Application of Computational Fluid Dynamics (CFD) for Simulation of Acid Mine Drainage Generation and Subsequent Pollutants Transportation through Groundwater Flow Systems and Rivers

Full text

Page 1

5
Application of Computational Fluid Dynamics
(CFD) for Simulation of Acid Mine Drainage
Generation and Subsequent Pollutants
Transportation through Groundwater Flow
Systems and Rivers
Faramarz Doulati Ardejani1, Ernest Baafi2, Kumars Seif Panahi3,
Raghu Nath Singh4 and Behshad Jodeiri Shokri5
1Faculty of Mining, Petroleum and Geophysics,
Shahrood University of Technology, Shahrood,
2School of Civil, Mining and Environmental Engineering, University of Wollongong,
3Shahrood University of Technology
4Nottingham Centre for Geomechanics, School of Civil Engineering,
University of Nottingham,
5Faculty of Mining and Metallurgical Engineering, Amirkabir University of Technology
(Tehran Polytechnic), Tehran
1,3,5Iran
2Australia
4UK
1. Introduction
Many environmental problems associated with the mining industry involve the
understanding and analysis of fluid or gas flow. Typical examples include groundwater
flow, transport of contaminants, heat transfer, explosions, fire development and dust
movements. Both experimental work and numerical models can provide the necessary
information for solution of any particular problem. The long-term pyrite oxidation, acid
mine drainage generation and transportation of the oxidation products are noted to be the
most important problems that can be modelled in order to predict the transport of the
contaminants through groundwater and rivers flow systems, to interpret the geochemistry
and achieve a better understanding of the processes involved.
The use of computational fluid dynamics (CFD) to simulate flow problems has risen
dramatically in the past three decades and become a fairly well established discipline shared
by a number of engineering and science branches. Associated with the widespread
availability of high performance and advanced computers and computational methods, CFD
is rapidly becoming accepted as a cost-effective design and predictive tool. Numerical
solution methods may be used for CFD analysis for the simulation of fluid flow, heat and
mass transport problems when it is expressed in terms of partial differential equations

Page 2

Computational Fluid Dynamics Technologies and Applications

124
(PDEs). Recent improvement of CFD codes enables researchers to visualise easily the local
velocity, temperature, concentrations of solutes and pressure fields in a domain by means of
graphic facilities (Edwards et al., 1995).
CFD is a very powerful tool and applicable to a wide range of industrial and non-industrial
areas, including aerodynamics of aircraft, automotive, pollution control, power plant, turbo
machinery, electrical and electronic engineering, civil engineering, hydrology and
oceanography, meteorology and medical science (Versteeg & Malalasekera, 1995).
CFD has been recently used in many applications relating to environmental studies as a way
of estimating impacts and developing control strategies for the long-term impacts of mining
on the environment and many other activities. The low response time and cost associated
with the simulations compared to experiments are the main benefits provided that adequate
accuracy may be obtained by the computer erally used in CFD can be classified as finite
difference method (FDM), finite element method (FEM), finite volume method (FVM), or
spectral methods. The choices of a numerical method and a gridding strategy are strongly
interdependent to solving the CFD problems. For example, the use of finite difference
method is typically restricted to structured grids (Harvard et al., 1999).
Some researchers used the finite element method to solve the PDEs for modelling of solute
transport through groundwater flow systems (Pickens & Lennox, 1976; Pinder, 1973;
Rabbani & Warner, 1994; Wunderly et al., 1996). Furthermore, some researchers used the
finite volume method to solve the PDEs for modelling of pollutant transport through rivers
flow systems (Ani et al., 2009; Kachiashvili et al., 2007; Zhi-Qiang & Hoon-Shin, 2009). Green
and Clothier (1994) used the PHOENICS-code (Spalding, 1981) incorporating the finite
volume method to simulate water and solutes transport into unsaturated soils. Edwards et
al. (1995) have noted the applicability of the CFD analysis in mine safety and health
problems such as methane control, gas or coal outbursts, dust suppression and explosions.
Balusu (1993) developed a numerical model using a CFD code, FIDAP, to simulate airflow
patterns and the respirable dust concentration at a longwall face in underground coal mines.
The analysis of contaminant transport in groundwater systems using finite volume
techniques have been carried out by Putti et al., (1990) and Binning & Celia (1996). Singh &
Doulati Ardejani (2004) developed an one-dimensional numerical finite model using
PHOENICS code to simulate long term pyrite oxidation, acid mine drainage generation and
transportation of the oxidation products through the backfills of an open cut mine.
This chapter presents the application of computational fluid dynamics (CFD) using
PHOENICS code to simulate pyrite oxidation, acid mine drainage generation and
subsequent pollutants transportation through groundwater flow systems and Rivers. The
simulations have been conducted by developing one- and two-dimensional models.
2. Computational fluid dynamics
Computational fluid dynamics (CFD) is defined as the analysis of systems involving fluid
flow, heat transfer and mass transport and associated phenomena such as chemical reactions
using computer-based simulation. To predict the way in which a fluid will flow for a given
situation, a mathematical analysis of the fluid flow has to be made to formulate the
governing equations of flow, and the CFD code enables users to calculate numerical
solutions to these equations. To produce a solution, these equations have to be transformed
into numerical analogues using discretisation techniques such as finite difference method
(FDM), finite element method (FEM) and finite volume method (FVM).

Page 3

Application of Computational Fluid Dynamics (CFD) for Simulation of Acid Mine Drainage
Generation and Subsequent Pollutants Transportation through Groundwater Flow Systems…

125

Start


End
Yes
No

Improve mesh quality, inspect boundary conditions and fluid
properties, and inspect time steps, iterations, solution method,
and parameters controlling convergence




Numerical solution of the problem
Result evaluation
Pre-processing phase
Simulation phase
Post-processing phase
Specification of boundary conditions and
fluid properties
Construction of the problem domain
and grid generation
Deciding about the problems to be
analysed and deriving a conceptual model

Are predicted results and
measured data similar?


Fig. 1. Main stages in a CFD simulation
CFD codes such as FLUENT, PHOENICS, FLOW3D and FIDAP are now widely available
commercially, each with its own particular set of features to deal with fluid flow problems.
Edwards et al., (1995) and Doulati Ardejani (2003) has given a comparison of some
commercial CFD codes. In general, a flow analysis with CFD codes can be divided into three
main phases:

Page 4

Computational Fluid Dynamics Technologies and Applications

126
•
Pre-processing phase includes the input of a flow problem to a CFD package using an
operator-friendly interface and the subsequent transformation of this input data into an
appropriate form for use by the solver. This phase mainly consists of the definition of
the geometry of the problem of interest, mesh generation, specification of physical
properties of the fluid and appropriate boundary conditions. An unknown flow
variable such as velocity, pressure and temperature is solved at nodes inside each cell.
The accuracy of a CFD solution is governed by the number of cells in the grid. The
accuracy is improved by increasing the number of cells. Optimal meshes are often non-
uniform. A finer mesh is constructed in areas where large variations occur from point to
point and a coarser grid is used in regions with relatively small change (Versteeg &
Malalasekera, 1995).
Simulation phase including solution of the governing equation for the unknown flow
variable.
Post-processing phase including domain geometry and grid display, vector plots, surface
plots, x-y graphs, line and shaded contour plots and animation for dynamic result display.
•
2.1 General stages of a CFD analysis
To produce a CFD simulation, a number of key steps should be followed in order to
generate an exact picture of a particular problem. Figure 1 shows the main steps for a CFD
analysis.
2.2 PHOENICS as a CFD software
The PHOENICS program has a few different modules to perform all these three phases of
flow analysis, namely SATELLITE, EARTH, and post-processing facilities including VR
VIEWER, PHOTON, AUTOPLOT and RESULT (Spalding, 1981). Figure 2 shows the
solution performance in the PHOENICS package.
3. Finite volume method (FVM)
FVM (sometimes called the control volume method) is a numerical technique for solving
governing equations of fluid flow and mass transport that calculates the values of the
conserved variables averaged across the control volume. The calculation domain is broken
down into a finite number of non-overlapping cells or control volumes such that there is one
control volume surrounding each grid point. The partial differential equation is integrated
over each control volume. The resulting discretisation equation containing the values of a
variable φ for grid points involves the substitution of a variety of finite-difference-type
approximations for the different terms in the integrated equation representing flow and
transport processes such as convection, diffusion and source terms. The integral equations
are therefore, converted into a system of algebraic equations. The algebraic equations
obtained in this manner are then solved iteratively.
It is possible to start the discretisation process with a direct statement of conservation on the
control volume. Alternatively we may start with the differential equation and integrate it
over the control volume. The numerical solution aims to provide us with values of φ at a
discrete number of points in the flow domain. These points are called grid points, although
we may also see them referred to as nodes or cell centroids, depending on the method
(Murthy, 2002).

Page 5

Application of Computational Fluid Dynamics (CFD) for Simulation of Acid Mine Drainage
Generation and Subsequent Pollutants Transportation through Groundwater Flow Systems…

127
Drawing tabular
plots of unknown
variables
Plot x-y graphs for
comparison of
PHOENICS
solutions with
field data

Viewing
streamlines,
vectors,
iso-surfaces,
contour plots

Viewing grid
streamlines, vector
plots Surface plots
contour plots
Discretisation of finite
volume equations, iterative
solution of finite volume
equations, output of results

RESULT
AUTOPLOT

PHOTON
VR VIEWER

Post-processor

EARTH
PHOENICS
(solver)

Q1 file

Pre-processor
GROUND (routine)
EARDAT file

Definition of geometry, grid
generation, selection of the
physical and chemical
phenomena that need to be
modelled, definition of fluid
properties, specification of
boundary conditions










Fig. 2. The procedure of solution in PHOENICS (CHAM, 2008)
FVM has become popular in CFD as a result, primarily, of two advantages. First, they
ensure that the discretisation is conservative, i.e., mass, momentum, and energy are
conserved in a discrete sense. While this property can usually be obtained using a finite
difference formulation, it is obtained naturally from a finite volume formulation. Second,
FVM does not require a coordinate transformation in order to be applied on irregular
meshes. As a result, they can be applied on unstructured meshes consisting of arbitrary
polyhedral in three dimensions or arbitrary polygons in two dimensions. This increased
flexibility can be used to great advantage in generating grids about arbitrary geometries
(Harvard et al., 1999).
Although the finite element method is mainly used for the simulation of the contaminant
transport problems through groundwater flow systems (see Barovic & Boochs, 1981; Bignoli