Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
Application of ventilation simulation to spontaneous combustion control
in underground coal mine: A case study from Bulianta colliery
Liang Yuntao
a
, Zhang Jian
b,c,
⇑
, Ren Ting
c
, Wang Zhongwei
c
, Song Shuanglin
a
a
Shenyang Branch of China Coal Technology & Engineering Group (CCTEG), Shenyang 110016, China
b
College of Safety Science and Engineering, Henan Polytechnic University, Jiaozuo 454000, China
c
School of Civil, Mining & Environmental Engineering, University of Wollongong, NSW 2522, Australia
article info
Article history:
Received 2 January 2017
Received in revised form 13 April 2017
Accepted 26 May 2017
Available online 16 December 2017
Keywords:
Ventilation simulation
Spontaneous combustion
Longwall operation
Pressure differential
Ventsim
abstract
Spontaneous combustion of residual coal in longwall goaf is a long standing hazard. Airflow leakage into
goaf is a major driver to the hazard and this issue deteriorates where longwalls are operating in multiple
seams and shallow covers because mining-induced cracks are very likely to draw fresh airflow into goaf
due to presence of pressure differential between longwall face and surface. To study the problem more
critically, a ventilation simulation package ‘‘Ventsim” is used to conduct a case study from Bulianta col-
liery. It was found that isolating and pressurizing active longwall panel can mitigate the problem and the
pressure differential can be adjusted by varying performance of auxiliary fan and resistance of ventilation
regulator. A booster ventilation system can also mitigate the problem by adjusting fan duties. Ventilation
simulation is a powerful tool to study spontaneous combustion control in underground coal mine.
Ó2017 Published by Elsevier B.V. on behalf of China University of Mining & Technology. This is an open
access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction
Coal, as a carbonaceous material, is capable of being oxidized
and generating heat from ambient temperatures [1–4]. Self-
heating or even spontaneous combustion of coal mass is likely to
outbreak under favorable circumstances during many processes
of coal extraction and utilization [2,5,6]. Especially underground
coal mine fires have been identified as one of the most devastating
mining hazards for posing a great threat to miners, burning out
valuable coal mine assets, and giving off toxic and greenhouse
gases [7,8]. Reviewing Australian coal mining history more than
125 fire incidents have been recorded in New South Wales whilst
at least 68 incidents have been reported in Queensland from
1960 to 1991 and most of them occurred in underground workings
[9]. From 1990 to 1999, approximately 17% of the 87 total reported
fires for U.S. underground coal mines were caused by self-heating
[10]. In India 75% of the coal mine fires occurs due to spontaneous
combustion [11]. In China more than 50% of coal mines have had
self-heating incidents and there are estimated to be 360 fire inci-
dents each year caused by the spontaneous combustion within
only several key coal mines [12]. A third of the 254 mine fires
reported during the period from 1970 to 1990 was caused by spon-
taneous combustion of coal in South Africa [13].
Generally several internal and external factors can contribute to
spontaneous combustion of coal in underground coal mine [14].
Intrinsic factors like coal properties and geological conditions are
beyond control of coal operators. While Extrinsic factors such as
longwall (LW) panel layout, ventilation deployment, and mine
planning can be managed by coal operators. Among those external
factors ventilation arrangement is possibly of the utmost impor-
tance because airflow leakage into goaf from ventilation in LW
working is a necessary element of fire. The primary duties of mine
ventilation are to dilute hazardous accumulation of gas and dust, to
dissipate heat primarily produced by mining machines, and to sup-
ply respirable air to underground working force [15–17]. A proper
ventilation network is capable of fulfilling this duty in an econom-
ical means while a poorly managed ventilation system is very
likely to fail the duty and even worse, to facilitate development
of some mining hazards. Spontaneous combustion is one of them
as coal mine ventilation is inevitably feeding oxygen rich air into
longwall goaf where a significant amount of coal is left. Today
there is a strong move to longer panels, wider faces, greater extrac-
tion heights, increased production rates, more efficient ventilation
and decreased personnel in longwall coal mine [18]. The coal
seams in newly developed mines or sections are generally thick
and the risk of spontaneous combustion increases significantly
during longwall mining due to the large quantities of broken coal
https://doi.org/10.1016/j.ijmst.2017.12.005
2095-2686/Ó2017 Published by Elsevier B.V. on behalf of China University of Mining & Technology.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
⇑
Corresponding author at: College of Safety Science and Engineering, Henan
Polytechnic University, Jiaozuo 454000, China.
E-mail address: jz164@uowmail.edu.au (J. Zhang).
International Journal of Mining Science and Technology 28 (2018) 231–242
Contents lists available at ScienceDirect
International Journal of Mining Science and Technology
journal homepage: www.elsevier.com/locate/ijmst
left behind the chocks and its exposure to high oxygen levels in the
goaf [19]. Due to depletion of the first coal seam, many coal mines
in China have extracted the second seam or mined multi-seams
simultaneously. The trending can now be found in Australian min-
ing industry as well. It undoubtedly will pose more complexities to
ventilation circuits and difficulties to manage coal spontaneous
combustion because mining-induced cracks are more developed
and more likely to propagate to surface to draw more air leakage
for multi-seam LW operations. In exhaust ventilation system fresh
air is drawn from surface to LW working face through the intercon-
nected mining-induced cracks and vice versa for the force ventila-
tion system. The pressure differential between LW working and
surface is the major driver for the leakage, so minimizing the pres-
sure differential is another important duty of ventilation for LWs
operated in multiple coal seams and under shallow cover. A popu-
lar philosophy in dealing with spontaneous combustion hazard is
prevention is always better than cure. Although many advances
in gas monitoring techniques, sealing and stopping construction,
and proactive inertisation plan have been achieved, a more compe-
tent ventilation system which can reduce the leakage into goaf is
the first and also the most important shield to the hazard. To quan-
tify the pressure differential and investigate the issue with more
details, a ventilation simulation program called ‘‘Ventsim” is used
to perform a case study based on a real ventilation network of
Bulianta colliery. The colliery is one of the most productive LW
operations in China and also a very representative LW operated
in multiple coal seams and under shallow cover.
2. Project description
2.1. General introduction
Bulianta colliery is situated 13 km south to Ordos city of Inner
Mongolia Autonomous Region of Northern China (see Fig. 1). The
colliery is operated in Shendong coalfield which is featured with
flat and thick coal seam under shallow cover. Mining area of the
colliery is approximately 34 km
2
and the total proven reserve
exceeds 506 million tons of coal. Due to recent upgrade of mining
technology and equipment, extraction height of LW working face
has increased to 7 m and annual production of the coal mine has
exceeds 15 million tons of coal. Bulianta colliery and several other
coal mines in Shendong coalfield have become the most productive
underground LW operations in China.
2.2. Geological conditions
According to the data interpretation of drilling core, the outcrop
of strata and the proven geological information of the coalfield, the
stratigraphy of the colliery is estimated. Fig. 2 shows a simplified
distribution of the strata. Main extraction coal seam 1
2
seam is
located in upper part and the other two main seams 2
2
and 3
1
seam are distributed in middle part. The average thicknesses of
three coal seams are 4.1 m, 6.8 m, and 3.2 m respectively. Spacing
of them is approximately 32 m between 1
2
seam and 2
2
seam
and 28 m between 2
2
seam and 3
1
seam, respectively. The min-
ing region is part of the Ordos early-middle Jurassic coal bearing
basin and no big faults are found in the basin. The basin is devel-
oped in the platform on the basis of inheriting type basin in which
the strata lies towards the N20°to 30°W and the tendency is S60°-
70°W. The incline of the strata varies slightly from 0°to 3°and the
floor of coal seam has slight fluctuation with gentle lift in the east.
It is noticeable coal seams in this colliery are closely distributed
and operated under very shallow cover.
2.3. Problem identification
Currently the coal mine is extracting two coal seams, namely
1
2
coal seam and 2
2
coal seam. 3
1
coal seam is on standby.
Fig. 3 shows the overall layout of Bulianta coal mine. The whole
mine is divided into five sections with several longall panels within
each of section. 1
2
coal seam and 2
2
coal seam has been totally
extracted in section one and two. At present section four and sec-
tion five are mining 1
2
coal seam and section three is mining 2
2
coal seam as 1
2
coal seam has been extracted and it is believed
overlying goaf has been interconnected via mining-induced cracks.
Contaminated air is taken out of pit via two main exhaust fans. One
is installed in north exhaust shaft and another one is installed in
south exhaust incline. Fresh air is mainly taken from intake incline
and intake shaft (see Fig. 3). Intake shaft serves to section five and
main intake inclines serves to section three. Fresh air is supplied
from both intake shaft and intake incline for section four.
Since the commencement of extraction of panels in section
three, several serious coal oxidation and self-heating incidents
have occurred and culminated in one open fire incident at
LW22306 working panel (see Fig. 3). It was found the fire origi-
nated from overlying 1
2
coal seam goaf because high concentra-
tion of CO (exceeds 10,000 10
6
) was initially detected from
several boreholes drilled to overlying 1
2
coal seam goaf. The fire
caused closure of the panel for more than six months and costed
hundreds of millions of dollars to quench it by slurry injection
through hundreds of downholes. After undertaking investigation
and incident review, the possible reason of the occurrence of the
fire incident was revealed and can be illustrated in Fig. 4.Asis
widely accepted, a major consequence of coal extraction is ground
subsidence and creation of factures and cracks to the overlying or
underlying strata. In this case, after 1
2
seam was mined, the
induced cracks may have already developed to surface due to shal-
low cover of the coal seam. As 2
2
coal seam was further extracted,
more developed and wider cracks were likely to be induced
because of higher mining height of this coal seam. These channels
are very likely to become interconnected and propagate to surface.
Fig. 5 presents real images of mining-induced cracks developed to
surface. These channels can function as air leakage path from sur-
face to active working face if any pressure differential presents. As
this mine is currently using an exhaust ventilation method, the
pressure of the airflow in working face is possibly much lower than
surface atmospheric pressure. Therefore, the pressure differential
is very likely to draw a certain amount of fresh air from surface
to active longwall face through these channels. In addition, the
immediate roof of Shendong coalfield is very fragile, and as a result,
approximately 0.5 m top coal is reserved to facilitate chock support
and will be left in goaf as LW advances. With continual supply of
fresh air, smoldering of coal developed to an open fire as the heat
generated from coal oxidation is not sufficiently dissipated. The
pressure differential not only aggravates the self-heating process
Fig. 1. Location of Bulianta colliery.
232 Y. Liang et al. / International Journal of Mining Science and Technology 28 (2018) 231–242
of coal but also promotes the ingress of goaf gas into the working
face. The ingress of oxygen deficient and high concentration of CO
gas poses a great threat to the safety of working crew at longwall
face. Many practices have been exercised to control the problem.
One direct solution is to seal these cracks with grout or slurry
injection. However, the solution is still prohibitive for two reasons.
Fig. 2. A simplified stratigraphy of Bulianta colliery.
Fig. 3. Overall layout and ventilation network of Bulianta colliery.
Fig. 4. Schematic illustration of the occurrence of fire incident at Bulianta colliery.
Y. Liang et al. /International Journal of Mining Science and Technology 28 (2018) 231–242 233
One is the cost would be substantial as there are a large number of
cracks required to be treated. Another difficulty is many concealed
cracks are hard to be detected. Therefore, minimizing the pressure
differential between surface and ventilation circuit by improving
ventilation performance would be a promising solution and this
is also an important initiative of this project.
3. Development and validation of ‘‘Ventsim model
A rational solution to the air leakage through mining-induced
cracks is to minimize the pressure differential between surface
and ventilation circuit. To quantify the pressure differential and
to investigate this issue more critically, a ventilation simulation
program ‘‘Ventsim” is used to conduct the case study. ‘‘Ventsim”
is one of the most sophisticated software packages in underground
mine ventilation simulation and is widely used in many Australian
underground mining operations. ‘‘Ventsim” can be utilized to assist
a range of mine ventilation related operations including mine ven-
tilation design, mine network analysis and optimization, prediction
of recirculated ventilation, and economical analysis on mine
ventilation.
3.1. Model development
Geometric model is established by importing Autocad DXF file
which incorporates the real plan map of the whole mine. After a
complex DXF drawing is imported to ‘‘Ventsim”, it is very likely
disconnected or overlapping airways are incorporated within raw
model data. It is necessary to run geometry repair or simplification
before editing any airway. After tiding up the model, the next step
is to assign various parameters including airway profile, geometric
dimension, and frictional factor to every single airway. In addition,
proper resistance factors should also be assigned to ventilation
control devices like airlocks, belt seals, and ventilation air doors.
Editing of airways can be accomplished by accessing into edit
box (see Fig. 6). Then two exhaust fans are installed at main
exhaust incline and north exhaust shaft, respectively. In this appli-
cation two main exhaust fans curve are interpolated by seven data
points based on site measured fan curve data (see Fig. 7). Fig. 8
shows an overview of the established base model.
3.2. Model validation
The base model is validated through two steps. One is via quan-
tity of airflow at most of the important locations and another one is
to check pressure loss along critical ventilation paths. Fig. 9 com-
pares the site measured air flow quantity of critical locations with
the computed data and it can be seen the differential is marginal.
Bulianta colliery undertook a comprehensive ventilation survey
in 2013 and the survey was performed along three airflow paths,
namely LW22305 path, LW12409 path, and LW12519 path. The
three paths represented three panel sections (section three, section
four, and section five, respectively) and the pressure loss is vali-
dated through the three paths. The detail of the validation is only
presented at section three as LWs in section three are the most
risky area in terms of coal spontaneous combustion and ingress
of goaf gas. The follow-up studies will focus on LW22307 path as
well. Fig. 10 is a simplified airflow path through LW22307. Most
fresh air is taken from main intake incline and then directs to
2
2
coal seam intake main. Part of the fresh airflow in 2
2
seam
intake main is taken to section three longwalls via section three
intake main. Then fresh air in LW22307 is supplied through both
LW22307 maingate and LW22308 tailgate. Contaminated air is
drawn to section three exhaust main via LW22307 tailgate and it
is then delivered to 1055 level exhaust main before the contami-
nated air is discharged through fan at south exhaust incline.
Fig. 11 shows the validation of pressure loss along LW22307 path.
It can be seen the overall trend of it resembles that of measured
LW22305 path. It underwent a slight increase of pressure loss com-
paring to that of LW22305 path due to its longer flow path and the
major of pressure loss occurs at exhausting airways. It can be also
observed at working face the pressure loss exceeds 200 Pa and as a
result, the pressure differential between working face and surface
is more than 200 Pa. A large amount of fresh air will be attracted
Fig. 5. Two photographic views of mining-induced cracks at Bulianta colliery.
Fig. 6. Airway parameter editing dialog box in ‘‘Ventsim”.
234 Y. Liang et al. / International Journal of Mining Science and Technology 28 (2018) 231–242
Fig. 7. Fans installation at south exhaust incline and exhaust shaft.
Fig. 8. Overview of the base model in ‘‘Ventsim”. Airways in blue colour represent fresh airflow intake, airways in red colour stands for exhaust airflow, and black colour
denotes sealed airways or virtual fringe of LW goaf.
Fig. 9. Comparison of site measured airflow quantity with computed data at critical airways.
Y. Liang et al. /International Journal of Mining Science and Technology 28 (2018) 231–242 235
into LW working face once mining-induced channels propagate to
surface due to presence of the 200 Pa pressure differential. This
pressure differential also provokes migration of goaf gas into work-
ing face and poses immediate danger to underground miners.
4. Solutions and discussion
4.1. Possible solution one: Modify ventilation network within panel
It has been studied different ventilation modes within LW panel
may induce different pressure differential across LW face and
therefore affect air leakage into goaf [18,20]. To reduce pressure
differential along LW22307 face, the first possible measure is to
modify ventilation network within panel. In this study, total seven
scenarios are proposed and analyzed (see Table 1). The results of
the simulation are presented in Figs. 12 and 13.
Fig. 12 depicts the pressure loss paths of various ventilation
modes within panel. It can be seen the pressure differential is
not able to be eliminated. The pressure differential at face shows
little difference except for two Bleederless ventilation modes. To
meet comfortable working conditions at LW face, no less than
30 m
3
/s fresh air is required in this colliery. As can be seen from
Fig. 13, ‘‘Homotropal U” and ‘‘Bleederless homotropal U” are clearly
not suitable due to low quantity of airflow across LW22307 even
though ‘‘Bleederless homotropal U” can slightly reduce the pres-
sure differential. ‘‘Bleeder return” and ‘‘Double return” ventilation
modes are able to supply sufficient air to working face but the
pressure differentials become greater than that of on-site ventila-
tion mode. ‘‘Bleederless on-site U” can considerably reduce the
pressure differential but the airflow quantity is slightly less than
the requirement. In addition, Bleederless ventilation network
would increase overall resistance and therefore it is inappropriate
to be used for a long period of time. No matter how the mode of
network is modified, the pressure differential would not be elimi-
nated. It is an intrinsic flaw for exhausting ventilation network.
Once mining-induced cracks develop to surface, fresh air will
always be drawn to working face and vice versa for a forcing ven-
tilation system.
4.2. Possible solution two: Pressurize LW panel
To decrease pressure differential and meanwhile to deliver suf-
ficient fresh air to LW working face, a solution called pressurizing
LW panel is proposed. The essence of the solution is to provide a
positive pressure at the start of panel intake to offset the pressure
lost in the past airways. To achieve the positive pressure, an auxil-
iary fan and several ventilation control devices are required (see
Fig. 14). Fig. 14 shows a possible deployment plan to pressurize
LW22307. At the start of the panel intake, an auxiliary fan is
installed at one LW22308 recovery roadway and a ventilation reg-
ulator is installed at one cut-through between LW22306 maingate
and LW22307 tailgate to adjust the pressure and airflow. The ven-
tilation regulator is essentially a ventilation door with adjustable
opening. Fig. 15 shows the pressure loss trend along ventilation
Fig. 10. Simplified airflow path of LW22307.
Fig. 11. Validation of pressure loss along LW22307 airflow path.
236 Y. Liang et al. / International Journal of Mining Science and Technology 28 (2018) 231–242
path with varying resistance factor and fan duty. It is noticeable
pressurizing LW22307 working face would considerably reduce
pressure differential and in addition, ideally true balance could
be acquired by adjusting the resistance factor or fan duty. Fig. 16
shows pressure differential and airflow quantity across
LW222307 with varying resistance factor (R) and auxiliary fan duty
Table 1
Different ventilation modes of LW22307. Blue lines denote fresh airflow, red lines denote contaminated air flow, and green lines denote isolated or sealed airways.
Scenario Illustration Notes
On-site U Fresh air is taken from both LW22307 maingate and LW22308
tailgate while contaminated air return to LW22307 tailgate and then
LW22306 maingate
Homotropal
U
Fresh air is taken from LW22308 tailgate and allows for a split of air
return to LW22307 maingate. The rest of the air returns to LW22307
tailgate after passing working face
Reverse U Inverse to the on-site U mode, fresh airs is taken from LW22307
tailgate and polluted air returns to LW22307 maingate
Bleederless
on-site U
Section three bleeder is sealed. Fresh air is provided from LW22307
maingate and LW22308 tailgate. Most of the contaminated air returns
to section three exhaust main through LW22307 tailgate and rest of
the polluted air returns through LW22308 panel
Bleederless homotropal U
Section three bleeder is still sealed and fresh air is taken to LW22307
via LW22308 tailgate. A part of air is directed to return at the in by
split location and another part of fresh air flow through LW22307 face
to LW22307 tailgate and LW22306 maingate
(continued on next page)
Y. Liang et al. /International Journal of Mining Science and Technology 28 (2018) 231–242 237
(F). It is obvious airflow quantity across LW22307 would increase
with more powerful fan and/or less resistance from ventilation reg-
ulator. The pressure differential evolves from negative to positive
with more powerful fan and/or more resistance from ventilation
regulator, and therefore, ideally a neutral point where there is no
pressure differential can be extrapolated. However, maintaining
LW panel pressure slightly positive is conducive to contain toxic
gas within goaf hence improving working conditions at LW work-
ing face.
4.3. Possible solution three: A booster ventilation system
Instead of locally isolating and pressurizing LW22307 panel,
utilization of a booster ventilation system (Blower-Exhaust) can
initially provide airflow with a positive pressure to overcome the
pressure loss along ventilation path before airflow arrives at
LW22307 working face. The mechanism can be explained in
Fig. 17. For an exhaust ventilation system, neutral point locates
at start of the intake opening and the negative pressure grows
along path and reaches peaks at fan. While for a force system, at
intake fan provides airflow with a positive pressure to overcome
the pressure loss along path and neutral point lies at the end of
the ventilation path. The booster ventilation system takes advan-
tages of the two previous systems and the neutral point can be
adjusted to any point in the middle of the ventilation path. By this
way the pressure differential between LW face and surface can be
significantly reduced. Therefore, the performance of a booster ven-
tilation system is simulated with varying fan duties and the results
are shown in Figs. 18 and 19.
A forcing fan with similar capacity is installed at intake incline
and several other openings are regulated with air locks by increas-
ing magnitude of resistance factor. The solution is demonstrated by
various duties of two fans and the result is illustrated in Figs. 18
and 19. From Fig. 18, it can be observed airflow is given a positive
pressure initially by the forcing fan to offset the pressure loss along
path. The pressure differential significantly decreases if fan dust is
adjusted properly. The neutral point can move to any point by
manipulating fan duties. As can be seen in Fig. 19, it is obvious
the airflow quantity reduces slightly with fan duty dropping but
the pressure differential is very sensitive to fan duty. Ten percent
alteration of fan duty may have dramatically changed the pressure
differential across LW face.
4.4. Discussion
If whole mine ventilation system is governed by only exhaust-
ing or forcing system, the pressure differential would not be elim-
inated no matter how the network is modified. Bleederless
ventilation system may improve the problem but increase diffi-
culty of ventilation and may cause even more pressure differential
at other return airways. Hence, solution one is not suitable at least
for long term operation. Pressurizing LW panel can tackle both the
problems including sufficient air supply and less pressure differen-
tial at LW face. The two parameters are adjustable with varying
auxiliary fan duty and resistance factor of ventilation control
device. The major flaw of this method is as follows. In case auxil-
iary fan fails, goaf toxic gas would migrate instantly to working
face and pose a great danger to working crew. Therefore, extra pre-
cautions and sufficient risk assessments must be put in place
before the implementation of this solution. With more and more
LWs trending to be operated in underlying coal seams, a booster
ventilation system may find its rationale because it is less complex
than locally isolating and pressurizing a LW panel. In addition, this
system has fewer disturbances to production operation comparing
to balancing pressure differential of LW panel one by one. How-
ever, additional capital and operational cost is incurred by instal-
ling and running extra fans and in addition, all mine accesses
Table 1 (continued)
Scenario Illustration Notes
Bleeder
return
Fresh air is taken from LW22307 tailgate and dusty air is directed to
LW22308 tailgate and back return to section through bleeder
Double
return
Similar to bleeder return mode, fresh air is taken from LW22307
tailgate. Air return both from LW22307 maingate and section three
bleeder
Fig. 12. Ventilation pressure loss paths of various ventilation modes.
238 Y. Liang et al. / International Journal of Mining Science and Technology 28 (2018) 231–242
need to install airlocks or conveyor seals. For an on-going LW oper-
ation in multi-seam, pressurizing LW panel might be a better
option to minimize pressure differential and air leakage to goaf.
A booster ventilation system should receive more considerations
upon planning an underground mining operation as this system
has less interference to production operations.
5. Field demonstration
LW22307 of Bulianta colliery commenced the operation on July,
2014. Initially the ventilation system used within the panel is
purely exhausting and no measures have been taken to mitigate
potential fire occurrence although a fire incidence has occurred
in adjacent goaf. With the detection of several possible self-
heating developments and growing severity of ingress of oxygen
deficient gas into working face, the mine determined to pressurize
the LW panel (solution two) to reduce the air leakage at the end of
2014. After reviewing the failure modes and conducting risk
assessments, an auxiliary fan was employed to provide the positive
pressure and associated ventilation regulators were also
Fig. 13. Pressure differentials and airflow quantities across LW22307 of different ventilation modes.
Fig. 14. Schematic illustration of isolating LW22307 panel with positive pressure.
Fig. 15. Pressure loss paths with varying resistance factors and auxiliary fan duties.
Y. Liang et al. /International Journal of Mining Science and Technology 28 (2018) 231–242 239
constructed to adjust the pressure and airflow across working face
(see Fig. 14). Quantity of air leakage and gas composition at LW
face were continually monitored after excising this control mea-
sure. As can be seen from Fig. 20, quantity of airflow across work-
ing face underwent a slight growth and the airflow leakage from
goaf was substantially reduced after balancing the pressure differ-
ential. Fig. 21 shows gas monitoring data of a sampling point at
conjunction of working face and LW22307 tailgate. The oxygen
concentration increased considerably and an inverse trend was
found for the nitrogen concentration. Goaf gas ingress was con-
strained and the working conditions were greatly improved after
the control was exercised. Therefore, locally pressurizing LW panel
is concluded as an effective measure to control pressure differen-
tial and spontaneous combustion hazard for an on-going LW oper-
ation with exhausting ventilation system.
As this mine is an on-going operation, the large ventilation sys-
tem is unlikely to be changed during extraction of LW22307 panel.
Therefore, demonstration of solution three in Bulianta colliery is
impossible to be accomplished. However, this method has been
successfully used in a LW operation in Hunter Valley, Australia
[9]. This coal mine is also operated in multi-seam and shallow
cover, which resembles the LW operation in Bulianta colliery.
Hence if LWs are operating under alike conditions (shallow cover
or multiple coal seams which are closely distributed), a booster
ventilation system could be a better solution as it is less complex
Fig. 16. Pressure differential and airflow quantity across LW22307 with varying resistance factors and auxiliary fan duties.
Fig. 17. Schematic view of pressure loss along path of three ventilation systems.
Fig. 18. Pressure loss paths with varying fan duties of the booster ventilation
system.
240 Y. Liang et al. / International Journal of Mining Science and Technology 28 (2018) 231–242
and more flexible to adjust the neutral point. Clearly this requires
more demonstrations and field trials to benchmark the solution.
6. Conclusions
In this paper, a ventilation simulation package ‘‘Ventsim” is
used to undertake a case study to investigate air leakage problem
in LW operated in multi-seam and under shallow cover. After
development and calibration of the base model, three solutions
are proposed to mitigate the pressure differential issue. The follow-
ing conclusions are drawn:
(1) Pressure differential between LW face and surface is an
intrinsic flaw if a purely exhausting ventilation system is
used in a mine. The pressure differential is not able to be
eliminated no matter how the ventilation circuit is modified.
(2) Isolating and pressurizing active LW panel can provide
working face sufficient amount of fresh airflow and mean-
while reduce pressure differential. This is accomplished by
deployment of an auxiliary fan and several ventilation regu-
lators. Ideally the pressure differential can be removed if the
resistance factors of ventilation control devices and duty of
auxiliary fan are adjusted properly. This solution was justi-
fied in Bulianta colliery.
(3) Extra precautions and sufficient risk assessments must be
put in place before locally pressurizing LW panel because a
large amount of goaf toxic gas would instantly strike work-
ing face and pose a great danger to working crew in case
auxiliary fan fails. This solution is more suitable to an on-
going LW operation.
(4) A booster ventilation system (Blower-Exhaust) can also
reduce the pressure differential by adjusting duties of the
two main fans. Theoretically the neutral point can be dis-
tributed any location along the pressure loss path by manip-
ulating performances of two fans. The solution has been
demonstrated in a LW operation in Australia.
(5) The booster ventilation solution is more recommended for
LW operations in planning as this system has fewer distur-
bances to production operation comparing to solution two.
However additional capital and operational cost is imposed
by installing and running extra fans and in addition, all mine
accesses need to install ventilation control devices.
(6) No doubt ventilation simulation is a powerful tool to study
spontaneous combustion issues in underground coal mines.
Acknowledgment
Financial supports for this work, provided by the University of
Wollongong (4390490), Australia and the Shenhua Group Innova-
tive Technology Research Fund (SHGF-13-07), State Key Laboratory
Cultivation Base for Gas Geology and Gas Control (Henan Polytech-
nic University) Open Funding (WS2017A01) are gratefully
acknowledged.
Fig. 19. Pressure differential and airflow quantity across LW22307 with varying fan duties of the booster ventilation system.
Fig. 20. Comparison of airflow leakage before and after pressurizing LW22307 at
Bulianta colliery.
Fig. 21. Comparison of oxygen and nitrogen concentration at conjunction of LW
and return before and after pressurizing LW22307 at Bulianta colliery.
Y. Liang et al. /International Journal of Mining Science and Technology 28 (2018) 231–242 241
References
[1] Wang HH, Dlugogorski BZ, Kennedy EM. Coal oxidation at low temperatures:
oxygen consumption, oxidation products, reaction mechanism and kinetic
modelling. Prog Energy Combust Sci 2003;29(6):487–513.
[2] Zhang J, Ren T, Liang YT, Wang ZW. A review on numerical solutions to self-
heating of coal stockpile: mechanism, theoretical basis, and variable study.
Fuel 2016;182:80–109.
[3] Liang YT, Tian FC, Luo HZ, Tang H. Characteristics of coal re-oxidation based on
microstructural and spectral observation. Int J Min Sci Technol 2015;25
(5):749–54.
[4] Peng C, Huang FJ, Fu Y. Performance of water-based foams affected by chemical
inhibitors to retard spontaneous combustion of coal. Int J Min Sci Technol
2016;26(3):443–8.
[5] Zhang J, Liang YT, Ren T, Wang ZW, Wang GD. Transient CFD modelling of low-
temperature spontaneous heating behaviour in multiple coal stockpiles with
wind forced convection. Fuel Process Technol 2016;149:55–74.
[6] Wang LK, Ren T, Nie BS, Chen Y, Lv CQ, Tang HY, et al. Development of a
spontaneous combustion TARPs system based on BP neural network. Int J Min
Sci Technol 2015;25(5):803–10.
[7] Song ZY, Kuenzer C. Coal fires in China over the last decade: a comprehensive
review. Int J Coal Geol 2014;133:72–99.
[8] Stracher GB, Taylor TP. Coal fires burning out of control around the world:
thermodynamic recipe for environmental catastrophe. Int J Coal Geol 2004;59
(1):7–17.
[9] MDG-1006. Technical reference for MDG 1006 spontaneous combustion
management guideline. New South Wales: Mine Safety Operations Branch;
2011.
[10] Rosa D, Maria I. Analysis of mine fires for all US underground and surface coal
mining categories: 1990–1999. Pittsburgh: Pittsburgh Research Laboratory,
National Institute for Occupational Safety and Health; 2004.
[11] Singh RVK, Singh VK. Status of mine fire of Jharia coalfield and suggestions for
prevention & control. Coal Min Technol Manage 2004;9(6–8):38–44.
[12] Zhou FB, Ren WX, Wang DM, Song TL, Li X, Zhang YL. Application of three-
phase foam to fight an extraordinarily serious coal mine fire. Int J Coal Geol
2006;67(1–2):95–100.
[13] Gouws MJ, Knoetze TP. Coal self-heating and explosibility. J S Afr Inst Min
Metall 1995;95(1):37–44.
[14] Morris R, Atkinson T. Seam factor and the spontaneous heating of coal. Min Sci
Technol 1988;7(2):149–59.
[15] Massanés MB, Pera LS, Moncunill JO. Ventilation management system for
underground environments. Tunn Undergr Space Technol 2015;50:516–22.
[16] Kursunoglu N, Onder M. Selection of an appropriate fan for an underground
coal mine using the Analytic Hierarchy Process. Tunn Undergr Space Technol
2015;48:101–9.
[17] Ren T, Wang ZW, Cooper G. CFD modelling of ventilation and dust flow
behaviour above an underground bin and the design of an innovative dust
mitigation system. Tunn Undergr Space Technol 2014;41(1):241–54.
[18] Gillies S, Wu HW. Australian longwall panel ventilation practices. In:
Proceedings of 13th coal operators’ conference. Wollongong: University of
Wollongong; 2013.
[19] Ren T, Rao B. Proactive goaf inertisation for controlling longwall goaf heatings.
Procedia Earth Planet Sci 2009;1(1):309.
[20] Smith AC, Diamond WP, Mucho TP, Organiscak JA. Bleederless ventilation
systems as a spontaneous combustion control measure in US coal mines. Bureau
of Mines, IC, Pittsburgh: US Department of the Interior; 1994. p. 9377.
242 Y. Liang et al. / International Journal of Mining Science and Technology 28 (2018) 231–242