GAS CONTROL IN TG MOTOR REGION IN LONGWALL GASSY MINES

A number of longwall mines suffer significant gas delays due to high methane gas concentration levels (i.e. above 1.0% statutory limit for methane gas level) near the tailgate (TG) motor area of the longwall face. Therefore, in addition to development of effective goaf gas drainage and management strategies for highly gassy mines, development of face control strategies to minimize methane gas concentration levels near the tailgate motor of the longwall face is highly critical to achieve high production rates in high capacity longwall mines. To address this issue, detailed CFD models of longwall face were developed and extensive modelling simulations were carried out to investigate the effectiveness of various control options on gas concentration distribution near the tailgate motor area. To investigate the effect of mining and ventilation parameters on effectiveness of various control strategies, simulations were carried out with CFD models of Mine A. The details and results of all these studies are presented in this paper.


INTRODUCTION
Goaf gas emissions have increased substantially over the years and are set to increase in the near future due to higher production rates, increased depth of mine and industry's trend towards wider and longer panels. In highly gassy mining conditions and in coal seams prone to spontaneous combustion, management of these two key issues is critical for successful operation of longwalls, particularly in longer panels. Theoretically, gas management should be such that the methane gas drainage rate from goaf holes should be equivalent to the goaf gas emissions rate. It's challenging and practically not possible develop such ideal gas management systems. Generally, the undrained goaf gas will be drained through the ventilation air and resulting in high methane gas levels in the TG region. When the concentrations reach 2%, the sensor loaded on the TG equipment's trips off the TG drive and hamper all the longwall operations.
CFD modelling has been used in the coal mining industry in a number of areas by various researchers. In the last two decades, CFD techniques were widely used to model gas and dust flows in longwall mines. Aziz et al (1993) used numerical techniques to understand the ventilation mechanisms, gas and dust distributions in coal mines. Balusu et al (2002Balusu et al ( , 2004 carried out numerical investigations for understanding goaf gas distribution in gassy mines and proposed various inertisation strategies for prevention of spontaneous combustion in those mines. Two-dimensional CFD models based on the theory of natural convection and heat transfer in porous media were used to study the flow and temperature fields in underground coal fires (Huang et al. 2001). National Institute for Occupational Safety and Health (NIOSH) and Mine Safety and Health Administration (MSHA) investigated the temperature characteristics of mine fire (Trevist et al. 2009). Coal dust explosions are one of the most significant hazards in underground mines. A series of comprehensive studies were conducted (Proud et al. 2015;Collecutt et al. 2009) using CFD simulations to model coal dust explosions and the effectiveness of active explosion barriers. Various other investigations simulating auxiliary ventilation layouts in rapid heading development (Monoley et al. 1998), mine fires and explosions (Brune and Saki, 2017), control of methane and spontaneous heating , spontaneous combustion and heating in longwall , and in other applications in mineral processing (Fletcher et al. 1995).
CFD techniques were used for investigating the gas management options near the TG region of the longwall top coal caving face (Tanguturi and Balusu, 2014). In this paper, the results of the research work carried out for controlling the gas in the TG regions using CFD techniques are presented. Initially, investigations were carried out for determining the gas concentration near the TG motor and later, various gas management strategies using curtains at various locations in the longwall were investigated numerically.

CFD MODEL DEVELOPMENT
The CFD models were developed based on the dimensions of the longwall panel at Mine A. Geometric coordinates from the working seam floor contours were used for developing the CFD models. The longwall panel width is 310 m and the roadway width on both maingate (MG) and tailgate (TG) sides of the face is 5.4 m. The longwall extraction thickness is 3.6 m, which also represents the face height in the model. The goaf height up to 80 m above the working seam and the floor strata down to 12 m below the working seam is included in the models. The layout of the CFD model is shown in Figure 1.
MG and TG cut-throughs of 5m in width and 75m in length have been incorporated into the CFD models, and these cut-throughs were spaced at 125m intervals along the panel in the models. One rear shaft of 2.1m diameter has also been included in the CFD model, which is located on the TG side at the start-up areas of the models. A number of goaf gas drainage holes of 0.3m diameter at different locations were incorporated into the micro model to investigate the effect of various goaf gas drainage strategies. Each individual goaf hole can be switched 'ON' and 'OFF' in various simulations, depending on the goaf gas drainage to be investigated. After development of the basic configuration for the CFD model, all the longwall face equipment was also incorporated into the model, based on the equipment drawings of the longwall shields, MG and TG shields, BSL, and shearer. The CFD models also contain details of the longwall face such as the BSL at the MG side of the panel, shearer located at the mid face, transitional shields at MG and TG sides of the face and the AFC motor drives at MG and TG areas etc., as shown in Figure 2.

BOUNDARY CONDITIONS-MINE A
Mine A uses U ventilation system for longwall panels. The ventilation layout and actual airflow quantities of Mine A are used in CFD simulations, as shown in Figure 3. In the CFD model, the velocity inlet boundary conditions were specified at the MG intake roadway (70 m³/s) and also at the TG intake roadways (40 m³/s). An outlet velocity condition was specified at the MG roadway (30 m³/s) and a free outflow condition was specified at the TG end for the total ventilation quantity of 80 m³/s leaving the face. The rear shaft which was incorporated as intake shaft was closed in the modelling simulations. The permeability distribution in the goaf area varies considerably at various locations depending on the stress distribution, and it ranged from 10e3 millidarcy (md) to 10e10 md in the goaf area and goes up to 10e12 md in the collapsed roadways just behind the face.

MATHEMATICAL MODEL
The instantaneous conservative equations i.e., continuity, Navier Stokes and the species equations were solved numerically using finite volume discretisation techniques (Patankar, 1980). These equations are solved in the goaf region where the flow is laminar. Note that the goaf region is treated as porous media with resistances varying in all the three directions.

Continuity equation for incompressible air flow
where subscript s represents properties of O2, CH4 and N2.
Turbulence Modelling The flow is assumed to be turbulent in the face region and the time averaged equations are solved. Two equation k-epsilon model was used to determine the eddy viscosity.

Time Averaged Governing Equations
Continuity Equation: Reynolds Averaged Navier Stokes Equation: where is the Reynolds stress tensor.

Turbulent Kinetic Energy-k equation:
where subscript j represents Einstein summation notation.
Turbulent dissipation-equation: where Cµ1 and Cµ2 are closure coefficient.
Reynolds Stress: where is the eddy viscosity and is Kronecker delta.
Eddy Viscosity: where is closure coefficient which is equal to 0.07.
The coupling between the velocity and pressure is done by SIMPLE algorithm.

Porous media model
The porous media model was used to simulate the flow though the porous goaf regions by the addition of a momentum source term to the standard fluid flow equations. The source term is composed of two parts: a viscous loss term (Darcy law), and an inertial loss term: | is the magnitude of the velocity and and are prescribed functions which are related to the distributions of permeability and inertial resistance coefficient of the longwall goaf. This momentum sink contributes to the pressure gradient in the porous cell, creating a pressure drop that is proportional to the fluid velocity in the cell. In the CFD model, the incorporation of goaf spatial permeability distribution and gas emission are via user defined function (UDF) that is linked to solver.

MODEL VALIDATION
In the base case model simulations, actual geometry and operating conditions of the initial longwall panels at Mine A were used. Boundary conditions as discussed in section Figure 3 are used in the base case simulations. The total goaf gas emissions were around 3,000 l/s of methane, with total goaf gas drainage flow rates of around 3,300 l/s. The results of base case simulations along with field measured values are presented in Figure  4, showing comparison of oxygen and methane gas concentration levels at various locations behind the longwall face. Simulation results show that oxygen ingress is high on the maingate side, with oxygen concentration levels over 15% even at 500m behind the longwall face (Figure 4 (a). Results show high methane gas concentration levels on the tailgate side of the panel due to gas buoyancy effects. Comparison and analyses of the results presented in Figure 4a and Figure 4b shows that simulation results are in close agreement with the field measured data from Mine A.
Based on the results of extensive CFD modelling investigations carried out during the course of the project, it was identified that a new goaf gas drainage strategy involving additional goaf holes on MG side would contribute to a significant increase in goaf gas drainage rates in longwall panels. These optimum goaf gas drainage strategies were implemented at Mine A as field demonstration studies. The results of the field studies showing the methane gas concentration levels and flow rates in various goaf holes located on TG and MG sides, and at start-up areas are presented in Figure 5. Field results show that methane gas concentration levels varies between 37% and 68% in TG goaf gas drainage holes, 92% in MG goaf hole, and 80% in the start-up area goaf hole.
Analyses of the field data shows that the methane gas concentration levels at the mine site are in close agreement with the simulated values.

Figure 5
Field study results -Goaf holes performance data -Mine A

RESULTS AND DISCUSSIONS
Analyses of the goaf gas distribution patterns presented in the above section indicates that in addition to goaf gas drainage there is a need to develop additional control strategies to manage gas concentration levels at the TG motor area. Based on the above gas distribution patterns, it was identified that one of the practicable ways to manage TG goaf gas distribution is to modify airflow patterns near the TG motor area. To achieve the objective of reduced gas levels at the TG motor area, a number of control options involving various curtain configurations were investigated. The results of various investigations showing the effect of various control strategies on TG gas distribution patterns under Mine A conditions are presented in this section.
The location and configuration of various curtain options is presented in Figure 6. All the curtains are oriented across the face, i.e. perpendicular to the airflow direction. The first three configurations involve installation of different size curtains in the chock shields at approximately 2m to 5m upstream of the TG motor, as shown in Figure 7. The fourth configuration involves installation of additional curtain in the transition shield adjacent to the TG motor. In summary, the configuration 1 involves installation of curtain above AFC, as shown in Figure 6 (a); configuration 2 involves installation of a long curtain over AFC and walkway region, as shown in Figure 6 (b); configuration 3 involves installation of a curtain in front walkway region of the chock shield, as shown in Figure  6 (c); and the fourth configuration involves installation of a curtain in walkway region of transition shield adjacent to the TG motor, as shown in Figure 6 (d).
(a) Curtain above AFC (b)Large size curtain above half AFC and in the walkway region (c) Small size curtain in walkway region (d) Curtain in the walkway region of transitional shield Extensive modelling investigations were carried out with above individual curtain configurations and a combination of curtain configurations to investigate their effect on gas distribution patterns at the TG motor area. The results of a few key investigations are presented in the following sub-sections.
This control option 1 involves configuration 2 (Figure 6b), i.e. installation of a long curtain over AFC and walkway region of the chock shield at 2 -5m upstream of the TG motor. As discussed previously, U ventilation system is used in Mine A, with face quantity of around 90 m3/s. The results of the simulations showing the effect of this curtain configuration on air velocity and methane gas distribution patterns near the TG corner area are presented in Figure 7. The curtain configuration is also shown in Figure  7, as inset figure.
The gas distribution patterns presented in Figure 7a shows that the methane gas concentration level is around 0.4-0.7% at the TG motor area, with this curtain configuration. The methane gas concentration at the TG motor area is still close to the 1.0% statutory limit. The velocity distribution presented in Figure 7b shows that the installation of long curtain results in very high turbulence and recirculation zones around the curtain and around the TG motor area. Analyses of the results indicate that even minor disturbances in the panel may result in gas concentration levels rising over 1.0% at the TG motor area. The results indicate that this curtain configuration may not be a best control option as its effectiveness in controlling methane gas concentration levels at TG motor seems to be uncertain. The gas distribution patterns presented in Figure 8a shows that the methane gas concentration level is below 0.3% at the TG motor area, with this curtain configuration. The velocity distribution presented in Figure 8b shows that the installation of walkway curtain results in splitting of airflow and diversion of significant stream of airflow into the goaf region. Results indicate that this diversion of airstream into the goaf assists in (b) Velocity distribution near TG motor area (a) Methane gas distribution near TG motor moving/pushing the gas fringe away from the TG motor area. There was no recirculation zone formation around the curtain. Analyses of the results indicate that this curtain configuration would be effective in significantly reducing methane gas concentration levels around the TG motor region. However, results also show that gas fringe with gas concentration levels above 1.0% is close to the TG motor area even after installation of this curtain. It is to be noted here that optimal positioning of this curtain configuration is also important to control gas concentration levels near the TG motor. The curtain should be located at about 2 -5m upstream of the TG motor location.  Figure 9. The curtains configuration in two chock shields, i.e. in standard shield and in transition shield, is also shown in the inset figure.
The gas distribution patterns presented in Figure 9a shows that the methane gas concentration level is below 0.2% at the TG motor area, with this curtain configuration.
Methane gas distribution results also shows that high concentration gas fringe is moved away by a few meters (around 3 to 5m) from the TG motor location. The velocity distribution presented in Figure 9b shows that the installation of two walkway curtains (b) Velocity distribution near TG motor area (a) Methane gas distribution near TG motor in two chock shields results in splitting of airflow and diversion of significant stream of airflow into the goaf region. Results indicate that this diversion of airstream into the goaf assists in pushing/moving the gas fringe further away from the TG motor area. There was no recirculation zone formation around the curtain. Analyses of the results indicate that this curtains configuration would be very effective in significantly reducing methane gas concentration levels around the TG motor region. It is to be noted here that optimal positioning of these curtains configuration is also important to control gas concentration levels near the TG motor.
To investigate the effect of ventilation airflow quantity on the effectiveness of various face gas control strategies, additional simulations were carried out with reduced ventilation under Mine A conditions. In the original simulation, the face ventilation was around 90 m³/s. In the second simulation, the face ventilation quantity was reduced to 50 m³/s. In both the cases, U ventilation was employed during modelling simulations.  Figure 10. The curtain configuration is also shown in the inset figure. The gas distribution patterns presented in Figure 10a shows that the methane gas concentration level is below 0.3% at the TG motor area with this curtain (a) Methane gas distribution near TG motor (b) Velocity distribution near TG motor area configuration under high airflow conditions of 90 m³/s. The velocity distribution presented in Figure 10b shows that the installation of walkway curtain results in splitting of airflow and diversion of significant stream of airflow into the goaf region under high airflow conditions. Results indicate that this diversion of airstream into the goaf assists in moving/pushing the gas fringe away from the TG motor area. Analyses of the results indicate that this curtain configuration would be effective in significantly reducing methane gas concentration levels around the TG motor region under high face airflow conditions.
The results of the simulations under 50 m³/s airflow conditions showing the effect of the walkway curtain configuration on air velocity and methane gas distribution patterns near the TG corner area are presented in Figure 11. The curtain configuration is also shown in the inset figure. The gas distribution patterns presented in Figure 11a shows that the methane gas concentration level is around 0.6% to 0.8% at the TG motor area with this curtain configuration under low airflow conditions of 50 m³/s. The velocity distribution presented in Figure 11b shows that the installation of walkway curtain results in splitting of airflow and diversion of a low air velocity stream into the goaf region under low airflow conditions. Results indicate that this diversion of airstream into the goaf assists in moving/pushing the gas fringe a little bit away from the TG motor area. Analyses of the results indicate that this curtain configuration would not be very effective in significantly reducing methane gas concentration levels around the TG motor region under low face airflow conditions.
Analyses and comparison of the results indicate that the walkway curtain control strategy seems to be less effective in reducing methane gas concentration levels around the TG motor region under the reduced airflow conditions, when compared with its effectiveness under high airflow conditions. Analyses of the results also indicate that gas concentration levels around the TG motor area would be significantly higher under reduced face ventilation conditions, with TG motor area surrounded by 0.6% to 0.8% methane gas concentration levels.

CONCLUSIONS
A brief summary of the study findings and recommended face gas control strategies are as follows: • Results of Mine A simulations show that the gas concentration levels over the TG motor area would be around 0.7% -0.8%, and the gas fringe with gas concentration levels above 1.0% would be very close to the TG motor area.

•
Analyses of the results indicate that the control option involving a combination of curtain configurations would be successful in reducing the gas concentration levels near the TG motor area under Mine A conditions.

•
The optimum curtain configuration should include installation of a curtain in the walkway region of the chock shield at 2m upstream of the TG motor, and an additional curtain in the walkway region of the transition shields adjacent to the TG motor.
(b) Velocity distribution near TG motor area (a) Methane gas distribution near TG motor • Analyses of results indicate that the installation of these curtains would result in splitting of airflow and diversion of significant stream of airflow into the goaf region, and would assist in pushing the gas fringe further away from the TG motor area.

•
Analyses of the results indicate that the walkway curtain control strategy would be significantly less effective in reducing methane gas concentration levels around the TG motor region under high cutting height conditions, when compared with its effectiveness under low cutting height conditions.

•
Analyses of the results also indicate that the walkway curtain configuration would not be very effective in diverting significant airflow towards the goaf side, and in reducing the methane gas concentration levels around the TG motor under reduced airflow/ low air velocity conditions.
To address the issue of significant gas delays due to high methane gas concentration levels near the tailgate (TG) motor area of the longwall face, detailed CFD models of longwall face were developed for Mine A. Results of Mine A simulations show that the gas concentration level at the TG motor area is around 0.7%, and the gas fringe with gas concentration levels above 1.0% is very close to the TG motor area. Analyses of the results indicate that even minor disturbances in panel, such as change in barometric pressures or minor ventilation changes due to TG collapse and any reductions in face ventilation may result in gas concentration levels rising over 1.0% at the TG motor area.