1. INTRODUCTION

Coal resources have been the most important fossil energy and will remain so in the foreseeable future in China, where coal, as the primary energy source, has reached 70% contribution rate [1,2]. Presently in China, over 95% of coal resources are extracted from underground coal mines [3]. However, the underground coal mine is one of the most dangerous working environments, e.g., dust particles generated during coal mining activity can lead to serious occupational health risks to coal workers [4,5]. The tunneling face in an underground coal mine is one of the most serious dust-generating locations where most particulate matters are mainly rock dust, which is of high quart content [6]. Epidemiological studies have shown that exposure to rock dust (especially respirable dust whose diameter is less than 10 μm) and/or crystalline silica is the main reason for pneumoconiosis with initiation and progression of pulmonary fibrosis, which are closely linked to dust concentration, dust diameter distribution, and its composition they are subjected to [7,8]. Thus, it is necessary to continuously evaluate rock dust pollution, especially dust concentration in the tunneling face for the health of coal miners.

Currently, two main approaches have been adopted to evaluate the dust-pollution level in an underground tunneling face: field measurement and portable instantaneous monitoring [9,10]. The field measurement method mainly depends on dust sampling and sample analysis, which is widely used in underground coal mines for its high accuracy. However, the field measurement method cannot meet the demand of real-time monitoring and sampling, and the data obtained are discrete and discontinuous in the whole tunneling face. Meanwhile, some portable instantaneous monitoring equipment and methods are used to evaluate dust pollution, which are mainly based on the absorption and scattering properties of a dust particle system without considering the absorption of gases in an underground mine. Portable monitoring equipment and methods mainly consider the dust particle system to be homogeneous and uniform in its detection space [11]; however, the dust concentration and diameter distribution are nonuniform, which varies obviously, and the absorption of some gases (e.g., ${\mathrm{CO}}_2$, ${\mathrm{H}}_2\mathrm{O}$) cannot be ignored for their high content [12]. Meanwhile, the monitoring of dust particles through portable monitoring equipment is discrete and discontinuous. The current method cannot achieve the continuous real-time monitoring of rock dust particles in a tunneling face. With the development of computer technology, computer numerical simulation, in particular, the computational fluid dynamics (CFD) method is widely used to simulate and predict the dispersion and spatial distribution of dust particles in an underground coal mine, which has been proven to be consistent and agrees well with the field measurement data [13–15]. Thus, the CFD simulation method can be applied to obtain the forward spatial distribution of dust particles.

In recent years, the real-time monitoring technology of atmospheric particulate matter (PM) in an open atmosphere has been developed and widely used all over the world [16,17]. The radiative transfer properties of PM, especially aerosol optical depth, can be obtained by various satellite and ground-based optical remote-sensing detection technologies; thus, combined with certain retrieval methods, the PM pollution level in an open atmosphere can be evaluated [18–20]. Such optical monitoring method can also be modified and optimized to be used in the tunneling face for rock dust monitoring. Actually, the optical absorption of some gases (mainly ${\mathrm{CO}}_2$, ${\mathrm{H}}_2\mathrm{O}$) is quite strong under certain wavebands due to the gases’ high contents in an underground atmosphere whose absorption effect should be removed to obtain the pure optical signals of rock dust [21,22]. Dust pollution (mainly dust concentration) in an underground tunneling face is hundreds of times larger than that of near-ground atmosphere, resulting in possible invalidation of the optical signal under certain wavebands. The detection position and waveband should be further analyzed and determined considering the high rock dust pollution in an underground tunneling face.

In our study, the spatial distribution of rock dust was simulated and obtained by the computational fluid dynamics (CFD) method, combined with the complex refractive index of rock dust in our previous study [23], we report and analyze the spectral transmission properties of a dust-polluted mine atmosphere within $400–4000{\mathrm{cm}}^{-1}$. By removing the optical background of mine gases, the optical signals of rock dust were obtained. The determination of detection position and detection waveband selection for rock dust monitoring was conducted to avoid the elimination of the optical signal under high-concentration rock dust. Finally, two optical channels were selected, and an evaluation method on the concentration and diameter of rock dust was put forward.

2. SPATIAL DISTRIBUTION OF ROCK DUST

The spatial distribution of rock dust in a whole tunneling roadway is the basis for simulating the optical properties of rock dust. The data obtained by field measurement method and portable instantaneous monitoring are discrete and limited to certain sampling points, which cannot satisfy the needs of our study. Then, the CFD method was adopted in this study to simulate the spatial distribution of rock dust in the tunneling face, which has been validated to be effective and acceptable in previous studies [13,14].

A. CFD Method and Parameter Configuration

A 3D steady incompressible Navier–Stokes equation was adopted to describe the airflow in a mine’s atmosphere. The $k-\epsilon$ model of Reynolds was applied to describe the turbulence effect, ignoring heat transfer. Then, a discrete phase model was adopted to take dust movement into account, thus tracking dust movement by solving a differential equation for a discrete second phase in Lagrangian time. The second phases are spherical particles, which are dispersed in a continuous phase. This model considers the coupling effect between the phases and its impact on the discrete phase trajectories and the continuous phase flow.

Because of the complex working equipment in an underground tunneling face, including the road header, supporting equipment, air duct, and cable trays, it is impossible to establish an exact geometric model. Thus, the geometric model was simplified to be a cuboid calculation area of $100\mathrm{m}\times 4\mathrm{m}\times 3\mathrm{m}$, a road header of $5\mathrm{m}\times 2\mathrm{m}\times 1.4\mathrm{m}$, an air duct of cylinders with the length of 93 m and diameter of 0.3 m. Thus, the simplified geometric model of a tunneling face is depicted in Fig. 1 in which only 20 m of the whole length is shown. Meanwhile, six sampling lines (lines 1, 2, 3, 4, 5, and 6) along the tunneling face were also marked and illustrated in the simplified model, which was used in further analysis.

The main parameters and boundary conditions of the numerical simulation were determined as follows: the dust particles in the simulation were rock dust with a density of $2500\mathrm{kg}/{\mathrm{m}}^3$, which conforms to the diameter distribution of Rosin–Rammler with a flow rate of 0.01 kg/s. The inlet boundary type of an air duct was set to be a velocity inlet with a magnitude of 17 m/s; the outlet boundary type was set to be outflow.

 

Fig. 1. Schematic of underground tunneling face: (a) side view, (b) front view, (c) top view.

Download Full Size | PDF

B. CFD Results

Based on the established geometric model in combination with the CFD method, the numerical simulation on the dispersion and distribution of rock dust in a tunneling face was carried out. The 3D simulation results of dust distribution are shown in Fig. 2.

As shown in Fig. 2, rock dust particles disperse outward mainly under the effect of airflow. The turbulence of airflow is quite strong near the tunneling face, thus causing the spatial distribution of rock dust to be inordinate near the tunneling face. A vortex flow area exists around the outlet of the air duct. Then, the dust dispersion gradually remains stable and regular with the increasing dispersion distance.

 

Fig. 2. 3D cloud chart of the spatial distribution of rock dust: (a) XZ face, (b) YZ face, (c) XY face.

Download Full Size | PDF

The air flows out from the air duct, which is placed on the right side of tunneling face, thus causing the rock dust particles generated in the tunneling face to disperse mainly on the left side of the tunneling face, as shown in Fig. 2(a). Meanwhile, rock dust particles gradually settle down under the effect of gravity, which means that dust pollution near the ground of the tunneling face is more serious than that of the roof, as presented in Figs. 2(b) and 2(c). Thus, the dust pollution on the left side near the ground of the tunneling face is the most serious after a long dispersion of rock dust.

The dust particles that greatly threaten coal miner’s health are mainly respirable dust whose diameter is less than 10 μm. The diameter distribution variations of rock dust along the tunneling face in different sampling lines (1, 2, 3, 4, 5, and 6) are almost the same. Thus, the distribution of rock dust concentration in the six lines and the respirable dust proportion in line 5 along the tunneling face are given in Fig. 3.

 

Fig. 3. Distribution of rock dust concentration and respirable dust proportion along the tunneling face.

Download Full Size | PDF

Figure 3 indicates that the concentration of rock dust in the tunneling face is first increased and then gradually reaches the maximum point, after which the dust concentration decreases slightly and remains stable after 70 m. The proportion of respirable rock dust, whose diameter is less than 10 μm, is obviously increased with the increasing dispersion distance. When the dispersion distance reaches 70 m, the increase trend gradually slows down and remains stable.

As can be seen in Figs. 2 and 3, the dust concentration near the ground in lines 3 and 4 is obviously larger than the other four lines, thus indicating that the optical signal is more likely to be eliminated under the attenuation and scattering effect of rock dust. Thus, such position is not appropriate for optical monitoring of rock dust, which is further analyzed and discussed in Section 3B.

3. SIGNAL PROCESS AND SPECTRAL ANALYSIS

The rock dust particles were sampled in the underground tunneling face and have been tested on its basic optical properties by Fourier Transform Infrared Spectroscopy (FT-IR) in our previous work. The refractive complex index of rock dust in the tunneling face was simulated and obtained as shown in Fig. 4 [23].

A. Removal of the Optical Background of Mine Atmosphere

The optical signal can be obviously attenuated and scattered by rock dust particles in the whole tunneling face. Apart from the attenuation and scattering of rock dust in the tunneling face, the infrared optical signal also can be absorbed by mine gases, mainly including ${\mathrm{CO}}_2$ and ${\mathrm{H}}_2\mathrm{O}$, because of their high volume content and absorption band. The strong absorption effect cannot be ignored because the content of ${\mathrm{CO}}_2$ and ${\mathrm{H}}_2\mathrm{O}$ is higher than that of open atmosphere, especially for ${\mathrm{H}}_2\mathrm{O}$, considering the extremely moist environment in an underground coal mine.

Based on the complex refractive index and the spatial distribution of rock dust, combined with the Mie scattering model, the attenuation coefficient of rock dust was obtained [24,25]. The line-by-line calculation was adopted to simulate the absorption coefficient of mine gases based on the airflow field in the tunneling face [26]. Then, considering the scattering of rock dust and absorption of mine gases, the radiative transfer model of participating air-dust media was established. Among the various numerical solution methods [27–29], the discrete ordinate method was solved on a 1D direction to obtain the transmission properties of a dust-polluted mine atmosphere [30]. The calculation process in this study is depicted in Fig. 5.

 

Fig. 4. Complex refractive index of rock dust.

Download Full Size | PDF

 

Fig. 5. Schematic of the calculation process.

Download Full Size | PDF

According to the field condition of the underground tunneling face, the temperature in the underground tunneling face was set to be 300 K, and the atmospheric pressure was set to be 103000 Pa because of the forced ventilation in the tunneling face. The volume concentrations of the main gases in the underground tunneling face were set as follows: ${\mathrm{N}}_2$ (76.1533%), ${\mathrm{O}}_2$ (20.1%), ${\mathrm{CO}}_2$ (0.04%), ${\mathrm{H}}_2\mathrm{O}$ (3.7067%). The spectral resolution and total number of spectral lines implemented in this simulation are $10{\mathrm{cm}}^{-1}$ and 360, respectively.

Considering the scattering of rock dust and the absorption of mine gases, the transmission properties of a dust-polluted mine atmosphere and the absorption spectrum of mine gases were calculated by Mie scattering model, line-by-line calculation, and the discrete ordinate method, as shown in Figs. 6 and 7.

 

Fig. 6. Transmittance of dust-polluted mine atmosphere.

Download Full Size | PDF

 

Fig. 7. Absorption spectrum of mine gases.

Download Full Size | PDF

Figure 6 presents that the transmittance of a dust-polluted mine atmosphere is relatively higher under the waveband of $400–900{\mathrm{cm}}^{-1}$ and $1300–2000{\mathrm{cm}}^{-1}$, thus indicating the weaker attenuation and absorption of rock dust and mine gases. The sharp volatility under the waveband of $1300–2000{\mathrm{cm}}^{-1}$ is mainly due to the absorption of mine gases. The absorption spectrum of mine gases under the waveband of $400–4000{\mathrm{cm}}^{-1}$ is shown in Fig. 7.

Three major absorption bands can be obviously observed in the wavenumber range of $400–4000{\mathrm{cm}}^{-1}$ in Fig. 7. The first absorption band ($1365–1835{\mathrm{cm}}^{-1}$) and the third absorption band ($3665–3935{\mathrm{cm}}^{-1}$) mainly reflect the absorption effect of ${\mathrm{H}}_2\mathrm{O}$, and the second absorption band under the wavenumber of $2315–2335{\mathrm{cm}}^{-1}$ is mainly the absorption of ${\mathrm{CO}}_2$. Other gases in a mine atmosphere (e.g., ${\mathrm{N}}_2$, ${\mathrm{O}}_2$) almost have no absorption within $400–4000{\mathrm{cm}}^{-1}$.

The optical signal obtained in a dust-polluted mine atmosphere is the combined effect of rock dust and mine gases. The transmittance of rock dust without the absorption of mine gas is the optical signal, which can be used to evaluate the dust pollution level. However, the pure optical transmittance of rock dust cannot be directly obtained in the field tunneling face. Thus, the optical background of mine gases must be removed first to obtain the accurate rock dust signal. The mine gases’ composition is quite stable in tunneling face under the normal driving advancement work without sudden-onset disaster (e.g., gas outburst, mine fire), the absorption properties and spectral transmittance of mine gases can be obtained based on the normal gases composition, local ambient temperature, and pressure. Therefore, the transmission signal of rock dust can be simulated from the optical transmittance of a dust-polluted mine atmosphere by removing the optical background of mine gases. The retrieved transmittance by removing optical background of gases can be obtained by the formula:

 

Fig. 8. Directly simulated and retrieved transmittance of rock dust.

Download Full Size | PDF

Figure 8 shows that the simulated and retrieved transmittance of rock dust agrees quite well both in variation and magnitude, which indicates that the interaction of rock dust and mine gases on the optical transmission is quite weak considering the field engineering application. Thus, by removing the optical background of mine gases, we can obtain an accurate, effective optical signal of rock dust. The transmission signal of rock dust was further used in the following analysis.

B. Determination of Detection Line

The dust pollution in an underground tunneling face is quite serious, which is hundreds of times larger than that of open atmosphere. The strong attenuation and scattering of rock dust can eliminate the optical signal under certain wavebands. Because the distribution of rock dust is quite different under a different sampling line, it is necessary to compare and analyze the transmission properties of rock dust under different observation position. Six detection lines, as can be seen in Fig. 1, were selected considering the field condition to determine the optimal detection position for rock dust pollution evaluation.

Figure 9 presents that the transmittance of rock dust under detection lines 3 and 4 are relatively weaker whose optical signal are more likely to be invalid, which is consistent with the results in Figs. 2 and 3. Further comparison of the remained detection lines (lines 1, 2, 5, and 6) is shown in Fig. 10.

 

Fig. 9. Detection line comparison in vertical direction: (a) line 1, line 4, and line 5; (b) line 2, line 3, and line 6.

Download Full Size | PDF

 

Fig. 10. Detection line comparison in horizontal direction: (a) line 1 and line 2; (b) line 5 and line 6.

Download Full Size | PDF

The comparison of the remained lines in horizontal direction indicates that detection lines 1 and 5 have better transmission properties of rock dust than lines 2 and 6. This is due to the fact that most dust particles are below left lines 2 and 6 because the air duct is placed on the right side. Meanwhile, line 5, which is 1.5 m high, is at the breathing level of coal workers, which can more scientifically represent the threat of dust particles. Considering the difficulty of detection equipment installation and the working environment of a tunneling face, detection line 5 is eventually determined to be the optimal detection position in this study.

C. Waveband Selection

Based on the optimal detection position, further analysis was carried out to select the best detection waveband under the waveband of $400–4000{\mathrm{cm}}^{-1}$. The transmittance of rock dust under different detection lines are shown in Section 3B where the dust distribution varies obviously, as shown in Fig. 2. Then, the impact of dust diameter on the transmittance was also calculated and analyzed. Adopting the dust concentration distribution of detection 5, simplify the dust diameter to 2, 5, 10, 12, 15, and 20 μm, respectively, to simulate the transmittance of rock dust. To study the influence of diameter, a dimensionless parameter ${\gamma }_{\text{ratio}}$ was put forward as follows:

Because most larger particles settle down soon under the effect of gravity during the dispersion of mixture rock dust. The dust particles suspended in the tunneling face are mainly respirable particles, which are consistent with our CFD simulation results in Section 2. Meanwhile, the dust particles that greatly threaten a coal miner’s health are respirable dust whose diameter is less than 10 μm; larger particles cannot be breathed in because of filtration of the nasal cavity. Thus, six kinds of uniform dust particles with the diameter of 2, 5, 10, 12, 15, and 20 μm were adopted in our study as the comparison group.

The ${\gamma }_{\text{ratio}}$ of rock dust under different uniform diameters in Fig. 11(a) indicates that the transmittance obviously and strongly varies with the variation of dust diameter except the “singular” point of $1515{\mathrm{cm}}^{-1}$ and around. The ${\gamma }_{\text{ratio}}$ under the “singular” point and around approaches zero, thus indicating the transmittance is almost unaffected by the variation of dust diameter, especially when the diameter is less than 20 μm. A partial enlarged diagram was presented in order to analyze the “singular” point in more detail, as can be seen in Fig. 11(b).

 

Fig. 11. Influence of dust diameter on transmittance.

Download Full Size | PDF

The partial enlarged diagram shows that the ${\gamma }_{\text{ratio}}$ of different diameters gradually approach zero when the wavenumber comes to $1500{\mathrm{cm}}^{-1}$ and then move away when it is over $1530{\mathrm{cm}}^{-1}$. The most optimal wavenumber for ${\gamma }_{\text{ratio}}$ approaching zero is at $1515{\mathrm{cm}}^{-1}$. By adopting the optimal optical channel $1505–1525{\mathrm{cm}}^{-1}$, we quantify the ${\gamma }_{\text{ratio}}$ of different diameters, as shown in Table 1.

Table 1. Quantification of the ${\mathsf{\gamma }}_{\mathsf{\text{ratio}}}$ within $\mathsf{1505}\mathsf{–}\mathsf{1525}{\mathsf{cm}}^{-\mathsf{1}}$

View Table | View all tables in this article

As can be seen in Table 1, the ${\gamma }_{\text{ratio}}$ within the optical channel of $1505–1525{\mathrm{cm}}^{-1}$ (1505, 1515, and $1525{\mathrm{cm}}^{-1}$) are almost all within $-0.1\sim 0.1$ especially for the dust particle whose diameter is less than 20 μm. When the diameter reaches 20 μm, the ratio would gradually increase. Considering that the dust particles dispersed in the tunneling face are mainly respirable dust whose diameter is less than 10 μm, the optical channel of $1505–1525{\mathrm{cm}}^{-1}$ can be used for the evaluation of dust concentration for its dullness of dust diameter variation.

The variation of dust concentration can be obtained by the analysis within $1505–1525{\mathrm{cm}}^{-1}$. Then, by utilizing another optical channel, which is sensitive to dust concentration and dust diameter, we can obtain the variation of dust diameter. The first optical channel for determining dust concentration is fixed as $1505–1525{\mathrm{cm}}^{-1}$ in this study, while the second channel is not constant within $400–4000{\mathrm{cm}}^{-1}$, as long as the transmittance signal is quite sensitive to the variation of dust concentration and dust diameter under this channel. According to such selecting criteria, several appropriate channels could be found such as $2245–2265{\mathrm{cm}}^{-1}$, $3025–3045{\mathrm{cm}}^{-1}$, $3505–3525{\mathrm{cm}}^{-1}$. Here, an optical channel within $2245–2265{\mathrm{cm}}^{-1}$ is taken as an example to further introduce the dual-band method. Such dual-band method, by using two optical channels ($1505–1525{\mathrm{cm}}^{-1}$ and $2245–2265{\mathrm{cm}}^{-1}$), can be considered for evaluation of rock dust pollution level in tunneling face. Further analysis is conducted in the discussion of Section 4.

4. DISCUSSION

The detection position determination and spectral analysis have been conducted in Section 3, and the two optical channels ($1505–1525{\mathrm{cm}}^{-1}$ and $2245–2265{\mathrm{cm}}^{-1}$) were selected based on the spectral analysis; thus, the dust pollution evaluation method was given and analyzed based on the obtained optical signal in this section.

The optical channel of $1505–1525{\mathrm{cm}}^{-1}$ is almost only sensitive to the variation of dust concentration; thus, further analysis on the relation of dust concentration and optical signal was conducted. Aiming at the respirable dust dispersed in the tunneling face, taking the attenuation rate as optical signal, two baselines of uniform particles with different mass concentration (2 and 10 μm) were simulated and given to show the correlation between dust concentration and attenuation rate within the optical channel of $1505–1525{\mathrm{cm}}^{-1}$. Detection lines 1, 2, 3, 4, and 6 were adopted to calculate and analyze the relation between dust concentration and attenuation rate based on their spatial distribution of rock dust and by doubling and tripling their dust concentration while keep the diameter distribution unchanged. The attenuation rate is an equivalent integral result of the whole optical channel ($1505–1525{\mathrm{cm}}^{-1}$), as shown below:

 

Fig. 12. Correlation of dust concentration and attenuation rate.

Download Full Size | PDF

Based on the correlation of dust concentration and attenuation rate for the comparison group and the two baselines (2 and 10 μm), further quantitative analysis was conducted to compare the relative errors on dust concentrations between comparison group and the two baselines at the same attenuation rate, as shown in Table 2. It presents that all the relative errors (ignored plus or minus signs) are within 0.04, which can be acceptable in engineering application. Figure 12 and Table 2 demonstrate that the variations of the comparison group (the field detection lines, which are of nonuniform diameter distributions) are consistent and agree well with the two baselines for uniform particles (2 and 10 μm). The attenuation rate is almost the only function of dust concentration and is hardly influenced by the variation of dust diameter. Thus, the optical channel of $1505–1525{\mathrm{cm}}^{-1}$ proved to be effective in evaluating the dust concentration in the tunneling face based on the attenuation rate in the tunneling face.

Table 2. Quantitative Analysis of the Relative Errors Between Comparison Group and Baselines

View Table | View all tables in this article

Based on the obtained dust concentration, the evaluation method of dust diameter distribution was subsequently discussed. The spectral attenuation coefficient of rock dust at every sampling point can be retrieved from the spectral transmittance. By adopting the second optical channel of $2245–2265{\mathrm{cm}}^{-1}$, the equivalent integral attenuation coefficient was calculated by

The variation of attenuation coefficient, dust concentration, and respirable dust proportion along the tunneling face was calculated and depicted in Fig. 13 in order to analyze the influence of dust concentration and diameter on the optical signal of attenuation coefficient.

Figure 13 indicates that the attenuation coefficient is mainly influenced by dust concentration and linked to dust diameter distribution as well. The attenuation coefficient of rock dust under $2245–2265{\mathrm{cm}}^{-1}$ at every sampling point can be obtained from the transmittance of rock dust in the tunneling face; the dust concentration at every sampling point has been determined through the optical channel of $1505–1525{\mathrm{cm}}^{-1}$; thus, the influence level of dust concentration on attenuation coefficient within $2245–2265{\mathrm{cm}}^{-1}$ can be known; by assuming the dust diameter distribution model, the diameter distribution of rock dust at every sampling point can be retrieved.

 

Fig. 13. Variation of attenuation coefficient and dust distribution.

Download Full Size | PDF

Assuming the dust concentration in the tunneling face were all the same (the average dust concentration in detection line 5 of $42.7\mathrm{mg}/{\mathrm{m}}^3$), aimed at the variation of dust diameter, a forward simulation was carried out to obtain the influence of diameter distribution on the attenuation coefficient based on the same dust concentration, as shown in Fig. 14.

 

Fig. 14. Variation of dust diameter on attenuation coefficient.

Download Full Size | PDF

As can be seen in Fig. 14, the attenuation coefficient is obviously linked to the diameter distribution of rock dust based on the same dust concentration. The proportion of respirable dust, whose diameter is less than 10 μm, was obviously increased under the effect of gravity with the increase of dispersion distance causing the increase of attenuation coefficient. The retrieval method of diameter distribution can be feasible when the attenuation coefficient and dust concentration at every sampling point have been determined.

As the main method presented in this study, how the dual-band method works is further illustrated and summarized in the flow chart in Fig. 15, which presents that, based on the measured transmittance, the dust concentration distribution can be determined by using the optical information of the first channel within $1505–1525{\mathrm{cm}}^{-1}$. Combined with the complex refractive index and the assumed dust diameter distribution-based CFD results, a Mie scattering model and discrete ordinate method are used to calculate the simulated transmittance within the second optical channel ($2245–2265{\mathrm{cm}}^{-1}$.). By comparing the simulated and measured transmittance, the satisfying dust diameter distribution can be obtained if the error is less than the given threshold. Otherwise, improve the parameters of diameter distribution and redo the calculation.

 

Fig. 15. Flow chart of dual-band detection method.

Download Full Size | PDF

5. CONCLUSION

The CFD method, which is based on a two-phase fluid model, was applied in this study to simulate the spatial distribution of rock dust in the tunneling face of underground coal mine. Based on the complex refractive index in our previous study, we simulated the transmission properties of a dust-polluted mine atmosphere within the spectral range of $400–4000{\mathrm{cm}}^{-1}$ by the Mie scattering model, line-by-line calculation, and discrete ordinate method. Aimed at the evaluation of rock dust pollution, the optical background of mine gases was removed to obtain the pure optical signal of a rock dust system in order to determine the optimal detection position and waveband in this study.

Six detection positions along the tunneling face were selected and compared on their transmission properties to determine the best detection line. Avoiding the elimination of the optical signal, considering the convenience of detection equipment installation, detection line 5 along the tunneling face, which is 1.5 m high and 0.3 away from the right wall, was determined to be the optimal detection position. Furthermore, to obtain a valid optical signal, for optical evaluation of rock dust pollution, waveband selection was also conducted. Spectral analysis indicates that an optical channel ($1505–1525{\mathrm{cm}}^{-1}$) was proved to be sensitive almost only to the variation of dust concentration and dull to the change of dust diameter especially for respirable dust. Such optical channel of $1505–1525{\mathrm{cm}}^{-1}$ was used for the evaluation of dust concentration in tunneling face. The dust concentration can be obtained by analyzing the transmission properties with the optical channel ($1505–1525{\mathrm{cm}}^{-1}$), another optical channel of $2245–2265{\mathrm{cm}}^{-1}$, which is sensitive to both dust concentration and dust diameter, can be selected for the evaluation of dust diameter distribution. Such dual-band method, by adopting $1505–1525{\mathrm{cm}}^{-1}$ and $2245–2265{\mathrm{cm}}^{-1}$, can be used for the comprehensive evaluation of rock dust pollution in the tunneling face of a coal mine.