Implementation of a Finite Element Model to Generate Synthetic data for Open Pit’s Dewatering

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 04 Issue: 12 | Dec-2017 www.irjet.net p-ISSN: 2395-0072
© 2017, IRJET | Impact Factor value: 6.171 | ISO 9001:2008 Certified Journal | Page 751
Implementation of a Finite Element Model to generate synthetic data
for open pit’s dewatering
Sage NGOIE1, Adalbert Mbuyu2, Jean Felix Kabulo3, Albert Kalau4
1Philosophiae Doctor, IGS, University of the Free State, Republic of South Africa
2Junior Lecturer, Dept. of Geology, University of Kolwezi, Dem. Rep. of Congo
3Junior Lecturer, Dept. of Geology, University of Likasi, Dem. Rep. of Congo
4Lecturer, Dept. of geology and mining engineering, ISTA Kzi, Dem. Rep. of Congo
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Synthetic data have long been employed in
geohydrology for model development and testing. The
objective of this chapter is to generate a synthetic data set of
geohydrological responses in an open pit environment. This
data set will be used to represent real data that could be
recorded at an open pit mine.
The synthetic data set is generated by using a numerical
model. In the model, different pumping scenarios are
considered. The model uses nine observation points
(piezometers) and three, six, nine and 12 pumping wells in the
four different pumping scenarios. The purpose of the pumping
wells is to dewater the open pit under different pumping
conditions. The response of the aquifer to these different
pumping scenarios is examined.
Key Words: Finite Elements Methods; Synthetic data;
Dewatering; Open pit mine.
1. INTRODUCTION
Synthetic data have long been employed in geohydrologyfor
model development and testing. The objective of this paper
is to generate a synthetic data set of hydrogeological
responses in an open pit environment. This data set can be
used to represent real data that could be recordedatanopen
pit mine.
The synthetic data set is generated by using a numerical
model. In the model, different pumping scenarios are
considered. The model uses nine observation points
(piezometers) and three, six, nine and 12 pumping wells in
the four different pumping scenarios. The purpose of the
pumping wells is to dewater the open pit under different
pumping conditions. The response of the aquifer to these
different pumping scenarios is examined.
2. MODEL DESCRIPTION
Aquifers are complex and not often directly visible. For
better understanding these aquifers for modellingpurposes,
they have to be represented by simplified versions in the
form of conceptual models. The conceptual model may
influence the choice of numericalmethodusedforsimulating
the behaviour of the aquifers. For example, a conceptual
model with complex aquifer boundaries may have to be
modelled using FEM instead of FDM, since the rectangular
cells used in FDM do not allow for adequate refinement of
the modelling grid.
If the conceptual model gives an accurate representation of
the real aquifer, the numerical model will also be more
accurate (Anderson and Woessner, 1992).
The conceptual model of the current investigation includes
information on the pit geometry, geomorphology, rainfall,
surface water bodies, and aquifer units as derived from the
geological layers.
2.1 Geometry of the modelled open pit mine
The modelled open pit mine is assumed to be excavated in a
sedimentary deposit with the top and bottom elevations at
1 250 mamsl and 1 166 mamsl, respectively. The plan view
of the pit can be compared to a smooth closedcurve,whichis
symmetric about its centre with the transverse, and
conjugate diametersof 880 m and 370 m, respectively(refer
to Fig-1).
The mine is exclusively excavated in the first geologicallayer
(dolomite), which is 160 m thick. The pit is assumed to be
excavated in an unconfined aquifer, since it is assumed that
water in the voids and fractures of the dolomite is in contact
with the atmosphere and is therefore under atmospheric
pressure.
The vertical distance between the highest point on the
perimeter of the pit and the pit floor is 84 m. The pit hasnine
benches with an average bench height of 9.3 m (see fig-2).
2.2 Topography and hydrography of the modelled
area
The general topography of the region is gentle. The pre-
mining topography shown in fig-3 is an existing topography
of a tropical area in the Democratic Republic ofCongo(DRC).
This particular area was chosen because of the variation in
the surface topography (higher elevations in the
southwestern parts and lower elevationsinthenortheastern
parts). Since groundwater elevations generally emulate the
surface topography, topographic gradients are often also
associated with hydraulic gradients and thus with
groundwater movement (Haitjema and Mitchell-Bruker,
2005). In this research, it is, therefore, assumed that the

groundwater flows in the direction of the topographic
gradient.
Fig - 1: Plan view of the open pit of the model
Fig - 2: Cross-section through the open pit of the model
Fig - 3: Pre-mining topography of the model
The open pit is located on the watershed between two
catchments, each catchment drained by a river flowingfrom
south-west to north-east. These two rivers are simulated as
constrained head boundaries. The constrained heads were
assigned valuesthat ensure a gradient in the direction ofthe
river flow (down-gradient, according to the ground
topography). It wasfurther assumed that the waterfromthe
river infiltrates the aquifer at a constant rate of 30 m3/h.
This latter infiltration rate was chosen because it was
observed by Norris (1983) in the Scioto River in south-
central Ohio (from 0.06 to 0.19 million gallons per day for
one acre) and also in the Dipeta River in the Democratic
Republic of Congo (30 m3/h for a river with a length of
1.3 km and a width of 3 m).
Based on the topography of the model, some of the surface
runoff drains directly into the open pits. Such runoff water
could pose problems to the management of surface water at
real mines. However, in this paper, surface runoff water
entering the pit will not be considered in the synthetic
model, since the aim is to model pit dewatering by using
abstraction wells.
2.3 Geometry of the groundwater model
The model domain is 1 126 m long, 574 m wide and 240 m
high. Asshown in fig-4 , the geology of the regionis assumed
to be sub-horizontal, consisting of only two layers, namely:a
dolomite layer (160 m thick), overlying a shale layer (80 m
thick). No prominent tectonic features, such as faults, occur
within the model domain. The open pit mine is excavated
exclusively in the dolomite layer to depth of 84 m.
2.4 Hydraulic parameters
In fig-4 , the spatial distribution of the hydraulicparameters
is shown. It is seen that these parametersare directlyrelated
to the geological units, and that these parameters do not
vary within the geological units. As indicated in Table-1, the
hydraulic conductivity is the only hydraulic parameter that
differsfor the two layersin the model. It is also seen that the
vertical hydraulic conductivities (KZZ) of the layers are
significantly smaller than the horizontal hydraulic
conductivities (KXX and KYY). These hydraulic conductivity
values are based on the work of Morris and Johnson (1967)
who conducted studies on the hydraulic parameters of
several rock types. The specific storages, and the specific
yields of the two layers are taken as the default values for
dolomites and shales, as defined in the software.
Table 1: Hydraulic parameters of the synthetic model
2.5 Recharge
The main recharge of the aquifer is through rainfall. The
mean annual rainfall (MAR) in the modelled area is assumed
to be 1 200 mm, corresponding to the rainfall figures in a
tropical climate. A large percentage of the rainfall flows to
rivers as runoff. In Feflow, rainfall is modelled as aerial
groundwater recharge by using sink/source formulations.
Recharge values for carbonate rocks such as limestone and
dolomite range from three to 10% (MWR, 2009). This
boundary condition was applied to the top of the first
geological layer of the numericalmodel.Rechargecalculation

is then performed automatically according to the
hydrogeological parameters (permeability, storativity, etc.)
of the layers in the model.
2.6 Dewatering and observation wells
While performing the dewatering simulations, the behavior
of the aquifer will be observed at nine observation points
(OBS_1 to OBS_9) spatially distributed as shown in fig-5.
Monitoring point OBS_9 is used to evaluate the water
elevation according the bottom of the pit because it is
located right in the middle of the pit. Four dewatering
scenarios will be run with three, six, nine and twelve
dewatering wells. The dewatering wellsare numberedBH_1
to BH_12.
Fig - 4: The synthetic model set up with hydraulic
conductivity distribution
Fig - 5: Spatial distribution of observation points and
dewatering well
2.7 Boundary conditions
The base of the model (the bottom of the shale layer) is
assumed impermeable. The numerical model used in this
study incorporates the following boundary conditions:
- Recharge (3 to 10% of the MAR) is represented by
areal fluxes applied at the top slice of the synthetic
model (the top of the dolomite layer);
- The well boundary conditions applied to the
dewatering wells describes the impact of water
abstraction at a single node in m3/d;
- Themodel assumesthat the riversandgroundwater
are in dynamic connection.Hydraulicheadboundary
condition with flow-rate constraints were used for
definition of rivers.
- Constant head boundary conditions are assigned to
the boundariesof the modeldomain.Theseconstant
heads were determined by considering the surface
topography at the boundaries.
3. MODEL DEVELOPMENT
3.1 Model package
The finite element software Feflow® v6.2 from DHI-WASY
wasused to simulate the behavior of groundwater. Feflowis
a three dimensional finite element package able to simulate
unsaturated and saturated flow. It also has a mesh
generation method which allows for flexible and quick
editing of the model. This code allows rapid execution,
development and analysis of the model (Diersch, 2004).
The capabilitiesof Feflow to interact with ArcGIS (ESRI)and
spreadsheets is one of the important features of this
software. Its flexibility is the reasons why it is one of the
modelling packagespreferred by scientists (Knapton,2009).
3.2 Spatial discretization
The discretization of the model is done with the Feflow®
package. Meshes are generated by applying the automatic
triangle algorithm (Shewchuk, 2002). This algorithm is very
versatile and extremely fast, and can deal with complex
geometrical setups of polygons, lines, and points.
The mesh of the current model has 169 386 elements with
84 873 nodes. The regional mesh was refined in the
synthetic model using the Mesh Geometry Editor. The
resulting mesh used in the modelling is presented in fig-6.
3.3 Model settings
The synthetic model assumed saturated and unconfined
conditions, and also assumed only groundwater flow (not
masstransport). The total duration of the modelling wasfor
a period of 5 months, from 01/01/2015 to 01/06/2015.
Fig - 6: Finite element mesh used in the synthetic model

3.4 Dewatering strategy and model results
a. Pre-mining groundwater levels
The natural pre-mining hydraulic gradient in the vicinity of
the pit is shown in fig-7. It can be seen that under natural
conditions, the groundwater generally flows from south-
west to northeast within the model domain.
Fig - 7: Pre-mining hydraulic heads within the model
domain
b.Static groundwater levels after mining
After excavating the pit, and allowing equilibrium (static)
conditions to be reached, the bottom level of the mine is
located at an elevation of 1 166 mamsl, while the highest
hydraulic head within the model domain is at 1 200 mamsl.
Fig - 8 shows the water elevations in all the observation
wellsunder static (no groundwater abstraction) conditions.
Asexpected, all the wellsdisplay constant heads (horizontal
lines), because, under static conditions, thewatertableisnot
impacted by dewatering. The difference betweenthehighest
(OBS_1) and lowest (OBS_9) hydraulic heads at the
observation wells is 8 m within the boundary of the model,
as shown in fig-8.
Fig - 8: Modelled hydraulic heads of the observation wells
when no abstraction takes place
Under conditions of no abstraction, a pit lake occurs with a
water elevationof 1 195.58 mamsl (adepthofapproximately
30 m as measured from the bottom of the pit), as shown in
Fig - 9.
Fig - 9: East-west profile of the pit for the model at initial
conditions
c. Dewatering using three abstraction wells
One or more dewatering strategiescould be appliedtolower
the water level. In this research, vertical pit boreholes are
used in the dewatering strategy. Each borehole pumps at a
constant rate of 300 m3/h. Four scenarios, taking into
account three, six, nine and 12 dewatering wells, runningfor
a 5-month abstraction period, were considered during the
modelling of pit dewatering.
To lower the water level, the first scenario consists of
installing three wells (BH_1 to BH_3) along the iso-
potentiometric line on the eastern ramp of the open pit in
order to decrease the water inflow to the mine. After
pumping commences on 01 January 2015, the water
elevations at all the monitoring points decrease due to the
formation of cones of depression around the abstraction
wells (refer to fig-10). However, from 17 March 2015
(approximately two and a half months after pumping
commenced) all observation points indicate stable water
levels, as equilibrium conditions are attained.
After simulating three dewatering wells pumping for 5
months, the water level in the pit lake decreased to an
elevation of 1 191.6 mamsl, as shown by the fig-11. During
the initial conditions, the water level at monitoring point
OBS_9 was 1 195.6 mamsl. After simulating three wells
pumping for 5 months, the water level in the lake dropped
by approximately 4 m. The water in the pit lake then had a
depth of 26 m.
d.Dewatering using six abstraction wells
With six dewatering wells (BH_1 to BH_6) in the model
pumping for 5 months, the depth of the water in the pit lake
was reduced to 7.8 meters (observation point OBS_9 in the
pit had a water elevation of 1173.8 mamsl). Thewaterlevels
in the observation wells during the 5-month period are
shown in fig-12 while a cross-section through the pit
showing the groundwater elevation is presented in fig-13. It
can be seen that the pit is still flooded after 5 months of
pumping from the six dewatering boreholes. Under these
circumstances, it would therefore be difficult to re-start
mining operations unless some additional dewatering wells
are installed.

for the model using three dewatering wells
Fig - 11: East-west profile of the pit for the model using
three dewatering wells
for the model using six dewatering wells
Fig - 13: East-west profile of the pit for the model using six
dewatering wells.
e.Dewatering using nine abstraction wells
The third scenario takes into account nine dewatering wells
(BH_1 to BH_9). Abstracting water from these wellsovera5-
month period reduced the water level of the pit lake (as
observed at monitoring point OBS_9) to 1166.6 mamsl. The
graphs of the water levels in the observation wells (fig-14)
show that the impact of the dewatering for 5 months is
significant, with a steep cone of depression around the
boreholes, but that the water level in the pit is not reduced
enough to allow the extraction of minerals under dry
conditions.
The water depth in the pit lake has now beenreducedtoonly
0.6 meters (see fig-15). Although this water level is low, it is
still not possible to extract minerals without further
dewatering procedures.
for the model using nine dewatering wells
Fig - 15: East-west profile of the pit for the model using
nine dewatering wells
f. Dewatering using 12 abstraction wells
Since nine abstraction wellswere not able to dewater thepit
completely, another modelling scenario with more
abstraction wells is required. This scenario takes into
account 12 dewatering wells to lower the water level up to
one bench lower than the bottom of the pit. After 5 months
of dewatering, the water level at OBS_9 in the pit stabilizesat
1151.2 mamsl (refer to fig-16). This elevation is 14.8 m
below the bottom elevation of the pit floor.

A cross-section through the pit after 5 months of pumping
with 12 abstraction wells is shown in fig-17 The
groundwater level is now below the bottom of the pit and
the extraction of minerals can commence.
for the model using 12 dewatering wells
Fig - 17: East-west profile of the pit for the model using 12
dewatering wells
4. DISCUSSION AND CONCLUSION
In
Fig - 18, the results of the different modelling scenarios are
summarized by plotting the pit water level (OBS_9) against
the number of abstraction wells used in the dewatering
strategy. From this figure, it is clear that the different
modelling scenarios had significantly different impacts on
the groundwater and pit water levels. The modelresultsalso
showed under which conditions complete dewatering ofthe
pit will be attained.
The model results provide valuable datasets of hydraulics
heads measured against time for the different pumping
scenarios. These synthetics data sets can be used for
numerous purpose in hydrogeology and geotechnical
engineering such as training, testing and validating Artificial
Neural Networks for mining operation purposes.
Fig - 18: Summary of the dewatering impact relative to the
bottom of the pit
REFERENCES
[1] F. Abdulla, M. Al-Khatib, Z. Al-Ghazzawi,
“Development of groundwater modeling for the
AZROQ basin”, Environ Geol. 40(1/2):11–18, 2000.
[2] M. Anderson and W. Woessner, “Applied
Groundwater Modeling, Simulation of Flow and
Convective Transport”, Academic San Diego,
California, 1991.
[3] K. Anthony, “An integrated surface – groundwater
model of the Roper River Catchment, Northern
Territory”, dept. of Natural resources, Env. Art and
Sport, Australian gov. 69 p., 2009.
[4] M. Bakker, “Simulating groundwater flow in multi
aquifer systems with analytical and numerical
Dupuit models”, J. Hydrol. 222, 55-64, 1999.
[5] Carslaw and Jaeger, “Conduction of Heat in Solids.
Oxford University”, 1959.
[6] P. Domenico and F. Scwartz, “Physical and chemical
hydrogeology”, second Edition, Wiley, 1998.
[7] FemLab User Guide, “An introduction to FEMLAB’s
Multiphysics modeling capabilities”, Burlington,
40p. 2015.
[8] P. France, “Finite element analysis of three
dimensional groundwater flowproblems”,J.Hydrol.
, 21, 381-398, 1974.
[9] M. Heinl and P. Brinkmann, “A groundwater model
of the Nubian aquifer system”, Hydrol. Sci. J.,
34:425–447, 1989.
[10] H. Karahan and M. Ayvaz, “Transient
groundwater modeling using spreadsheets”, Adv.
Eng. Software: 36, 374-384, 2005.

[11] H. Karahan and M.Ayvaz,“Time-Dependent
Groundwater Modeling Using Spreadsheet”,
Computer applications in engineering education:
13, 192 -199, 2005.
[12] L. Kipata, “Brittle tectonics in the Lufilian
foldand- thrust belt and its foreland An insight into
the stress field record in relation to moving plates
(Katanga, DRC)”, PhD thesis KU Leuven, faculty of
science, p.160, 2013.
[13] D. Morris and A. Johnson, “Summary of
hydrologic and physical properties of rock and soil
materials asanalyzed by the HydrologicLaboratory
of the U.S. Geological Survey”, U.S.GeologicalSurvey
Water-Supply Paper 1839-D, 42p., 1967.
[14] J. Shewchuk, “Delaunay refinement
algorithms for triangular mesh generation”,
Computational geometry: theory and application,
Amsterdam, Volume 22: 21-74, 2002.
[15] P. Wang and M. Anderson, “Introductionto
Groundwater Modeling”, W. H. Freeman and
Company, San Francisco. 237 pp., 1982.
[16] P. Wang and Z. Chunmaio, “An efficient
approach for successively perturbed groundwater
models”, Adv. Water Res., 21: 499-508, 1998.
[17] T. Winter, “The concept of hydrologic
landscapes”, J. Am. Water Resources Assoc.:
37:335–349, 2001.
[18] W. Woessner and M. Anderson,“Thehydro-
malapropos and the ground water table”, Ground
Water, vol. 40, no. 5, p. 465, 2002.
BIOGRAPHIE
Sage Ngoie was born in Democratic
Republic of Congo. He obtained a degree
in Geology and a Master’s Degree in
Geotechnical and Hydrogeological
Sciences. He holdsa PhD in Geohydrology
from the University of the Free State in
South Africa where he specialized in
Artificial intelligence and mathematical
modeling applied to groundwater.

Implementation of a Finite Element Model to Generate Synthetic data for Open Pit’s Dewatering

More Related Content

What's hot (19)

Similar to Implementation of a Finite Element Model to Generate Synthetic data for Open Pit’s Dewatering (20)

More from IRJET Journal (20)

Recently uploaded (20)

Implementation of a Finite Element Model to Generate Synthetic data for Open Pit’s Dewatering