Available online at www.sciencedirect.

com

ScienceDirect
Energy Procedia 36 (2013) 163 – 168

Simulation of water filtration in porous zone based on Darcy's

law
Mohammed Amine Kendoucia,b a*, Benali Kharroubia , Rachid Khelfaouib , Ali
Bendidaa , Brahim Dennaib and Abdelhak Maazouzib
a
Laboratory of Hydrology and management of Water resources (HYDRE), University of sciences and technology, Oran, Algeria.
b
Laboratory ENERGARID, university of Bechar, Algeria

Abstract

Modeling of water filtration through a porous zone subjected to a steady flow with free surface. Under certain
assumptions, it is possible to use a predictive process for understanding the mechanisms of filtration of water in a
porous zone such as sand dune. The approach is based on the formulation of Darcy porous zone-grained. We
examined the validity of Darcy model with the experimental results of a test bench realized in the laboratory.

This work was done using the software Fluent family CFD (computational fluid dynamics), which is a commercial
code programmed in C and based on the numerical method of finite volumes. For numerical modeling was used both
CFD softwares: Gambit (the mesh) and Fluent (solver and postproceseur). Gambit and Fluent are software
simulations for 2D or 3D fluid mechanics from the construction of the mesh with Gambit in solving Navier-Stokes
equations and post-processing with Fluent.

The experimental device used for the testing of filtration. The driver is a glass column of 5 cm diameter and 100 cm
height and cylindrical shape. The effective height of the filter bed 60cm, 30 cm is used for filtered water which is
kept constant throughout the experiment, the results obtained noted that the difference between the experimental and
simulated flow is order of 10-4.

The results allow us to conclude that the filtration rate at operating pressure depends on the turbidity of the water
directly affects the porosity of the filter and consequently its permeability.

Keywords: Law of Darcy, Fluent, porous zone, Permeability, Porosity, Flow

* Corresponding author. Tel.: +213790594126; fax: +213-764937.

E-mail address: amine.ken37@yahoo.fr

1876-6102 © 2013 The Authors. Published by Elsevier Ltd. Open access under CC BY-NC-ND license.
Selection and/or peer-review under responsibility of the TerraGreen Academy
doi:10.1016/j.egypro.2013.07.019
164 Mohammed Amine Kendouci et al. / Energy Procedia 36 (2013) 163 – 168

1. Introduction

The process of filtering the water through a filter bed of sand is much more complex than believed
since several factors may occur during filtration, thus physicochemical and mineralogical characteristics
of the bed, physicochemical properties and the nature of water treated.
Sand composition, the stability of the structure and chemistry of the water flowing through the pores of
the sand are key factors in the process of water filtration. The adsorption / desorption and precipitation /
dissolution are the types of interaction that can occur between the solid (sand) and an element dissolved in
the liquid. The adsorption and precipitation imply the disappearance of the chemical species without
power differential between the two phenomena; it allows us to include only the two phenomena under the
term sorption. Therefore, following the elimination of an element dissolved in water after passage through
the solid material (sand), adsorption and precipitation are the mechanisms that dominate the process.

2. Materials and methods

2.1. The porous medium

In South West Algeria expanses of sand dunes are known as 'Erg', and this area is covered with several
Ergs are: The Grand Erg Occidental, located in the north west of the Sahara 600km south of the
Mediterranean, along the 500km WSW to ENE 150 to 200 km wide and covers an area of nearly 100,000
km2 [1]. The sands are soft rock allochthons rocks derived from pre-existing flush consolidated altered in
a site. These stretches of sand known to give Ergs a landscape that characterizes the Sahara and gives it a
particular site, and to value this type of unique landscape, the media coverage of these sites is a great
contribution to the economy of the region. The Sahara sand dunes extend over large areas and seem to be
an obstacle in the development of urban infrastructure Saharan regions and an ecological barrier called
desertification. The study of these sites also has a sandy impact mapping, prospecting therefore bedrock is
difficult to detect [2] [4].
Alger
Oran

Béchar

Béni Abbès

Fig.1 Location of the town of Beni Abbes

The Ergs hidden lakes, reservoirs of oil, gas and even extensions heel high economic interest, thus
mapping and sampling hidden by sand provide additional data geo-economics. The oasis city of Beni
 Mohammed Amine Kendouci et al. / Energy Procedia 36 (2013) 163 – 168 165

Abbes claimed Saoura, is found about 1200 km from the capital Algiers. It is located 250 km south from
the capital Bechar (Fig.1).
The type of filter media used in this work, it is the blessed sand dunes abbots come directly from the
Grand Erg Occidental, the rate of insoluble material is about 97% (mainly composed of quartz), materials
weak organic and particle size analysis shows that the soil is very uniform [6].

Table 1. The physical characteristics of the porous medium (sand)

Diameter effective Uniformity Permeability density Porosity

( mm) coefficient (m/s) 10-4 kg/m3 %
0.17 1.76 7,26 2.63 42.01

The type of sand studied belongs to the category of fine sand , that is confirmed by its low porosity
42%, can also be quantified by the effective diameter d10 is the order of 0.17mm, the fineness modulus
(FM) 2.16, and their very low permeability is of the order of 7.26 10-4 ms-1 class according to [3] [5] and is
of good permeability as observed in Table 1 the uniformity coefficient is less than 2 we can conclude that
the sand has a uniform particle size, by convention, if CU is between 1 and 2, the sand is called uniform.

2.2. Experimental

The experimental device used for the testing of filtration. The driver is a glass column of 5 cm diameter
and 100 cm height and cylindrical shape. The actual height of the filter bed and 60cm, 30 cm is used for
the filtered water (distilled water), which is maintained constant throughout the experiment in order to
keep the hydraulic load on the same filter bed of sand.

2.3. The experimental results

The figure below shows the variation of throughput as a function of time for the filter bed:

0,001

0,0008
flow(l/s)

0,0006

0,0004

0,0002

0
7 12 16 20 24 28 33 39 45 49 52 55 57 59
Time(mn)

Fig.2 Evolution of flow as a function of time

166 Mohammed Amine Kendouci et al. / Energy Procedia 36 (2013) 163 – 168

The results are obtained for a water depth equal to 30cm, according to the hydrodynamic study of the
porous medium we obtained a rate of 0.95 10-6 m3 / s and permeability was recorded which is of the order
of 7.26 10-4 m / s this is justified as it has been noted in Table 1.

3. Simulation

Our work has been done using the software Fluent family CFD (computational fluid dynamics), which
is a commercial code programmed in C and based on the numerical method of finite volumes.
For numerical modeling was used both CFD software: Gambit (the mesh) and Fluent (solver and
postprocessor). Gambit and Fluent are software simulations for 2D or 3D fluid mechanics from the
construction of the mesh with Gambit in solving Navies-Stokes equations and post-processing with
Fluent.
"Fluent" is a computer code that simulates fluid flows with and without heat transfer in complex
geometries. It can solve flow problems with structured and unstructured meshes, some of the complexity
of the geometry, with relative ease [7].

3.1. Data of the filter bed

As a first step, create the geometry of the object of study, as shown in the following figure:
The characteristic dimensions of the porous medium model are identical to the studied experimental
device. A two-dimensional model is considered.

Input 103000
Pa

Porous media 60
cm
Diameter
5 cm

Output 100000
Pa
Fig.3 Diagram of the simulated
 Mohammed Amine Kendouci et al. / Energy Procedia 36 (2013) 163 – 168 167

Physical parameters introduced by Fluent are: porosity of the porous medium, equivalent diameter of
the porous medium (mm), the permeability (m2) and the inertial resistance factor (1 / m).

3.1. Results

The results of simulated fluent flow are as follows:

Table.2 Simulation results and the relative error

Experimental flow (l/s) 9.5 10-4

Simulated flow (l/s) 9.07 10-4
Relative error 0.00043
Relative error % 4.3

The results obtained noted that the difference between the experimental and simulated flow is about
4.3%, it is acceptable.

4. Conclusion

In this study, we conducted a simulation based on experimental data obtained in the laboratory, using
the computer code FLUENT. To verify the accuracy of the mesh, tested convergence and even data
processing simulation were performed, we obtained acceptable results.

The simulated discharges are obtained for resistance factors specific to the filter bed that rely heavily
on the mean diameter and specific porosity of the sample. We quote for the sand filter bed experimental
flow is 9.5 10-4 l / s and the value of the simulated flow is 7.9 10-4 l / s.

Numerical simulations yielded an estimate of the duration of the experiment and determine the
magnitude of the measured parameter.

Acknowledgements

The authors would like to express their gratitude to the MESRS of Algeria, for the support.
168 Mohammed Amine Kendouci et al. / Energy Procedia 36 (2013) 163 – 168

References
[1] Yann Callot, «Histoire d’un massif de dunes, le Grand Erg Occidentale (Algérie)»Géodynamique des milieux continentaux,
Sécheresse N°1, Vol2 1991.
[2] Maazouzi.A; Kettab.A et Badri.A; «Etude de procédés de filtration sur sable de la région de Béchar en pré traitement de
l’eau potable» Desalination 206 (2007) 358–368
[3] Vinay Kumar Tyagi, Abid Ali Khan, A.A.Kazmi, Indu Mehrotra, A.K.Chopra, «Slow sand filtration of UASB reactor
effluent: A promising post treatment technique» Desalination 249 (2009) 571–576
[4] Ahmed Benamar, HuaQing Wang, Nasre-dine Ahfir, Abdellah Alem, Nicolas Masséi, Jean-Paul Dupont, «Effets de la vitesse
d’écoulement sur le transport et la cinétique de dépôt de particules en suspension en milieu poreux saturé» Tribology International
43 (2005) 455–466
[5] Youcef Touil, Samir Taha, Rachid Issaadi, Abdeltif Amrane, « Plot plant for wastewater treatment involving septic pit and
biological filtration on sand of dunes of the Algerian Sahara» Desalination 10 (2009) 148–152
[6] Menoret. C « Use of recirculation Through medium size granular filters to treat small food processing industry effluents. »
Conference IWA apparait dans Wat.Sci.tech. 2001.
[7] A.Maazouzi, A. Kettab, R.Khelfaoui, A. Badri, B. Zahraoui , « Modélisation du procédé de prétraitement des eaux Par
filtration sur sable». Le Premier Colloque Maghrébin sur le Traitement et le Dessalement des Eaux (CMTDE) Tunis 2007