, 120°, 135°, and 150°—were integrated into the pipeline configuration, as illustrated in Figure 2. The positions of the pressure transducers are depicted in Figure 2a: P1 is located 200 mm downstream from the ignition source; P2 is positioned 200 mm upstream of the bend apex; P3 is located at the bend apex, which is defined as the intersection of the centerlines of the straight and inclined sections; P4 is situated 500 mm downstream from the bend apex; and P5 is placed 200 mm further downstream from P4. To ensure the reliability and statistical significance of the experimental results, each set of experiments was conducted three or more times. Following established scientific statistical principles, three sets of valid data were retained for each angle as the final experimental dataset.

3. Results and Discussion

3.1. Analysis of the Variation Law of Explosion Flame Front Velocity

Through experimentation, a substantial amount of data pertaining to explosion flame imagery and explosion pressure was collected. The explosion flame image data were systematically organized to extract relevant information regarding the velocity of the explosion flame front at various time intervals. A curve depicting the flame front velocity of the gas–coal dust coupling explosion as a function of time was generated, as illustrated in Figure 3. Additionally, the time required for the explosion flame to propagate from the ignition point to the outlet of the pipeline was measured, and a curve representing the maximum flame front velocity as a function of the turning angle was plotted, as shown in Figure 4.

The flame front velocity curve of the gas–coal dust coupling explosion over time is illustrated in Figure 3. From the figure, it is evident that the velocity curve of the explosion flame front exhibits a discernible trend throughout the coupling explosion of gas and coal dust. During the initial stage of the explosion, the velocity of the flame front remains relatively stable. However, as the explosion progresses, the velocity begins to show a marked upward trend in the range of 45–50 ms. Subsequently, the flame front velocity increases rapidly until it propagates to the outlet of the test pipeline. Moreover, the data indicate that as the turning angle increases, the velocity of the explosion flame front also tends to rise. The increase in turning angle significantly enhances the development of the explosion flame front velocity. Figure 4 presents the time taken for the explosion flame to propagate from the ignition point to the outlet of the pipeline, as well as the maximum flame front velocity as a function of the turning angle. Under varying turning angles of 30°, 45°, 60°, 90°, 120°, 135°, and 150°, the time taken for the explosion flame to reach the outlet of the pipeline was recorded as 71 ms, 66.5 ms, 63.5 ms, 61 ms, 59 ms, 57.5 ms, and 56.5 ms, respectively. As the turning angle increases, the total time for the explosion flame’s propagation gradually decreases. In terms of the maximum flame front velocity, the explosion flame front velocities within the pipeline at turning angles of 30°, 45°, 60°, 90°, 120°, 135°, and 150° were 97.92 m/s, 147.61 m/s, 171.86 m/s, 245.49 m/s, 300.74 m/s, 330.74 m/s, and 361.28 m/s, respectively. These data clearly demonstrate that the maximum flame front velocity consistently rises with increasing pipeline turning angle.

The variation in the pipeline turning angle significantly influences the turbulence inside the pipeline, which in turn has a profound effect on the velocity of the explosion flame front. As flames navigate bends, wall-induced flow reversal creates aerodynamic shear, thereby amplifying turbulent kinetic energy within the curvature zone. The concurrent formation of vortices distorts the flame topology, increases the reaction surface area, and enhances the entrainment of the fresh mixture—factors that are critical for intensifying combustion. Importantly, when flames interact with fuel-rich vortices, turbulent flame coupling occurs through localized entrainment and rapid mixing. Additionally, the turbulence induced by bends facilitates the fluidization of coal dust, resulting in optimal combustible suspensions that accelerate reaction kinetics. As the angle progressively increases, these turbulence-mediated mechanisms are systematically strengthened, leading to exponential flame acceleration downstream of the inflection points.

3.2. Analysis of the Variation Pattern of Explosion Flame Duration

In conducting a comprehensive analysis of explosive flames, the duration of the explosion flame serves as a crucial analytical indicator. In the pipeline section prior to the coal dust laying point (200 mm from the left ignition point), the low brightness of the explosion flame poses challenges for analysis. Observations indicate that this segment does not significantly affect the overall distribution of flame duration within the pipeline. Therefore, this section focuses on the segment extending from the coal dust laying point to the outlet of the pipeline. By organizing the image data of the explosion flame, the duration data for the gas–coal dust coupling explosion flame at various positions along the pipeline were extracted. Utilizing these data, curves representing the duration of the explosion flame under different turning angles were plotted, as shown in Figure 5. Additionally, curves illustrating both the maximum and minimum durations of the explosion flame within the pipeline as a function of the turning angle were generated, as depicted in Figure 6.

The duration curve of the explosion flame under various turning angles is presented in Figure 5. Analyzing the curve reveals that the duration of the explosion flame within the curved pipeline exhibits a discernible pattern. From the coal dust laying point to the outlet of the test pipeline, the duration of the explosion flame gradually decreases. Specifically, notable differences in the duration of the explosion flame are observed at different turning angles. As illustrated in Figure 6, the explosion flames at turning angles of 30°, 45°, 60°, 90°, 120°, 135°, and 150° all exhibit the longest durations at the coal dust laying point, measured at 240.5 ms, 224.67 ms, 191.5 ms, 115 ms, 79.67 ms, 71.33 ms, and 64.17 ms, respectively. Conversely, the shortest durations recorded at the pipe outlet are 18.5 ms, 13 ms, 7.33 ms, 5.67 ms, 3.83 ms, 3 ms, and 2 ms. Overall, it is evident that as the turning angle changes, the duration of the explosion flames within the pipeline follows a specific trend: larger turning angles correlate with shorter overall durations of the explosion flames.

As the turning angle of the pipeline increases, the overall duration of the explosion flame resulting from the gas–coal dust coupling gradually decreases. This phenomenon can primarily be attributed to the turbulent effects of airflow in the region of the pipeline’s turning point. The extinguishment of flames within the pipeline is largely a consequence of the inability of gas and coal dust to continue reacting due to various factors or the complete consumption of reactants. In this experiment, the turbulence generated at the turning point of the pipeline accelerates the reaction rates of the gas and coal dust. This turbulence leads to a rapid consumption of reactants, including gas, coal dust, and oxygen, thereby shortening the duration of the explosion flame. Furthermore, as the turning angle of the pipeline increases, the turbulence inside the pipeline intensifies, which further enhances the reaction efficiency of the reactants and reduces the duration of the explosion flame.

3.3. Analysis of the Overpressure Variation Law of the Explosion Shock Wave

A substantial amount of data concerning the overpressure generated by shock waves from gas–coal dust coupling explosions was acquired through experimental procedures. After organizing and analyzing the experimental data, we obtained the maximum explosion pressure distribution curve, which reflects the overpressure distribution of the gas-coal dust coupling explosion shock waves within the test pipeline. This distribution curve is illustrated in Figure 7.

Figure 7 illustrates the maximum pressure distribution curve of the shock wave overpressure generated by the coupling explosion of gas and coal dust in pipelines with varying turning angles. The analysis reveals that the overpressure distribution patterns in pipelines at turning angles of 30°, 45°, 60°, 90°, 120°, 135°, and 150° exhibit notable similarities. Specifically, in the straight section preceding the pipeline’s turn, the shock wave overpressure remains relatively stable, with only a slight decrease observed near the turning point. Following the transition through the turning section, the overpressure value experiences a rapid decline until it reaches the outlet of the pipeline. A detailed numerical analysis indicates that at a turning angle of 30°, the overpressure values at each measuring point are relatively low, with sensor 2 recording a maximum overpressure value averaging 7.07 kPa. As the turning angle increases to 150°, the maximum overpressure value at the same measuring point significantly rises to 88.40 kPa, while the overpressure values at various positions within the pipeline show a general increasing trend. Notably, regardless of variations in the turning angle, the peak overpressure consistently occurs at the location of sensor 2 before the turn. This finding suggests that this specific area is most significantly influenced by geometric changes and turbulence effects.

The pressure distribution resulting from the coupling explosion of gas and coal dust in a curved pipeline can be attributed to several key factors: (1) Sudden change effect of pipeline geometry: When the shock wave reaches the corner of the pipeline, the abrupt change in the direction of the pipeline’s cross-section leads to the reflection and refraction of the shock wave. The superposition of reflected waves and incident waves results in local pressure peaks. In contrast, refracted waves experience pressure attenuation in the latter part of the bend due to extended propagation paths and energy dissipation. As the turning angle increases, the intensity of the reflected waves also increases, leading to a significant rise in maximum overpressure at measuring point 2 (located in the area prior to the turn). (2) Turbulence-induced acceleration of combustion reaction: The strong turbulence generated at the corner enhances the mixing of gas and coal dust, which increases the combustion rate. The energy released during combustion is rapidly converted into shock wave energy, causing the overpressure in the front section of the turn to rise with the increasing angle. Additionally, the distortion of the flame front caused by turbulence increases the reaction area, further intensifying the rate of energy release. (3) Energy dissipation and differences in propagation paths: After the shock wave passes through a bend, the change in the propagation direction causes energy to disperse over a larger spatial area, while the friction and eddy current dissipation along the pipe wall are enhanced. Consequently, the overpressure value at measuring point 4 (located in the area following the turn) is generally lower than that at measuring point 2, with greater turning angles resulting in more significant pressure attenuation in the rear section of the turn. (4) Secondary dust raising effect of coal dust: The high-speed propagation of shock waves in the front section of a bend lifts sedimentary coal dust, forming a concentrated dust cloud. As the turning angle increases, the intensity of turbulence also rises, leading to an increase in the suspension of coal dust. This enhanced suspension allows for a more thorough mixing of coal dust with gas, resulting in intensified combustion reactions and further elevating the peak overpressure in the front section of the turn.

3.4. Analysis of Pressure Fluctuations at Measurement Points

To investigate the influence of the pipeline’s turning structure on explosion shock wave overpressure, this section analyzes the variations in shock wave overpressure in the upstream and downstream regions adjacent to the turning section. Pressure data from sensors 2 and 4, which are positioned before and after the turning angle, respectively, are selected for this analysis. The temporal changes in pressure data recorded by sensors 2 and 4 are presented in Figure 8 and Figure 9, respectively.

Figure 8 and Figure 9 illustrate the pressure value curves for measuring points 2 and 4 over time, respectively. It can be observed that the pressure at both measuring points gradually increases from 0 to 45 ms, with no significant differences in pressure values noted across varying turning angles. In the time frame between 45 and 50 ms, however, the pressure exhibits a sharp upward trend and peaks between 50 and 55 ms. Under the 150° turning condition, the maximum pressure recorded at measuring point 2 reaches 88.40 kPa, while the peak pressure at measuring point 4 is 71.21 kPa, the latter being significantly lower than the former. Notably, the peak pressure increases with the turning angle, indicating that the geometric conditions of the pipeline exert a significant regulatory influence on the intensity of energy release. Following the peak, the pressure curve rapidly declines to the negative pressure range from 55 to 60 ms, with both measuring points exhibiting a brief negative pressure phenomenon. This observation highlights the characteristics of pressure drop resulting from gas expansion and local flow effects in the later stages of the explosion. Several key factors contribute to the aforementioned phenomena: (1) Dynamic superposition of reflected waves and incident waves: In the region preceding the turn (measuring point 2), the incident shock wave and the reflected wave from the pipeline wall undergo coherent superposition, resulting in a pronounced increase in the pressure peak. As the turning angle increases, the interaction between the reflected wave and the incident wave intensifies, enhancing the superposition effect—particularly at a 150° corner, where the reflected wave approaches vertical incidence. Consequently, pressure at measuring point 2 escalates with increasing angle. Conversely, in the region following the turn (measuring point 4), the shock wave experiences energy dispersion due to the change in propagation direction, leading to a diminished intensity of reflected waves and a reduced superposition effect. This results in generally lower pressure values than those observed at measuring point 2. (2) Turbulence-induced combustion instability: The strong turbulence generated at bends not only accelerates combustion reactions but also results in intermittent fluctuations in the combustion rate, manifesting in oscillatory characteristics within the pressure curve. Measuring point 2 is situated in an area where turbulence is enhanced, exhibiting large amplitude, regular oscillations in pressure. In contrast, the turbulence experienced at measuring point 4 becomes increasingly complex due to energy dissipation, forming multi-scale vortices that contribute to smaller, stepped fluctuations in pressure. (3) Mechanism of negative pressure formation: In the later stages of the explosion, rapidly expanding and diffusing ignited gas creates a low-pressure zone in localized areas, induced by the swift flow of gas along the pipeline outlet.

Furthermore, the pressure fluctuation characteristics between the two measuring points exhibit distinct differences. Measuring point 2 (P2) predominantly displays periodic high-amplitude waveforms, while measuring point 4 (P4) shows complex modulations characterized by superimposed discrete perturbations within dominant oscillations. This disparity can be attributed to the following factors: (1) Interference effect of multiple reflected waves: Measuring point 4 is situated in the downstream section of the bend, where the shock wave experiences multiple reflections (for instance, due to interactions between the pipe wall and the coal dust deposition layer). This results in variations in the arrival times of different reflected waves and the formation of superimposed interference, which leads to stepped fluctuations in the pressure curve. (2) Uneven distribution of coal dust: Following the turn, the coal dust in the downstream section is characterized by a non-uniform suspension state due to turbulence. Sudden increases in coal dust concentration in localized areas can cause intermittent intense combustion, which is manifested as brief peaks (step-like fluctuations) in the pressure curve. (3) Vortex separation effect: In the case of large angle bends (e.g., 150°), the airflow separates in the downstream section of the bend, resulting in the formation of periodic shedding vortices. These vortices can entrain unburned gas and coal dust into the downstream area, potentially causing local secondary explosions and further exacerbating the complexity of pressure fluctuations.

The mechanisms of pressure changes within pipelines attributed to reflection-induced superposition and combustion enhancement exhibit distinct characteristics. Reflection-induced superposition: The abrupt geometric change at the bend leads to the reflection of shock waves. The phase alignment between the incident and reflected waves results in instantaneous pressure spikes. This phenomenon is dependent on the pipeline’s geometry and operates independently of combustion processes. Turbulence-enhanced combustion: The turbulence generated by the bend accelerates mixing and reaction rates, thereby contributing to a sustained increase in pressure over extended durations.

4. Conclusions

This study established a transparent organic glass pipeline test system with various turning angles to systematically investigate the pressure and flame propagation characteristics associated with gas–coal dust coupling explosions within bent pipelines. The following conclusions were drawn:

(1) The increase in the turning angle of the pipeline significantly enhances the propagation speed of explosive flames. As the turning angle increases from 30° to 150°, the maximum velocity of the flame front rises from 97.92 m/s to 361.28 m/s, while the total propagation time decreases from 71 ms to 56.5 ms. The airflow generates vortices in the turning region, which increases the surface area of the flame front and accelerates the mixing and combustion reaction rates of unburned gases. Additionally, the secondary dust effect of coal dust further amplifies the energy release process.

(2) The duration of the explosion flame significantly decreases as the turning angle increases. At the coal dust deposition point, the flame duration in the 30° turning pipeline is 240.5 ms, whereas it is reduced to 64.17 ms in the 150° turning pipeline. The intense turbulence in the turning region accelerates the consumption rate of reactants, leading to an overall reduction in flame duration. Moreover, the flame duration at the outlet of the pipeline is notably shorter (ranging from 2 to 18.5 ms), which indicates that increasing the turning angle can effectively mitigate the residual risk of flames at the end of the pipeline.

(3) The peak overpressure of the shock wave significantly increases with the turning angle. The maximum overpressure measured in the pipeline at 30° and 150° turns is 7.07 kPa and 88.40 kPa, respectively. The overpressure values in the front section of the turn are generally higher than those in the rear section, primarily due to the superposition effect of reflected waves resulting from geometric discontinuities and the enhancing effect of turbulence on combustion. Additionally, the secondary suspension of coal dust, along with turbulent mixing, further elevates the peak overpressure in the front section of the bend, whereas energy dissipation in the rear section of the bend results in intensified pressure attenuation.

(4) The pressure curve in the front section of the turn displays regular oscillations, whereas in the rear section of the turn, the pressure fluctuations exhibit a stepped and complex pattern due to the interference of multiple reflected waves, uneven distribution of coal dust, and vortex separation effects. Large angle turns can induce periodic vortex shedding, which may trigger local secondary explosions and exacerbate the instability of downstream pressure fluctuations.

(5) While the square pipes and small-scale designs utilized in this study cannot fully replicate the actual underground tunnel environment of coal mines, they do elucidate the core mechanism of geometric asymmetry during the initial stage of an explosion. In future work, numerical simulations and mesoscale experiments may be combined to quantify the effects of Reynolds number and size on propagation dynamics. Additionally, the selected concentration (7.5% CH₄ and 100 g/m³ coal dust) represents only a typical mixed scenario observed in mine accidents, and the conclusions drawn from this research have certain limitations. Further investigations will focus on the dependence of explosion dynamics on varying concentrations.

(6) Principles of safety design based on experimental data: In the design of underground coal mine tunnels, it is advisable to limit the turning angle to ≤90°, which strikes a balance between the risk of explosion propagation and engineering feasibility. Explosion suppression devices and pressure monitoring points should be prioritized for installation in the upstream vicinity of tunnel corners. When it is unavoidable to employ a corner angle greater than 120°, explosion-resistant lining plates should be installed in the corner area of the tunnel, and turbulence suppression grilles should be positioned downstream within a distance of 20 times the inner diameter of the tunnel.