Tunnelling and Underground Space Technology 42 (2014) 206–215

Contents lists available at ScienceDirect

Tunnelling and Underground Space Technology

journal homepage: www.elsevier .com/ locate / tust

Trenchless Technology Research

Simulation of a novel intermittent ventilation system for undergroundmines

http://dx.doi.org/10.1016/j.tust.2014.03.0090886-7798/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author at: Department of Mining and Materials Engineering,McGill University, 3450 University Street, Frank Dawson Adams Bldg, Room 115,Montreal, QC, H3A2A7, Canada. Tel.: +1 514 398 3788; fax: +1 514 398 7099.

E-mail addresses: kurnia.jc@gmail.com (J.C. Kurnia), agus.sasmito@mcgill.ca(A.P. Sasmito).

Jundika C. Kurnia a,b, Agus P. Sasmito a,c,⇑, Arun S. Mujumdar a,d

a Minerals Metals and Material Technology Centre, National University of Singapore, Engineering Drive 1, Singapore 117576, Singaporeb Mechanical Engineering, Masdar Institute of Science and Technology, Masdar City, P.O. Box 54224, Abu Dhabi, United Arab Emiratesc Department of Mining and Materials Engineering, McGill University, 3450 University Street, Frank Dawson Adams Bldg, Room 115, Montreal, QC H3A2A7, Canadad Department of Bioresource Engineering, McGill University, 111 Lakeshore Road Sainte-Anne-de-Belleveu, Quebec H9X3V9, Canada

a r t i c l e i n f o a b s t r a c t

Article history:Received 16 October 2012Received in revised form 28 January 2014Accepted 7 March 2014

Keywords:Energy savingIntermittentMethane controlMine ventilationSafety

With increase of energy costs and implementation of carbon tax in many countries, a cost-effective mineventilation system has become highly desirable in underground mine operations. In this study, a novelintermittent airflow ventilation system is proposed and evaluated via simulation with the goal of reduc-ing the energy cost whilst maintaining methane level in the mining face below the allowable level. Para-metric studies are conducted to investigate the effects of various factors influencing the effectiveness andperformance of this novel intermittent ventilation system, in particular the effect of intermittency period,air velocity, and application of multiple outlet nozzles with intermittent on-off flapper valves for air flowcontrol. Significant energy savings and air handling requirements are shown to be possible through thescheme proposed.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

During underground mining operation (especially undergroundcoal mine), large amounts of methane are released in the face area.This hazardous greenhouse gas has for decades been recognized asone of the prime causes of underground mine disasters around theglobe. Several major incidents and accidents related to the pres-ence of methane in underground coal mines with fatalities havebeen reported (Torano et al., 2009). To eliminate such incidentsand accidents, a good ventilation system is mandatory. Accordingto US regulations, methane concentration should be maintainedbelow 3% v/v while in other countries an even lower methane con-centration is mandatory: 1% in Germany, 1.25% in UK, 2% in Franceand 2.5% in Spain (Noack, 1998). It is expected that these require-ments will become increasingly stringent in future.

Most ventilation systems installed in underground coal minesnowadays supply excess fresh air to ensure safe methane concen-tration as required by local codes. A study by Reddy (2009) revealsthat up to 60% of the total mining operating cost is attributable tomine ventilation cost, highlighting the huge amount of electrical

energy consumed to drive fresh air to various locations in under-ground mine. With increase of energy costs and implementationof carbon tax in many countries, a cost-effective mine ventilationsystem which ensure a safe and productive environment in anunderground mine whilst keeping energy usage and operating costat minimum has become highly desirable for the mining industry.As such, a concept of ventilation-on-demand has received consid-erable attention (Tuck et al., 2006; Allen and Keen, 2008; O’Connor,2008). The idea behind this concept is straightforward: supply ade-quate fresh air to a certain location only when it is needed. Appli-cation of this system in underground mine, however, will requirecomplex control system and sophisticated monitoring devicesand sensors.

In tandem with experimental design and studies, numericalmodeling has come to play an important role in designing andexamining mine ventilation system in underground mine. Model-ing allows innovation at very low cost. Among the first researchersconducting computational fluid dynamics (CFD) modeling to inves-tigate ventilation airflow, methane emission and dust dispersion inunderground mine were Heerden and Sullivan (1993). They exam-ined ventilation airflow patterns around continuous miner in anactive mining area and its effect on the methane and dust distribu-tion. Next study was conducted by Srinivasa et al (1993). By utiliz-ing commercially available CFD tools, they investigate flowbehavior and dust movement in a longwall face. First validatedCFD model on underground mine ventilation was developed by

Fig. 1. Shcematic of an underground mine tunnel with ventilation duct.

J.C. Kurnia et al. / Tunnelling and Underground Space Technology 42 (2014) 206–215 207

Uchino and Inoue (1997) by using experimental data from bothfull-scale heading and a scale model. This model was extended tostudy methane distribution by Tomata et al (1999). Nakayamaet al (1999) conducted similar studies by investigating methanegas distribution in a mining face by utilizing CFD software LA-SAR95/98. Their model prediction achieved relatively good agree-ment with the experimentally measured counterparts. Wala et al(2003, 2007) developed a CFD model for longwall as well as roomand pilar mines and validated their model with the experimentaldata obtained from a lab-scale set-up. They extend their study byincluding the effect of scrubber operation on the face ventilation(Wala et al., 2008).

A multiphase Eularian model to predict dust behavior in a com-plex mine geometry was developed by Canoo (2004). Another com-putational study was conducted by Parra et al (2006). Aftervalidating their model, they investigated two scenario of additionalventilation system in a cul-de-sac mine: blowing and exhausting.Hargreaves and Lowndes (2007) numerically investigated the ef-fect of the drivage of continuous miner on the flow behavior inunderground mining face. Zheng and Tien (2008) conducted CFDmodeling of diesel particulate matters dispersion in a mining faceand examine the performance of blowing and exhausting ventila-tion system in reducing DPM concentration in the mining face.Torano et al (2009, 2011) conducted numerical study to evaluatemethane and dust behavior in mining area and validate it withexperimental data obtained from an underground mine in Spain.Recently, Sasmito et al (2013) reported a numerical study examin-ing various auxiliary ventilation equipment’s to provide betterventilation whilst keeping low energy usage.

In this study, a novel and original intermittent airflow ventila-tion system is proposed and evaluated via simulation with the goalof reducing the energy cost whilst maintaining methane level inthe mining face below the allowable level. A computational fluiddynamic (CFD) approach is utilized to investigate the flow behaviorand methane dispersion on mine tunnels with an intermittent flowventilation system. In our physical model, methane is uniformlyreleased from ten sources (10 � 10 cm2) in the mining face with to-tal flow rate of 0.05 m3 s�1. Parametric studies are conducted toinvestigate effects of various factors influencing the effectivenessand performance of this novel intermittent ventilation system:intermittency period, air velocity, and application of multiple

outlet nozzles with intermittent on-off flapper valves for air flowcontrol. It is noted that current mining codes refer only to steadyflows. So, if the proposed scheme were to be applied, changesmay be needed to local mining codes.

2. Mathematical formulation

A three-dimensional underground coal mining model is devel-oped for a typical mine tunnel which is the simplest and most usedin underground coal mining these days (please refer to Fig. 1). Themine tunnel 36 m long 3.6 m wide and 2.9 m high. A ventilationduct with a diameter of 0.6 m is hung at 1.9 m height from the floorand 0.6 m from the tunnel wall on the access road. Its setback dis-tance from the mining face is 6 m.

2.1. Governing equations

In the tunnel flow, simultaneous mass, momentum, energy andspecies transport occur. Methane is released from specified dis-crete sources in the mining face and it is dispersed by the ventila-tion airflow. Conservation equations for mass, momentum, andspecies can be expressed as

@q@tþr � qU ¼ 0; ð1Þ

@

@tðqUÞ þ r � qUU ¼ �rpþr � sþ qg; ð2Þ

@

@tðqcpTÞ þ r � ðqcpUTÞ ¼ r � ðkeff þ

cplt

PrtÞrT; ð3Þ

@

@tðqxiÞ þ r � ðqxiUÞ ¼ r � qDi;eff þ

lt

Sct

� �rxi: ð4Þ

where q is the fluid density, U is the fluid velocity, p is pressure, s isthe viscous stress tensor, g is gravity acceleration, cp is the specificheat of the fluid, keff is the effective fluid thermal conductivity, T isthe temperature, xi is the mass fraction of species i (O2, CH4 andN2), Di,eff is the effective diffusivity of species i, lt is turbulent vis-cosity and Sct is the turbulent Schmidt number and Prt is the turbu-lent Prandtl number.

Table 1Validation of numerical results for various turbulence models with experimental data.

Measurement point (see Parraet al. 2006)

Distance from miningface (m)

Spalart–Allmaras(Error, %)

Standard k-epsilon(Error, %)

Standard k-omega(Error, %)

Reynolds Stress Model (RSM)(Error, %)

1 4 43 33 63 371 12 51 48 90 131 18 51 51 81 132 4 36 44 56 52 12 28 31 60 322 18 57 28 66 323 4 52 44 93 213 12 15 41 14 283 18 60 15 27 284 4 41 51 66 704 12 56 68 35 724 18 48 56 35 725 4 72 15 37 65 12 30 48 47 785 18 68 30 43 78

R2 0.95 0.96 0.92 0.89

208 J.C. Kurnia et al. / Tunnelling and Underground Space Technology 42 (2014) 206–215

2.2. Constitutive relations

The viscous stress tensor for fluid can be expressed as

s ¼ ðlþ ltÞðrUþ ðrUÞTÞ � 23½ðlþ ltÞðr � UÞIþ qkI�; ð5Þ

where l is the dynamic viscosity of the fluid, q is the fluid density, Iis the identity or second order unit tensor and k is turbulent kineticenergy.

In the tunnel, a ternary species mixture, xi, comprising of oxy-gen, water vapor and methane is solved. The interaction betweenthe species is captured in the mixture density which followsincompressible ideal gas law as

q ¼ pMRT

; ð6Þ

where R is the universal gas constant and M refers to the mixturemolar mass given by

M ¼ xCH4

MCH4

þxO2

MO2

þxH2O

MH2OþxN2

MN2

� ��1

; ð7Þ

Here, M is the molar mass of species i. Mass fraction of nitrogen iscalculated as

xN2 ¼ 1� xO2 þxH2O þxCH4

� �; ð8Þ

The molar fractions, xi, are related to the mass fraction, given by

xi ¼xiMMi

ð9Þ

The fluid mixture viscosity is calculated by

l ¼X

i

xiliXj

xiUi;j

with i and j ¼ CH4;O2;H2O and N2 ð10Þ

where xi,j are the mole fraction of species i and j and

Ui;j ¼1þ ðli=ljÞ

1=2ðMj=MiÞ1=4h i2

8 1þ Mi=Mj� �� �� 1=2 ; ð11Þ

In-line with concentration unit commonly used in the regula-tion/law, methane concentration in this paper is presented in %v/v. In addition, fan power is calculated as

Pfan ¼ DPfan_Q fan; ð12Þ

where DPfan is pressure rise across the inlet and _Qfan is the volumet-ric flow rate of the fan. It should be noted that the actual fan powerwould be different depending on fan efficiency.

2.3. Turbulence model

The turbulence model is the key component in representingflow behavior in underground environment (Versteeg andMalalasekera, 1995). In selecting the appropriate turbulence mod-el, a comparison of flow behavior from various turbulence models,i.e., Spalart–Allmaras, k-epsilon, k-omega and RSM with the exper-imental data by Parra et al (2006) (see Table 1 for the comparison)was conducted. For sake of brevity, the details of the validation ofturbulence model can be seen in our previous work (Kurnia et al.,2014a and Kurnia et al., 2014b). Here, the most widely used modelin engineering field, k-epsilon model is chosen as it gives reason-able good prediction with R2 between CFD model and experimentaldata of 0.96; furthermore, the interest of the current study is notdetail of the flow in specific location but rather the overall flowbehavior in the mining tunnel for design purpose. In short, k-epsi-lon model considers two-equation model which solves for turbu-lent kinetic energy, k, and its rate of dissipation, e, which iscoupled with turbulent viscosity. This model is given as (Wilcox,1993)

@

@tðqkÞ þ r � ðqUkÞ ¼ r � lþ lt

rk

� �rk

� �þ Gk � qe; ð13Þ

@

@tðqeÞ þ r � ðqUeÞ ¼ r � lþ lt

re

� �re

� �þ C1e

eGk

kþ C2eq

e2

k;

ð14Þ

In above equations, Gk represents the generation of turbulencekinetic energy due to the mean velocity gradients, C1, and C2 areconstants, rk and re are the turbulent Prandtl numbers for k ande, respectively, and lt is turbulent viscosity given by

lt ¼ qClk2

eð15Þ

The values of C1e, C2e,, Cl, rk and re are 1.44, 1.92, 0.09, 1 and1.3, respectively.

J.C. Kurnia et al. / Tunnelling and Underground Space Technology 42 (2014) 206–215 209

2.4. Boundary conditions

The boundary conditions for the model are summarized as fol-lows: (i) At walls: the standard wall function is used in all simula-tions; (ii) inlet: air velocity of 12 m s�1 and various intermittentinlet velocity are prescribed at the duct outlet; (iii) At the miningface: methane is released at total flow rate of 0.05 m3 s�1; (iv) Atthe outlet: stream-wise gradient of the temperature is set to zeroand the pressure is set to standard atmospheric pressure (1 bar)with zero velocity gradient.

Fig. 3. (a) Cross-section average at 1 m from the mining face and (b) volume

3. Numerical methodology

The computational domains were created, meshed and labeledin Gambit 2.3.16 (Fluent 6.3 documentation). Three differentamount of mesh 5 � 105, 1 � 106 and 2 � 106 were implementedand compared in terms of local pressure, velocities, and methaneconcentration to ensure a mesh independent solution. It was foundthat the mesh amount of around 1 � 106 gives about 1% deviationcompared to the mesh size of 2 � 106; whereas, the results fromthe mesh size of 5 � 105 deviate up to 12% as compared to thosefrom the finest one. Therefore, a mesh of around 1 million elementswas sufficient for the numerical investigation purposes: a finestructure near the wall (with the minimum size of 10�3 m) andincreasingly coarser mesh (maximum size of 10�1 m) in the middleof the tunnel to reduce the computational cost.

The geometry and boundary conditions were solved using acommercial finite volume solver, Fluent 6.3.26. The equations weresolved with the well-known Semi-Implicit Pressure-Linked Equa-tion (SIMPLE) algorithm, first order upwind discretization andAlgebraic Multigrid Method (AGM). On average, each simulationrequired around 1000–3000 iterations for convergence toleranceof 10�6 for all variables. It takes around 6–8 h on workstations withsix core processor, requiring 6–8 GB RAM.

average methane concentration (% v/v) inside the mine tunnel for variousintermittent modes.

4. Results and discussion

Flow behavior and methane dispersion on mine tunnel withintermittent flow ventilation system are investigated. In the

Fig. 2. Inlet velocity for vario

following section, various possible configuration of innovativeintermittent ventilation system is presented and discussed.

us intermittent modes.

Fig. 4. Velocity contour (m s�1) at height 1 m from the mine floor for intermittent flow 5 min high (12 ms�1) 5 min low (6 ms�1).

Fig. 5. Methane concentration (% v/v) at 1, 8, 16, 24, 32 m from the mine face for intermittent flow 5 min high (12 ms�1) 5 min low (6 ms�1).

210 J.C. Kurnia et al. / Tunnelling and Underground Space Technology 42 (2014) 206–215

4.1. Various intermittent modes

To ensure sufficient oxygen supply for the miner and to main-tain concentration of methane below allowable maximum level,underground mine ventilation system generally drives excessiveamount of fresh air from the surface to various locations in under-ground mine. Large underground mines typically utilize large mainfans to drive the flow from the surface which is further distributedto smaller branches by using auxiliary ventilation system. Thisventilation system is working continuously during mining opera-tion. While majority of mining operator adopted this approach, itrequires large amount of electrical energy to power the fan. Here,an innovative way to save energy whilst ensuring safe environ-ment by introducing an intermittent ventilation system is pro-posed and evaluated. This approach can be implemented byinstalling control valves at the air flow branches distribution to di-rect fresh air towards one branch while blocking other one, this can

be done alternately so the intermittency can be achieved withoutaffecting main ventilation fans.

Here, several possible intermittency designs and their perfor-mance are compared and examined in terms of the methane con-centration and possible energy saving as compared to that withsteady ventilation flow. A 12 m/s steady flow from the main venti-lation shaft blows the mining face is compared with three inter-mittency scenarios: (case 1) 5 min high velocity (12 m/s) and5 min low velocity (6 m/s); (case 2) 5 min high velocity (12 m/s)and 10 min low velocity (6 m/s); and (case 3) 5 min high velocity(12 m/s) and 15 min low velocity (6 m/s), for which the dynamicinlet velocities are illustrated in Fig. 2.

Fig. 3 shows average methane concentration at the face 1 mfrom the mining face (Fig. 3a) and in the whole tunnel (Fig. 3b).Here several features are apparent: foremost among them is that,on average, methane concentration is maintained bellow the criti-cal explosive level of 5.5% v/v. Intermittency leads to dynamic

Table 2Maximum methane concentration (% v/v) for various intermittent modes.

Cases Cross-section Time (min)

0 2.5 5 7.5 10 12.5 15 17.5 20

Case 1 1 m 3.78 3.75 3.75 3.75 3.76 3.76 3.76 3.76 3.76Steady flow 5 m 1.98 1.97 1.97 1.97 1.97 1.97 1.97 1.97 1.97

10 m 1.71 1.70 1.70 1.70 1.70 1.70 1.70 1.70 1.70

Case 2a 1 m 3.75 3.75 3.75 6.73 6.83 3.79 3.75 6.78 6.805 min high 5 m 1.97 1.97 1.97 3.60 3.81 1.99 1.96 3.59 3.805 min low 10 m 1.70 1.70 1.70 2.98 3.14 1.73 1.69 2.99 3.14

Case 3a 1 m 3.75 3.75 3.72 6.74 6.83 6.83 6.81 3.79 3.765 min high 5 m 1.97 1.97 1.97 3.60 3.81 3.89 3.93 2.00 1.9710 min low 10 m 1.70 1.70 1.67 2.99 3.14 3.18 3.20 1.74 1.70

Case 4a 1 m 3.71 3.71 3.72 6.74 6.84 6.83 6.81 6.75 6.725 min high 5 m 1.97 1.97 1.97 3.60 3.82 3.90 3.93 3.95 3.9515 min low 15 m 1.67 1.67 1.67 2.99 3.14 3.18 3.20 3.22 3.24

a High = 12 ms�1 and low = 6 ms�1.

Table 3Energy saving for various intermittent modes.

Cases Description Pressure difference (Pa) Volumetric flow rate (m3 s�1) Fan power (Watt) Energy savingb (%)

Case 1a Steady flow (12 ms�1) 89.89 3.39 304.72 0

Case 2 5 min high (12 ms�1) 89.89 3.39 304.72 43.515 min low (6 ms�1) 23.42 1.69 39.58

Case 3 5 min high (12 ms�1) 89.89 3.39 304.72 58.0110 min low (6 ms�1) 23.42 1.69 39.58

Case 4 5 min high (12 ms�1) 89.89 3.39 304.72 65.2615 min low (6 ms�1) 23.42 1.69 39.58

a Case 1 is the base case used as reference for energy saving calculation.b Power saving is calculated based on 1-h operation.

J.C. Kurnia et al. / Tunnelling and Underground Space Technology 42 (2014) 206–215 211

behavior of flow (Fig. 4) and methane concentration (Fig. 5). Thestep changes in ventilation velocity changes the overall velocitybehavior inside the tunnel: when high velocity applies (Fig. 4aand c), a relatively high flow velocity develops throughout the tun-nel dispersing methane emission and forcing it to leave the tunnel(Fig. 5a and c); conversely, when low air velocity applies, air flowvelocity throughout the tunnel reduces significantly (up to 70% atthe outlet region, see Fig. 4b and d); this is further mirrored bythe rise in the methane concentration up to twice (Fig. 5b and d)and then reduces back to low methane concentration as the inter-mittency is periodically applied. It is also noteworthy to mentionthat during one period of intermittency, methane concentrationat high velocity becomes slightly increase (�10%, see Fig. 5a andc for detail comparison).

Another important finding is that intermittency duration playsimportant role to the methane removal. The results suggest thatshorter intermittency duration, case 1, yields the lowest averagemethane concentration as compared to other intermittent scenar-ios (Fig. 3). This is attributed to the shorter period of methane accu-mulation during low velocity ventilation which alleviates highermethane concentration. Looking to the maximum methane con-centration throughout the tunnel in Table 2, it is noted that themaximum methane concentration increases to almost double oncethe intermittency (reduce air velocity to half) is applied. At the dis-tance of 1 m from the mining face where the miners typicallyworks, methane concentration for all cases considered goes up tobeyond its allowable level at some points limit (more than 5.5%)which may trigger explosion when it mixes with oxygen and spark.These results indicates that for the mine considered in this studyintermittency by reducing air velocity into half may not be feasibleto maintain the methane concentration bellow its allowable limit.

Despite its inferior performance on handling methane removal,intermittency offers potential for energy saving as given in Table 3.

It is noted that significant amount of energy saving can be achievedup to 43.5%, 58% and 65% for case 1, 2 and 3, respectively. Theamount will even higher when it is translated to the annual oper-ating cost saving and, to some extent, company can claim for car-bon emission trading as well. Clearly, it can be deduced thatintermittency has potential to be implemented for energy saving;on the other hand, an improved intermittency design should bedeveloped and more studies is required to enhance methane re-moval and optimize the operating condition.

4.2. Shorter intermittent period and higher ventilation velocity

In the previous section, it is found that the intermittent flowventilation tested could not maintain methane concentration atsafe level. Methane concentration jump to dangerous level whenthe velocity falls to half indicating higher velocity and shorter per-iod of intermittence are required. Here, application of intermittentflow at higher velocity and shorter interval of intermittency areinvestigated. Two cases are considered: (case 2) 2.5 min highvelocity (12 m/s) and 5 min low velocity (9 m/s); and (case 3)5 min high velocity (15 m/s) and 5 min low velocity (9 m/s), asseen in Fig. 6. Similar to the previous discussion, steady flow ven-tilation will be used as reference (case 1).

Fig. 7 depicts the average methane concentration at the cross-section 1 m distance from mining face (Fig. 7a) and inside thewhole tunnel (Fig. 7b). It have been shown earlier in Fig. 3 thatintermittency can maintain the average methane concentrationbellow its explosive limit of 5.5% v/v; this is indeed the case, ascan be inferred from Fig. 7, where the average methane concentra-tion is maintained bellow its explosive limit. On closer inspection,it is revealed that the maximum average methane concentration islesser (around 1.45% v/v at 1 m distance from mining face and1.8% v/v for the whole tunnel) as compared to the previous scenarios

Fig. 6. Inlet velocity for intermittent ventilation system with higher velocity and short intermittence period.

Fig. 7. (a) Cross-section average at 1 m from the mining face and (b) volumeaverage methane concentration (% v/v) inside the mine tunnel for intermittentventilation system with higher velocity and short intermittence period.

212 J.C. Kurnia et al. / Tunnelling and Underground Space Technology 42 (2014) 206–215

in Fig. 3 (around 2.2% v/v at the 1 m distance from mining face and2.8% v/v for the whole tunnel). Now, by focusing at the local max-imum methane concentration in Table 4, it is found that for allcases, the maximum methane concentration is maintained belowits explosive level of 5.5% v/v with the maximum methane concen-tration is about 4.82% v/v for case 2 at the 1 m distance from min-ing face five min after intermittency is applied. This indicates thatintermittency is potential to be applied in underground mines dueto its capability for handling methane removal – of course furtheroptimization is required to achieve better performance.

Besides methane removal, energy saving becomes our concernin this innovative concept; Table 5 shows the potential energy sav-ing. It is observed that case 2 offers potential for energy saving upto �28% which shows potential to be implemented in real under-ground mine due to its ability for methane removal performancesimilar to one with steady ventilation. On the other hand, despiteits better performance on methane handling as compare to steadyventilation, case 3 performs worse in term of energy saving as itconstitutes to a negative saving (more power is required) due tohigher air velocity (15 m/s), which is mirrored by higher parasiticload. Hence, it can be concluded that case 2 offers balanced perfor-mance, reduced energy usage while keeping methane below itsexplosive level and shows potential for practical application.

4.3. Multiple duct outlet

Thus far it have been shown that intermittency ventilation haspotential to reduce energy consumption whilst keeping methaneconcentration below its allowable limit. In this Section, instead ofintermittency by reducing supply air to the mining face, intermit-tency is introduced by branching the outlet of ventilation duct into2 (case 2) or 3 branches (case 3) and direct the flow to thesebranches alternately, as shown in Fig. 8. The ventilation perfor-mance will again be compared with the steady flow ventilation(case 1) as the reference.

Keeping methane concentration at the lowest possible is ofinterest in order to ensure safety aspect in the mines. The average

Table 4Maximum methane concentration (% v/v) for intermittent ventilation system with higher velocity and short intermittence period.

Cases Cross-section Time (min)

0 2.5 5 7.5 10 12.5 15 17.5 20

Case 1 1 m 3.78 3.75 3.75 3.75 3.76 3.76 3.76 3.76 3.76Steady flow 5 m 1.98 1.97 1.97 1.97 1.97 1.97 1.97 1.97 1.97

10 m 1.71 1.70 1.70 1.70 1.70 1.70 1.70 1.70 1.70

Case 2a 1 m 3.75 3.68 4.82 3.68 4.80 3.68 4.79 3.68 4.792.5 min high 5 m 1.92 1.92 2.56 1.97 2.57 1.97 2.58 1.97 2.582.5 min low 10 m 1.61 1.61 2.12 1.67 2.13 1.67 2.13 1.67 2.13

Case 3b 1 m 3.00 3.00 4.81 3.02 4.80 3.01 4.80 3.01 4.802.5 min high 5 m 1.54 1.54 2.55 1.56 2.55 1.55 2.55 1.55 2.552.5 min low 10 m 1.30 1.30 2.12 1.31 2.12 1.31 2.12 1.31 2.12

a High = 12 ms�1 and low = 9 ms�1.b High = 15 ms�1 and low = 9 ms�1.

Table 5Energy saving for intermittent ventilation system with higher velocity and short intermittence period.

Cases Description Pressure difference (Pa) Volumetric flow rate (m3/s) Fan power (Watt) Energy savingb(%)

Case 1a Steady flow (12 m/s) 89.89 3.39 304.72 0

Case 2 2.5 min high (12 m/s) 89.89 3.39 304.72 28.672.5 min low (9 m/s) 51.10 2.54 129.97

Case 3 2.5 min high (15 m/s) 139.00 4.24 589.22 �18.012.5 min low (9 m/s) 51.10 4.24 129.97

a Case 1 is the base case used as reference for energy saving calculation.b Power saving is calculated based on 1-h operation.

(a) (b)

(c)

Fig. 8. (a) Schematics of mine tunnel with multiple ventilation duct outlet.

J.C. Kurnia et al. / Tunnelling and Underground Space Technology 42 (2014) 206–215 213

methane concentration for the proposed branching ventilation, asshown in Fig. 9, yields better performance as compared to the stea-dy ventilation. It is found that methane concentration is relativelyfar below explosive level of 5.5% v/v: the average concentration for

case 2 is �1.05% and 1.38% at the distance 1 meter from the miningface and inside whole tunnel, respectively; whereas, methane con-centration for case 3 is slightly about 10–20% higher than that ofcase 2. Upon closer inspection, it reveals that case 2 (with two

Fig. 9. (a) Cross-section average at 1 m from the mining face and (b) volumeaverage methane concentration (% v/v) inside the mine tunnel with multipleventilation duct outlet.

214 J.C. Kurnia et al. / Tunnelling and Underground Space Technology 42 (2014) 206–215

branches at the duct outlet) is able to maintain the lowest methaneconcentration at any given locations inside the tunnel at any giventime among other designs, as listed in Table 6. On contrary, case 3performs the worst, even as compared to the steady ventilationcase. This may be due to the fact that branching the flow at smalleralternate time is unable to effectively dilute methane concentra-tion due to short period of dynamic alternate blowing flow to themining face. Clearly, branching ventilation duct into two outletsenhances methane removal in the underground tunnel whichshows potential application. If one concerns on the energy saving,however, this design unfortunately does not offer energy saving(energy consumptions is about the same for all cases) since flowis only directed to different outlet at the same flow rate. In

Table 6Maximum methane concentration (% v/v) for various blowing duct configurations.

Cases Cross-section Time (min)

0 2.5 5

Case 1 1 m 3.78 3.75 3.75Single duct 5 m 1.98 1.97 1.97Outlet 10 m 1.71 1.70 1.70

Case 2 1 m 3.70 370 3.76Two duct 5 m 1.93 1.93 1.93Outlets 10 m 1.54 1.54 1.56

Case 3 1 m 3.85 3.85 3.89Three duct 5 m 1.88 1.88 1.93Outlets 10 m 1.53 1.53 1.56

addition, this design may add complexity to the system as flowcontroller, actuator and additional branching duct installation isrequired. However, if methane removal is of paramount impor-tance, e.g. in gassy mines, this design can be a desirable choice tomaintain safety level at high.

5. Conclusion

In this study, a novel and original intermittent airflow ventila-tion system is proposed and evaluated via simulation with the goalof reducing the energy cost whilst maintaining methane level inthe mining face below the allowable level. Parametric studies areconducted to investigate effects of various factors influencing theeffectiveness and performance of this novel intermittent ventila-tion system: intermittency period, air velocity, and application ofmultiple outlet nozzles with intermittent on-off flapper valvesfor air flow control.

The results are promising based on the CFD results obtained todate for a model underground mine, whereby intermittent ventila-tion offers over 25% electrical energy savings despite its slightlyinferior methane handling performance (where methane concen-tration increases by about 28% at some locations) compared tosteady flow ventilation. In addition, for mines where methane isnot a primary issue (non-coal underground mines), intermittentflow with 5 min high velocity and 15 min low velocity could offerup to 65% of energy savings. Such energy savings will not only re-duce expenses in electricity bill but also will result in very signifi-cant savings from carbon tax credits. More parametric studies arenow being carried out to obtain an optimum ventilation designwhich could save energy usage and in turn carbon credit tax asso-ciated with it even more. In addition this study will be extended toinvestigate application of intermittent flow ventilation system innon-coal underground mine where diesel emission and oxygendepletion are the prime issues rather than methane as in under-ground coal mine. We are currently seeking designs of real under-ground mine configurations for prediction of energy savingsutilizing our new design concept without compromising safety ofminers. Since current mine codes are based on steady flows,changes to local codes will be necessity if this novel idea is to beimplemented widely. This study is conducted in specific under-ground mine size, parametric study on the effect of intermittencywith regard to the different mining size/geometry will be carriedout in future work.

Acknowledgement

This work was financially supported by Singapore EconomicDevelopment Board (EDB) through Minerals Metals and MaterialsTechnology Centre (M3TC) Research Grant R-261-501-013-414.

7.5 10 12.5 15 17.5 20

3.75 3.76 3.76 3.76 3.76 3.761.97 1.97 1.97 1.97 1.97 1.971.70 1.70 1.70 1.70 1.70 1.70

3.64 3.63 3.66 3.72 3.64 3.631.86 1.86 1.98 1.96 1.86 1.861.62 1.62 1.58 1.57 1.63 1.62

3.82 3.83 3.80 3.80 3.86 3.871.98 2.00 1.87 1.87 1.91 1.901.64 1.63 1.55 1.55 1.56 1.55

J.C. Kurnia et al. / Tunnelling and Underground Space Technology 42 (2014) 206–215 215

References

Allen, C., Keen, B., 2008. Ventilation on demand (VOD) project – Vale Inco Ltd.,Coleman Mine. In: Proceeding of the 12th North American Mine VentilationSymposium 2008, pp. 45–49.

Canoo, B., 2004. STAR-CD digs miners out of trouble. CD Adapco Dyn. Fall 2004, 27–28.

Fluent 6.3 documentations, www.fluent.com. (accessed 2008).Hargreaves, D.M., Lowndes, I.S., 2007. The computational modeling of the

ventilation flows within rapid development drivage. Tunn. Undergr. SpaceTechnol. 22, 150–160.

Heerden, J., Sullivan, P., 1993. The application of CFD for evaluation of dustsuppression and auxiliary ventilation systems used with continuous miners. In:Proceeding of the 6th US Mine Ventilation Symposium. SME, Littleton, pp. 293–297.

Kurnia, J.C., Sasmito, A.P., Mujumdar, A.S., 2014a. Dust dispersion and managementin underground mining faces. Int. J. Min. Sci. Technol. 24, 39–44.

Kurnia, J.C., Sasmito, A.P., Mujumdar, A.S., 2014b. CFD simulation of methanedispersion and innovative methane management in underground mining faces,Appl. Math. Modell. in press, http://dx.doi.org/10.1016/j.apm.2013.11.067.

Nakayama, S., Kim, Y.K., Jo, Y.D., 1999. Simulation of methane gas distribution bycomputational fluid dynamics. In: Xie, H.P., Golosinski, T.S. (Eds.), Mining andScience Technology. Balkema Publisher, Brookfield, pp. 259–262.

Noack, K., 1998. Control of gas emissions in underground coal mines. Int. J. CoalGeol. 35, 57–82.

O’Connor, D.F., 2008. Ventilation on demand (VOD) auxiliary fan project – Vale IncoLimited Creighton Mine, In: Proceeding of the 12th North American MineVentilation Symposium 2008, pp. 41–44.

Parra, M.T., Villafruela, J.M., Castro, F., Mendez, C., 2006. Numerical andexperimental analysis of different ventilation systems in deep mines. Build.Environ. 41, 87–93.

Reddy, A.C., 2009. Development of a Coal Reserve GIS Model and Estimation of theRecoverability and Extraction Costs, Master of Science Thesis, Department ofMining Engineering, West Virginia University.

Sasmito, A.P., Birgersson, E., Ly, H.C., Mujumdar, A.S., 2013. Some approaches toimprove ventilation system in underground coal mines environment – acomputational fluid dynamic study. Tunn. Undergr. Space Technol. 34, 82–95.

Srinivasa, R.B., Baafi, E.Y., Aziz, N.I., Singh, R.N., 1993. Three dimensional modelingof air velocities and dust control techniques in a longwall face. In: Proceeding ofthe 6th US Mine Ventilation Symposium. SME, Littleton, pp. 287–292.

Tomata, S., Uchino, K., Inoue, M., 1999. Methane concentration at heading faces withauxiliary ventilation. In: Proceeding of the 8th US Mine Ventilation Symposium.SME, Littleton, pp. 187–192.

Torano, J., Torno, S., Menendez, M., Gent, M., Velasco, J., 2009. Models of methanebehaviour in auxiliary ventilation of underground coal mining. Int. J. Coal Geol.80, 35–43.

Torano, J., Torno, S., Menendez, M., Gent, M., 2011. Auxiliary ventilation in miningroadways driven with roadheaders: validated CFD modeling of dust behaviour.Tunn. Undergr. Space Technol. 26, 201–210.

Tuck, M.A., Finch, C., Holden, J., 2006. Ventilation on demand: a preliminary studyfor Ballarat Goldfields NL, In: Proceeding of the 11th North American MineVentilation Symposium 2006, pp. 11–14.

Uchino, K., Inoue, M., 1997. Auxiliary ventilation at a heading of a face by a fan. In:Proceeding of the 6th US Mine Ventilation Symposium. SME, Littleton, pp. 493–496.

Versteeg, H.K., Malalasekera, 1995. An introduction to computational fluiddynamics – the finite volume method, Longman Scientific and Technical.

Wala, A., Jacob, J., Brown, J., Huang, G., 2003. New approaches to mine-faceventilation. Min. Eng. 55 (3), 25–30.

Wala, A.M., Vytla, S., Taylor, C.D., Huang, G., 2007. Mine face ventilation: acomparison of CFD results against benchmark experiments for the CFD codevalidation. Min. Eng. 59 (10), 1–7.

Wala, A.M., Vytla, S., Huang, G., Taylor, C.D., 2008. Study on the effect of scrubber onthe face ventilation. In: Proceeding of the 12th North American MineVentilation Symposium 2008, pp. 281–286.

Wilcox, D.C., 1993. Turbulence modeling for CFD. DCW industries Inc.Zheng, Y., Tien, J.C., 2008. DPM dispersion study using CFD for underground metal/

nonmetal mines. In: Proceeding of the 12th North American Mine VentilationSymposium 2008, pp. 487–493.