Inﬂuence of the Auxiliary Air-Duct Outlet and the Brattice Location on the Methane Hazard—Numerical Simulations

: The article presents the results of research into the inﬂuence of the location of auxiliary ventilation devices on the distribution of methane concentrations at the outlet of the longwall in an underground mine. Since this area is crucial from the point of view of explosion risk, the existence of an optimal arrangement of these devices could lead to improved safety of the crew working in the area. The aim of conducted study was to examine if the impact of this devices placement is signiﬁcant. The research was carried out with the use of computational ﬂuid dynamics (CFD) modeling—Ansys Fluent. The analyses took into account the location of the two most commonly used devices: a brattice and an auxiliary air-duct. The numerical model has been prepared and validated based on in situ measurements. Thirty-two cases of device conﬁgurations were analysed. The length and position of the brattice, as well as the height and position air-duct outlet along tailgate, were modiﬁed. It has been shown that although the presented solutions are an effective risk mitigation method, contrary to the common opinion of many practitioners, the impact of their exact placement, provided it is compliant with the regulations, is not signiﬁcant for the registered methane concentration distribution at a longwall outlet.


Introduction
The methane hazard is common during the exploitation of energy resources. At the same time, taking into account the flammable and explosive properties of methane, this threat should be considered as one of the most important [1][2][3][4][5][6]. The uncontrolled increase in methane concentration in the workspace may lead to loss-generating downtime or to events that threaten the health and life of employees in the area where it occurs. To limit the possibility of activating potentially dangerous events, the ventilation services of mines are obliged to control and combat dangerous accumulations of this gas [7][8][9]: • Active methane prevention methods by ensuring adequate ventilation of the endangered area, methane drainage from the deposit and the use of auxiliary ventilation devices; • Passive methane prevention methods, including constant monitoring of the methane content in the air with the use of properly arranged sensors and automatic anemometers.
In relation to the specific character of hazards occurring in underground mining and their mutual interactions, it is necessary in many cases to conduct simulation studies.
Research published in recent years have successfully utilized the possibilities offered by computational fluid dynamics (CFD) to investigate and design this particularly hazardous regions [10].
These methods are particularly applied in the study of methane hazard. Many aspects related to the methods of limiting and controlling this hazard curently are the subject of research. In the field of methane drainage, Botao et al. [11] and Qin et al. [12] examined the effectiveness of selected drainage pressures on the return air methane concentrations and fire hazard, while Zapletal and Kosowski [13] analyzed selected methane collector diameters, Cao and Li [14] and Skotniczy [15] studied the distribution of methane concentrations in the mining-induced fracture zone in terms of its extraction and Wang et al. [16] researched methane concentration distribution in goafs and its interaction with the fire hazard. The research subject are also dynamic phenomena, such as methane explosion initiation and development [17].
Another area of studies are the ventilation parameters and hazard level with or without auxiliary ventilation equipment in working faces and longwall areas. Hasheminasab et al. [18] and Zhou et al. [19] studied the effect of selected auxiliary equipment on methane distribution in drilling faces, Liu et al. [2] proposed a way to reduce energy consumpion in these workings and Mishra et al. [20] present the results of a model study of methane distribution and dispersion in the tailgate. The effectiveness of CFD simulations in predicting methane concentration distributions in working faces was also subject of research by Daloglu et al. [21] and Kurnia et al. [22]. An important contribution are also the proposed methodologies for conducting model studies and validation with real measurements presented by Wierzbinski [23] for U-ventillated areas and by Janoszek et al. [24] for Y-ventillated longwall areas.
Among the indicated prevention measures, auxiliary ventilation devices are an extremely effective method of local limitation of hazardous methane accumulations at the longwall outlet. Their basic tasks are to dilute the air-methane mixture to the level of permissible concentrations and to direct the stream flowing from the longwall towards the part of the excavation being liquidated, thus moving the dangerous concentrations away from the working zone of the longwall and the crossing with the tailgate. The configuration of these devices should be adjusted to the recognized hazard level. The basic element is a brattice with a ventilation air-duct. If necessary, the system of auxiliary ventilation devices can be additionally extended with other elements, such as auxiliary air-ducts, jets and cyclones or directional vents [8,9,18,25].
Statistics provided in the literature [26] indicate the great popularity of these solutions. In over 70% of longwalls in the analyzed mines in the given period, auxiliary ventilation devices were used. Moreover, it should be noted that when the ventilation methane capacity exceeds 5 m 3 /min, the most widely used variant was the brattice and one auxiliary air-duct configuration. This variant is shown in Figure 1 covering the last section of the longwall with the location of the scraper chain conveyor with its drive. The brattice and the auxiliary air-duct are located in the tailgate near the intersection with longwall. The hatched area indicates the goaf zone. The red color indicates the fresh air path and the blue color indicates the return air path. The given ranges of methane sensor locations result from legal regulations. Legal regulations and guidelines in many cases do not indicate detailed recommendations, leaving the decision to the investor [9,[27][28][29] or include only general recommendations for their placement, indicating that the cross-sectional area between the brattice and the opposite side of the excavation cannot be smaller than 6 m 2 [30,31]. On the other hand, the detailed definition of the scope and parameters is the responsibility of the mining ventilation services, based on the opinion of experts, in such a way that "their location and operating parameters take into account the local conditions of methane hazard" [31]. In addition, the system should be made under the principles of consistency of the ventilation system and operational stability to effectively reduce the level of risk. In practice, there is a conviction among the employees of ventilation services that the appropriate arrangement of the brattice and the outlet of the air-duct in the tailgate geometry is of key importance for the effectiveness of this prevention method. This claim is unjustified because this subject has not been sufficiently researched, both in the in situ and model-based studies.
The subject of the conducted research was to verify the mentioned thesis to possibly indicate the optimal range of distribution. This article presents the results of simulation studies of the influence of selected variants of the distribution of auxiliary ventilation devices on the recorded methane concentrations. The most commonly used variant was analyzed ( Figure 1).

Methods
The Ansys Fluent software [32] was used in the research work to perform model tests of air flows and methane emissions at the intersection of the longwall with the tailgate. It is currently one of the most popular software environments for CFD calculations and simulations [2,11,14,16,18,23].
The model used in the research was prepared based on an actual longwall carried out in a underground coal mine, in which a high methane hazard was determined. The longwall was ventilated in a U system along the coal seam, i.e., the maingate supplementing the fresh air and the tailgate as a return air way ( Figure 2). The average methane content in this part of coal seam was 15.1 m 3 CH 4 /Mg. The selection of the most advantageous mesh was based on the evaluation of different mesh quality ( Table 1). The authors adopted a tetrahedral grid. The quality parameter values of the adopted mesh are 0.20834 (good) for orthogonal quality and 0.79166 (good) for skewness [32]. The adopted simplifications of the model in relation to the real conditions did not significantly affect the quality of their reproduction, which was confirmed by experimentation and validation of the results based on the actual data.   The primary solver settings are summarized in Table 2. Based on the measurements and previous experience [35][36][37], the boundary conditions of the model were established. The inflow to the model is indicated in the following planes: Due to the short section of the modeled longwall, as a result of the tests performed, it was decided to omit the parameter of methane inflow from the face of the exposed seam. The estimated value of this inflow has been included in the inlet_longwall boundary condition.
The model calculations made for the geometric configuration and the longwall ventilation parameters obtained as a result of measurements are shown in Figure 5.  The plane located in the tailgate behind the ventilation partition (DD ) was omitted at this stage because the high heterogeneity of the methane-air mixture [37] in this location, which may result in high errors.
The obtained error values for the selected three time periods in the final model were acceptably low. They are presented in Table 3. To determine the impact of auxiliary ventilation devices placement in the selected configuration variant for the distribution of methane concentrations at the longwall outlet, a model study of thirty-two selected variants was carried out (Table 4). Four selected parameters were modified, while the geometric arrangement of auxiliary ventilation devices constituting the reference point was given above in the assumptions of the model:   Table 4 presents the results of simulated methane concentrations at selected measurement points for the analyzed variants of the location of auxiliary ventilation devices. Indicated results were given for the four analyzed locations of the actual methane sensors (A-D) along the result planes shown on Figure 6. The results are given for two heights in the axis of the excavation (max-under the roof and ½-half of the excavation height).

Discussion
Firstly, attention should be paid to the slight variability of the results received at the location of the sensors in the longwall above the conveyor drive (sensor A), in the tailgate opposite to the longwall outlet (sensor C) and behind the brattice (sensor D). The obtained results, depending on the variant used, are not regular, and the fluctuations of their values does not exceed the previously determined (Table 1) maximum absolute error of mapping the actual measurements. This confirms a slight influence, contrary to common opinion, of the changes in the geometric parameters of selected for the analysis ventilation devices on the distribution of methane concentrations at the longwall outlet. The influence on this distribution should be sought in terms of the selection and number of auxiliary ventilation devices applied or the possibility of potential failure states occurrence, rather than their geometric arrangement (provided that they were properly placed regarding the mining practice).
The only significant variation in the simulated methane concentrations was recorded on the result plane in the location of the methane sensor B, i.e., in the so-called mixing zone, near the tailgate caving line (Figure 7). The minimum and maximum values of the simulated concentrations in the set are bolded in Table 5. This is the result of moving the end of brattice towards the liquidated part of the tailgate, which is also associated with the shift of the mixing zone itself. Regrettably, such action may cause an increased outflow of methane from the goafs in the final section of the longwall from behind the powered support section, due to the increased aerodynamic drag in the tailgate. Result planes for selected variants are presented on Figures 7 and 8. Cross-sections of the remaining variants due to their number are available in the external repository Supplementary Materials.
Despite the lack of research on the impact of the auxiliary devices' location in the tailgate on the methane concentration distribution, the obtained results are consistent with suspected distribution. The general distribution coincides with the previously conducted in situ tests performed in a similar working equipped with a brattice and air-duct [37]. The greatest accumulation of methane occurs in the zones where methane sensors are usually located, i.e., under the ceiling in the mixing zone and on the tailgate wall opposite to the longwall outlet. It is also consistent with the results of in situ and model studies on methane concentration distribution in tailgates, in which no auxiliary ventilation devices were used [20,23].  However, it should be recognized that the model studies carried out are subject to certain simplifications both from the perspective of equipment and workings geometry and initial conditions. In particular, it concerns the goaf section, where due to its nature it is not possible to perform detailed measurements of methane concentration distribution. Hence, it is necessary to rely on theoretical distributions. In spite of these limitations, as shown by other research carried out in the longwall areas, the obtained results are characterised by a sufficient accuracy and make it possible to observe phenomena that are interesting for research purposes and further application. The methane hazard itself may also be highly dynamic, and with this in mind, the authors of this publication have conducted research for steady conditions in ventilationally stable periods.

Conclusions
The application of auxiliary ventilation devices, in particular a brattice and an auxiliary air-duct, is an undoubtedly effective method of mitigation the level of methane hazard in a longwall outlet. The cross-sections with simulated methane concentrations show a clear separation of dangerous methane accumulations from crucial working areas. These include, first of all, the outlet of the longwall excavation, by the sidewall, where the driver of the scraper conveyor is located. Nevertheless, the subject of this study was to verify the thesis that the arrangement of these elements in the crossing geometry had a significant impact on the effectiveness of methane hazard prevention. The conducted analyzes clearly show that, despite the common opinions, this thesis cannot be confirmed. The greater effectiveness of methane hazard prevention may be potentially achieved by supplementing the devices configuration with additional elements, thus increasing the aerodynamic drag of the longwall working at its end section. Such actions, however, require a more extensive examination, as too high resistance at longwall outlet cause undesirable migration of methane to the working in its earlier section. A slight undesirable influence of the parameters and the arrangement was demonstrated only in the configurations described M_9, M_12, M_13 and M_16. This variants concerned the extension of the brattice without changing its location or moving it away from the tailgate caving line without changing the position of the air-duct outlet in the working axis at the same time. Such actions reduced the effectiveness of the auxiliary air-duct operation. It should also be pointed out that the models M_11 and M_15, indicating a slight decrease in the simulated concentrations, may, however, be unfavourable from the point of view of fire prevention. Increasing the airflow path and shifting the mixing zone further towards the tailgate caving line may cause re-migration of air to the goaf and the risk of endogenous fire.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/en15103672/s1, Figure S1: Methane concentration distribution at the AA plane: configuration M_1; Figure S2: Methane concentration distribution at the AA plane: configuration M_2; Figure S3: Methane concentration distribution at the AA plane: configuration M_3; Figure S4: Methane concentration distribution at the AA plane: configuration M_4; Figure S5: Methane concentration distribution at the AA plane: configuration M_5; Figure S6: Methane concentration distribution at the AA plane: configuration M_6; Figure S7: Methane concentration distribution at the AA plane: configuration M_7; Figure S8: Methane concentration distribution at the AA plane: configuration M_8; Figure S9: Methane concentration distribution at the AA plane: configuration M_17; Figure S10: Methane concentration distribution at the AA plane: configuration M_18; Figure S11: Methane concentration distribution at the AA plane: configuration M_19; Figure S12: Methane concentration distribution at the AA plane: configuration M_20; Figure S13: Methane concentration distribution at the AA plane: configuration M_21; Figure S14: Methane concentration distribution at the AA plane: configuration M_22; Figure S15: Methane concentration distribution at the AA plane: configuration M_23; Figure S16: Methane concentration distribution at the AA plane: configuration M_24; Figure S17: Methane concentration distribution at the AA plane: configuration M_25; Figure S18: Methane concentration distribution at the AA plane: configuration M_26; Figure S19: Methane concentration distribution at the AA plane: configuration M_27; Figure S20: Methane concentration distribution at the AA plane: configuration M_28; Figure S21: Methane concentration distribution at the AA plane: configuration M_29; Figure S22: Methane concentration distribution at the AA plane: configuration M_30; Figure S23: Methane concentration distribution at the AA plane: configuration M_31; Figure S24: Methane concentration distribution at the AA plane: configuration M_32; Figure S25: Methane concentration at selected result planes: configuration M_1; Figure S26: Methane concentration at selected result planes: configuration M_2; Figure S27: Methane concentration at selected result planes: configuration M_3; Figure S28: Methane concentration at selected result planes: configuration M_4; Figure S29: Methane concentration at selected result planes: configuration M_5; Figure S30: Methane concentration at selected result planes: configuration M_6; Figure S31: Methane concentration at selected result planes: configuration M_7; Figure S32: Methane concentration at selected result planes: configuration M_8; Figure S33: Methane concentration at selected result planes: configuration M_17; Figure S34: Methane concentration at selected result planes: configuration M_18; Figure S35: Methane concentration at selected result planes: configuration M_19; Figure S36: Methane concentration at selected result planes: configuration M_20; Figure S37: Methane concentration at selected result planes: configuration M_21; Figure S38: Methane concentration at selected result planes: configuration M_22; Figure S39: Methane concentration at selected result planes: configuration M_23; Figure S40: Methane concentration at selected result planes: configuration M_24; Figure S41: Methane concentration at selected result planes: configuration M_25; Figure S42: Methane concentration at selected result planes: configuration M_26; Figure S43: Methane concentration at selected result planes: configuration M_27; Figure S44: Methane concentration at selected result planes: configuration M_28; Figure S45: Methane concentration at selected result planes: configuration M_29; Figure S46: Methane concentration at selected result planes: configuration M_30; Figure S47: Methane concentration at selected result planes: configuration M_31; Figure S48 writing-review and editing, A.P.N., G.P., Z.R., P.W. and D.M.; visualization, A.P.N., G.P., Z.R., P.W. and D.M. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by the Department of Mining, Silesian University of Technology, grant number 06/050/BKM18/0061.