ARTICLE IN PRESS
Building and Environment 41 (2006) 87–93 www.elsevier.com/locate/buildenv
Numerical and experimental analysis of different ventilation systems in deep mines M.T. Parra, J.M. Villafruela, F. Castro, C. Me´ndez Dpto. Ingenierı´a Energe´tica y Fluidomeca´nica, E. T. S. Ingenieros Industriales, Universidad de Valladolid, Paseo del Cauce sln, 47011 Valladolid, Spain Received 19 February 2004; received in revised form 10 December 2004; accepted 4 January 2005
Abstract The ventilation system plays an essential role in underground workings. On an experimental and numerical sense, the present paper deals with a study on ventilation systems working in the cul-de-sac of a coal mine. The measurements taken in a real mine gallery with a hot-wire anemometry, have been used to validate the numerical model. Three different types of ventilation systems have been numerically examined: blowing, exhaust and mixed ventilation. The numerical models provide detailed information about the flow field, mean age of air and the methane concentration. This information allows contrasting the ventilation system efficiency against diverse criteria: stagnation regions, contaminated air regions and risk of explosions regions. r 2005 Elsevier Ltd. All rights reserved. Keywords: Ventilation; Coal mine; CFD; Hot wire anemometry; Blowing; Exhaust; Mean age of air; Risk of explosions
1. Introduction Coal mining in underground mines of up to 1000 m depth, generates one of the most dangerous working environments for humans. Moreover, the presence of methane and coal dust produces a risk of explosion. In addition, there is a danger of serious damage to health in cases of prolonged exposure to these and other airborne pollutants. There are two main reasons for air pollution while mining [1]: the first is that carbon dioxide is exhausted from internal combustion engines. The second is that soil contains gases such as methane. Thus, methane, coal dust and other pollutants are mainly gathered in the mining area, next to the closed side of the cul-de-sac. Therefore, it represents the highest risk area. But pollutants are most efficiently reduced by their rapid dilution, i.e. by introducing fresh air through the ventilation system. To this purpose it is necessary to Corresponding author. Tel.: +34 983 42 33 13x44 14; fax: +34 983 42 33 63. E-mail address:
[email protected] (M.T. Parra).
0360-1323/$ - see front matter r 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.buildenv.2005.01.002
introduce continuous and uniform fresh air into the workings. The existence of areas inadequately ventilated may produce pollutants build-up with the risks involved. At this point an efficient ventilation system is clearly needed. Long ducts, air diffusers and fans that work as blowers or exhausters comprise the underground mines ventilation system. The main objective of these elements is to supply fresh air, and even more important, to remove contaminated air with pollutants and dust so as to avoid build-ups in workings. Both the working conditions improvement and safety at work depend on the ventilation system ability to removing the hazardous pollutants. It is a broadly admitted fact that a ventilation system is suitable when it guarantees in all areas the mean velocity specified in the governing regulation. Lately, there has emerged an attempt to seek further restrictive parameters when it comes to assess the ventilation system performances. The authors have found the application of these factors in public places but not yet in underground mines [2]. The ventilation quality is not to be quantified exclusively by global parameters such as air change
ARTICLE IN PRESS M.T. Parra et al. / Building and Environment 41 (2006) 87–93
88
rate. A parameter such as the percentage of regions with a velocity module inferior to the one given (dead zones), does not represent by itself a guarantee of a proper dilution. The mean age of air and local levels of pollutants’ concentration in risk areas become better parameters in terms of quantifying the ventilation quality. The aim of this study is the analysis of three ventilation systems: exhaust, blowing and mixed. Neither the humidity nor the heat transfer effects have been taken into account. To this purpose a numerical model has been developed. The numerical model solves the fluid field (velocities, pressures, concentrations, levels of turbulence,y) and the local mean age of air. Experimental data with hot-wire anemometry at a real mine have been obtained in order to validate the numerical model of the fluid field. By analysing the dead zones and the mean local age of air it is possible to compare the quality of every ventilation system. Likewise, by analysing those areas of high concentrations of methane, the risk of explosion will be determined. These results intend to shed some light on flow patterns of different ventilation systems and also to establish some trends to design a ventilation system.
2. Experimental set-up The purpose of the experimental analysis is to have available measures for numerical model validation. The measures have been executed in a real coalmine gallery (fifth south gallery of the Barredo Underground Experimental and Training Centre in Mieres, Asturias, Spain), Fig. 1. The gallery is 48 m long with a quasisemicircular cross-section of 9 m2, approximately. The 0.6 m diameter ventilation duct travels near the roof and parallel to the gallery. The ventilation duct end is placed at 6 m from the closed side of cul-de-sac (Fig. 2) and the driven airflow is Q ¼ 3:39 m3 =s: In Fig. 1, there are also two secondary pipes with smaller diameter: one for compressed air and the other for water drainage.
Fig. 1. Mine gallery.
An IFA 100 of TSIs hot-wire anemometer was used with two types of probes: an omni-directional one (DANTECs 55R49) for velocities under 0.8 m/s and the other was a unidirectional film probe (DANTECs 55R31) for velocities between 0.5 and 20 m/s. Hot wire measurements have been done on different cross-section between the closed side of the gallery and Z ¼ 18 m: Fifty-two points have been measured in each crosssection evenly distributed accordingly to a 0.65 m spacing in horizontal direction, X, and 0.25 m in vertical direction, Y.
3. Numerical model validation The previous experimental results are used to validate the numerical model. Therefore, the geometric domain of Fig. 1 has been reproduced as well as the real culde-sac ventilation conditions. The pipes for compressed air and water drainage were not considered because of their low influence on the flow field. To discretize the geometric domain, Fig. 2, a three-dimensional grid of 5 104 hexahedral cells, approximately, has been generated. On the duct end, uniform velocity of V z ¼ 12 m=s was imposed and on the head end of the gallery, the condition of atmospheric pressure was imposed. Navier Stokes equations for a three dimensional, steady, incompressible and isothermal flow have been solved. The Spalart-Allmaras [3] turbulent model of one equation has been selected in order to calculate the turbulent viscosity, mt : qrmt þ divðrV mt Þ ¼ Pmt þ divðGmt grad mt Þ Dmt qt
(1)
where the production and the destruction of turbulent viscosity, Pmt and Dmt ; are evaluated from empirical formulae. Spalart-Allmaras gives fairly good behaviour in both the near wall regions as well as the free shear flow domain. Second-order upwind numerical scheme has been used. Fig. 3 shows the Vz numerical isovelocity contours for different cross-sections, along with the experimental values in locations in which the velocity supplied by the anemometer is over 0.5 m/s. If the velocity is under this value, the applied experimental technique evidence high rate of errors [4]. For Zp6 and X14 m zones, the deviations obtained remained under 10%. At the middle zones the flow is affected by the presence of the duct end. Furthermore, the secondary flows are accountable for the increased deviations that go up to 20%. The good agreement between the experimental data and the numerical results guarantees the suitability of both the grid spatial resolution and the algorithm used for the simulation of this type of flow.
ARTICLE IN PRESS M.T. Parra et al. / Building and Environment 41 (2006) 87–93
89
Ventilation duct y
y 0.6
1.8 m
Q
2.9 m
1.2 m
1.9 m 1.1m z
x 6m 36 m
(b)
(a)
Fig. 2. Cul-de-sac scheme: (a) front view and (b) lateral view.
4. Parametrical study E –3.21 N –3.47
E 1.25 N 1.38 E 1.83 N 1.68
E –4.44 N –4.65
E 2.19 N 2.39
–8 –4 –2
2
–1
0
1
(a) 0 E 0.94 N 1.10 –0.5
0.5
1
E 0.96 N 0.85
E 0.95 N 1.07 1.5
E 1.24 N1.53
E 0.90 N 0.79
(b)
0.25
E 0.84 N 0.86
E 0.77 N 0.84
0.5
0
Once the numerical model has been validated with experimental data, different ventilation systems are to be analysed: blowing, exhaust, and different mixed layouts [5]. The blowing ventilation involves the driving of fresh air towards the closed side of the cul-de-sac (case Blow6m in Table 1). On the other hand, the exhaust ventilation sucks the contaminated air from the working area (cases Exh6m and Exh1m in Table 1). All cases have been simulated by the same geometry layout that corresponds to the one shown in Fig. 4 in absence of auxiliary duct. Improvements in pits ventilation require the installation of mixed ventilation. The mixed ventilation, apart from the main pipe, utilise an auxiliary duct, near the closed side, equipped with a fan (cases Mix1 to Mix3 in Table 1). The purpose of this second pipe is to induce an additional flow inside the gallery, but in the opposite direction as for the main duct, see Fig. 4. A simulation of a cul-de-sac of 48 m long has been done by using 8.5 104 cells. The diameter of main duct is d m ¼ 0:6 m and the auxiliary diameter is d aux ¼ 0:3 m: An airflow of Q ¼ 3:4 m3 =s goes through the main duct and an airflow of Q ¼ 0:85 m3 =s through the secondary one. In fresh air inlet surfaces, the uniform inlet velocity is imposed as boundary condition. In contaminated air outlet surfaces the static pressure is imposed. A momentum source term is imposed to model the internal fan into the auxiliary duct. The remainder of numerical model features overlap the ones described in the validation section.
E 0.79 N 0.73
E 0.80 N 0.79
0.75
1 (c) Fig. 3. Numerical and experimental values of Vz velocity: (a) Z ¼ 4 m; (b) Z ¼ 12 m and (c) Z ¼ 18 m:
5. Dead zones The Spanish regulation on mining safety establishes that ventilation system should guarantee a minimum velocity of air into the coal mine of 0.2 m/s. Dead zones
ARTICLE IN PRESS M.T. Parra et al. / Building and Environment 41 (2006) 87–93
90 Table 1 Ventilation systems and their efficiency Case
Zm (m)
Zaux (m)
Laux (m)
Main duct
Auxiliary duct
Gallery entry
Efficiency
Blow6m Exh6m Exh1m Mix1 Mix2 Mix3
6 6 1 25 15 2
— — — 5 2 5
— — — 30 23 10
Blowing Exhaust Exhaust Exhaust Blowing Exhaust
— — — Blowing Exhaust Blowing
Air Air Air Air Air Air
0.828 0.561 0.954 0.315 0.659 0.910
daux
exit entry entry entry exit entry
Zm
dm
x 0.9m 0.9m
2.0m
z 1.8 m
1.1m
2.9m
y
Zaux Laux
x
48 m
(a)
(b) Fig. 4. Gallery model for the parametric study: (a) front view and (b) upper view.
Percentage of dead zones (%)
100
Mix1
90
Mix2
80
Mix3
70
Exh6m
60
Exh1m Blow6m
50 40 30 20 10 0 0
10
20
30
40
Z (m) Fig. 5. Spatial distribution of dead zones.
are defined as those regions where velocity module remains under this value. By assessing the dead zones’ distribution along the gallery we may perform a first assessment on the quality of different ventilation systems. As mining operations are executed in the vicinity of the closed side of the cul-de-sac (Z small), this region shows the greatest pollutants’ concentration. Therefore, a right air renewal should be chiefly ensured at this area. Fig. 5 shows the dead zones percentages in every cross-section for the different systems examined. There is no air movement at the zone between the duct end and the closed side when applying an exhaust ventilation system, without an auxiliary duct support (Exh6m). The same happens when air is exhausting at 1 m of the closed side (Exh1m) though the stagnation area is smaller. The exhaust ventilation in this area
shows a slight improvement when a blowing auxiliary duct is available (Mix1 and Mix3). On the contrary, the blowing ventilation system is featured by the virtually absence of dead zones from the gallery closed side to the duct end (Blow6m). An additional improvement is achieved by the support of an auxiliary duct, (Mix2).
6. Air age and effectiveness Nevertheless, the analysis of ventilation quality based on determining dead zones may not be appropriate. There might be deficiently ventilated zones as a result of the recirculation, even if velocities are over 0.2 m/s. Furthermore, there are multiple researches, in which, the
ARTICLE IN PRESS M.T. Parra et al. / Building and Environment 41 (2006) 87–93
91
350
Mix1 Mix2
300
Mix3
250
Age (s)
Exh6m 200
Exh1m Blow6m
150 100 50 0 0
5
10
15
20
25
30
35
40
Z (m) Fig. 6. Spatial distribution of local mean age of air.
ventilation system study is based on local mean age of air analysis [2,6,7]. The local mean age of air, tp ; is calculated by solving a partial diferential equation just as Sandberg [8] established qtp q Vi ¼ qxi qxi
nt n qtp þ þ 1; st s qxi
(2)
where Vi is the component in the i direction of the mean velocity, n and nt are the laminar and turbulent kinematic viscosity and s and st are the laminar and turbulent Schmidty numbers, respectively. The value of local mean age of air is imposed as null in the fresh air inlet surfaces. Beforehand, it is convenient that air mean age was relatively low (fresh air), especially at the closed side vicinities. On comparing ventilation systems, the average of local mean age of air for each cross-section of the gallery is assessed. Fig. 6 shows the evolution of this average along the gallery for all the previous systems. Due to the boundary condition of the local mean age, only those systems in which fresh air inlet areas overlap are comparable, i.e. Blow6m with Mix2 amongst themselves; and Exh1m with Exh6m with Mix1 and with Mix3 amongst themselves. By analysing Fig. 6 we find that best performances of exhaust ventilation are obtained by exhausting air at 1 m of the closed side (Exh1m) or at 2 m with an auxiliary duct support (Mix3). When air is exhausted at higher distances, with or without auxiliary duct (Mix1, Exh6m), the area near the working face is occupied by very polluted air as a result of increased air ages. But in all circumstances from the area of fresh air inlet to the exhaust duct, the displacement of the flow is piston type, i.e. uniform velocity parallel to the mine’s axis. However, better results are obtained for the blowing ventilation, if the duct end is closer to the working face.
According to Roos [2] the global efficiency of ventilation system may be defined as ¼
tp;exit ; 2tp;total
(3)
where tp;exit is the local mean age in the outlet section and tp;total is the local mean age in the whole gallery. A perfect displacement flow would reveal an efficiency value of 1 and a perfect mixing flow has an efficiency value of 0.5 [2]. The definition of efficiency allows the comparison of any ventilation system, regardless the fresh air inlet section. Last column of Table 1 shows the efficiency values obtained for the studied systems. An important volume with high local mean age denotes low efficiency as it is possible to observe in Mix 1 and Exh6m cases in Fig. 6 and Table 1. The remaining cases as previously pointed, show a perfect displacement flow in the majority of the gallery, therefore the efficiency is close to 1. According to efficiency criterion, the optimal layout would be accomplished by air exhaust at 1 m of the closed side, Exh1m. The worst efficiency was obtained with the mixed layout by exhaust at 25 m of the working area, Mix1.
7. Explosion risk Coalbeds contain a mixture of gases made up an 80% to 99% of methane [9]. Poor gas ventilation allows methane to accumulate in amounts that could be ignited by a spark from mining equipment. Therefore, there is a third option in terms of studying the quality of ventilation system, apart from the dead zones and the mean ages of air. It consists of analysing methane dispersion. This section displays the results obtained from the constant emission of a methane flow rate of QCH4 ¼
ARTICLE IN PRESS M.T. Parra et al. / Building and Environment 41 (2006) 87–93
92
0:030 m3 =s through the closed side of cul-de-sac with exhaust and blowing ventilation (Exh6m y Blow6m.) The numerical model must be solved by a non-steady solver. The steady solution is considered to be accomplished when the mass flow emanated is the same as the one removed. Each country has its own mine safety regulation with different permissible limit of methane concentration, such as the 1% in Germany, 1.25% in the United Kingdom, 2% in France or 2.5% in Spain [10]. The comparison of both ventilation systems, exhaust and blowing, is carried out by the distribution of the zones at flammability risk according the Spanish regulation, Fig. 7. It is observed that the blowing model shows absence of these zones whereby the exhaust model offers no possibilities to avoid this risk near the closed side of the cul-de-sac (Zo7 m). Fig. 8 shows the isoconcentration contours for the cross-section Z ¼ 0:5 m: The buoyancy effect together 14
Exh6m Percentage of ignition zones (%)
12
Blow6m
10
8
6
4
with the low velocity field for the exhaust ventilation produce a methane build-up at the upper part of the gallery, Fig. 8b. The blowing system, Fig. 8a, allows a better methane dilution, as velocities are higher at the zone near the emission.
8. Summary and conclusions Measurements carried out in the interior of a cul-desac in a real mine have allowed validating a numerical model. The model analyses the quality of ventilation system under three different criteria: dead zone analysis starting from the velocity distribution; effectiveness starting from the distribution of local mean ages of air; and risk of explosion starting from the distribution of methane concentrations. All these criteria have been applied to different ventilation systems of common use in deep mines. The first two criteria provide quite similar results, but the most advisable is the one based on effectiveness. The dead zones criterion does not take into account the flow recirculation. The risk of explosion criterion requires a higher time for computing, as the simulation is non-steady one. However, it supplies additional information of pollutants’ behaviour inside airflow induced by the ventilation system. So, we can gather that systems are efficient if exhaust is taking place very close to the closed side of the gallery (about 1 m) or air is blown at a distance relatively far away from the working face. Likewise, results may imply that efficient ventilation may be accomplished in the vicinity of the working face without an auxiliary duct.
2
Acknowledgements 0 0
2
4
6
8
Z (m) Fig. 7. Spatial distribution of zones with the risk of explosion.
The authors wish to thank the Association for the Research and Industrial Development of Natural Resources AITEMIN for giving them the opportunity to achieve the experimental measures using the
4% 3% 2%
0.3 %
1% 0.5 %
0.4 %
0.25 %
0.6 % 0.8 %
(a)
(b)
Fig. 8. Methane concentration isocontours (Z ¼ 0:5 m): (a) Blowing (Blow6m) and (b) Exhaust (Exh6m).
ARTICLE IN PRESS M.T. Parra et al. / Building and Environment 41 (2006) 87–93
ventilation facilities of the Barredo Underground Experimental Centre. References [1] Likar J, Cade J. Ventilation design of enclosed underground structures. Tunnelling and Underground Space Technology 2000;15/4:477–80. [2] Roos A. On the effectiveness of ventilation. Ph.D. thesis dissertation, Eindhoven University of Technology, 1999. [3] Spalart PR, Allmaras SR. A one-equation turbulence model for aerodynamic flows. AIAA Paper 92-0439, 1992. [4] Hall E, Timko R. Determining the accuracy of instruments capable of measuring low air velocities. Proceedings of the 2005 society of mining engineers annual meeting. Salt Lake City, UT, USA, 2005.
93
[5] Kissell FN. Control of dust in hard-rock tunnels. Handbook for dust control in mining. NIOSH publication no. 2003-147, 2003. [6] Bartak M, Cermak M, Clarke JA, Denev J, Drkal F, Lain MD, Mardonald IA, Majer M, Stankov P. Experimental and numerical study of local mean age of air. Seventh international IBPAS conference, Brazil, 2001. [7] Karimipanah T, Awbi HB. Theoretical and experimental investigation of impinging jet ventilation and comparison with wall displacement ventilation. Building and Environment 2002;37/ 12:1329–42. [8] Sandberg M. What is the ventilation efficiency? Building and Environment 1981;16:123–35. [9] Flores RM. Coalbed methane: from hazard to resource. International Journal of Coal Geology 1998;35:3–26. [10] Noack K. Control of gas emissions in underground coal mines. International Journal of Coal Geology 1998;35:57–82.
