A SEMINAR PAPER PRESENTATION ON:

COMPUTATIONAL FLUID DYNAMICS (CFD) AS A TOOL TO

STUDY CUTTING TRANSPORT IN WELLBORES

BY:

ISU, DAVID OMIKO 20154998538

SUBMITTED TO THE

DEPARTMENT OF PETROLEUM ENGINEERING

SCHOOL OF POSTGRADUATE STUDIES

FEDERAL UNIVERSITY OF TECHNOLOGY, OWERRI

JULY, 2018
1.0 INTRODUCTION

Cutting removal is one of the critical components of drilling operation. Faster the
formed particles are displaced by the drilling fluid better the penetration rates
due to contact of bit with fresh formation surface. Adequate circulation rates
needed to displace cuttings formed. At the same time proper mud properties
must be maintained to avoid problems such as stuck pipe and increased torque.
An important mud property affected by cuttings is the density but this change is
not uniform throughout the wellbore and can lead to problems. The interaction
between particles and drilling fluids is a complex problem and existence of
different wellbore conditions further complicates any prediction for adequate
hole cleaning.

The study of cutting transport has been a general interest to several researchers.
Initial studies were conducted with experimental setups to investigate the drill
cutting transport in the annular sections of a wellbore. Experiments were also
conducted for directional and horizontal wells to study the transport
characteristics and the parameters affecting the cutting transport. Some of the
research included theoretical models to describe the two-phase solid-liquid flow
process.

The use of a CFD model is proven to be a successful tool in many areas of fluid
flow. More applications of CFD exist in the design and manufacture of drill bits
and the related bit hydraulics. Researchers simulated drilling conditions to
visualize velocity and pressure gradients for different wellbore and bit geometry.
Recent advances in computer hardware and increased capabilities allow
researchers to investigate more complex problems.

2
2.0 METHODOLOGY
Computational fluid dynamics (CFD) is a branch of fluid mechanics that uses
numerical analysis and data structures to solve and analyze problems that involve
fluid flows (Mme and Skale, 2012). CFD involves the solution of the governing
laws of fluid dynamics numerically. Computers are used to perform the
calculations required to simulate the interaction of liquids and gases with surfaces
defined by boundary conditions. With high-speed supercomputers, better
solutions can be achieved. Ongoing research yields software that improves the
accuracy and speed of complex simulation scenarios such as transonic or
turbulent flows.

The physical aspects of any fluid flow are ruled by three main principles:

1. The conservation of mass,

2. The second law of Newton and
3. The conservation of energy.

These basic principles may be expressed in terms of mathematical equations,

which are mostly partial differential equations. The computational fluid dynamics
technique tries to solve the equations that rule the fluids flow numerically, while
the mathematical solution advances in space and time to obtain the complete
numerical flow field description (Hussain, 2010).

The comparison of the results of numerical fluid dynamics with experimental data
has taken on an important role in validating and establishing limits of many
approaches for the ruling equations. Traditionally, it has been shown to be an
effective, low cost alternative for total scale measures. This situation has led to an

3
increase in the development of numerical simulation routines and commercial
codes (Abouzar and Bandar, 2008).

The flow in the well is simulated as annulus flow. The inner surface of the annulus
is the drilling pipe wall and the outer surface is the well wall. The equations
governing the flow are derived from Navier-Stokes and continuity equations. The
flow in the annulus is considered 2-D flow, laminar or turbulent depending on the
flow velocity, but the fluid is non-Newtonian. The fluid moves in z-direction and
there is no radial velocity component, i.e., Ur = 0 (Belavadi and Chukwu, 2014).

The differential forms of the continuity equation (law of mass conservation) and
the Navier Stoke’s equation (from Newton’s second law) respectively in equations
1 and 2.

𝜕𝜕𝜕𝜕 𝜕𝜕
+ (𝜌𝜌𝑢𝑢𝑖𝑖 ) = 0 (1)
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕𝑖𝑖

𝜕𝜕 𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕 2 𝜕𝜕𝜕𝜕𝑘𝑘
(𝜌𝜌𝜌𝜌𝑖𝑖 ) + �𝜌𝜌𝜌𝜌𝑖𝑖 𝑢𝑢𝑗𝑗 � = − + �2𝜇𝜇𝑆𝑆𝑖𝑖𝑖𝑖 − 𝜇𝜇 𝛿𝛿 � + 𝜌𝜌𝑓𝑓𝑖𝑖 (2)
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕𝑖𝑖 𝜕𝜕𝜕𝜕𝑗𝑗 𝜕𝜕𝜕𝜕𝑗𝑗 3 𝜕𝜕𝜕𝜕𝑘𝑘 𝑖𝑖𝑖𝑖

Sij and δij are the rate of strain tensor and the Kronecker delta-tensor
respectively. μ is the physical property of the fluid called molecular viscosity. The
molecular viscosity term in Equation 2 must be replaced by any model which
describes the visco-elastic behaviour of the non-Newtonian fluids. The Power law
model is chosen for this study. Power law model describes the relationship of
shear stress and shear rate as:

Kϒn
τ = Kϒn or μ = (3)
ϒ

Where

4
τ = Shear stress

μ = Fluid viscosity

K = Consistency index

γ = Shear rate

n = Flow behaviour index

Value of n indicates the type of the drilling mud. If n is lower than 1, then it will be
shear thinning fluid. If n is 1, then the fluid is Newtonian fluid. The value more
than 1 will correspond when the fluid is shear thickening fluid (Cho and Osisanya,
2002).

In Reynolds-Averaged Navier-Stokes (RANS) approach, Reynolds decomposition

applied to the Navier-Stokes equations, which decomposes the turbulent
variables into instantaneous (fluctuating) and mean (time averaged) components,
and flow equations are averaged over a time scale that is long compared to that
of the turbulent motion (Cho and Osisanya, 2002). The main idea of Reynolds
averaging is to decompose the flow to averaged and fluctuating component:

Ui = Ȗi + ui, P = Ṗ + p (4)

This process is called Reynolds decomposition. The upper case letters with
overbar represent the mean values, the lower case letters represent the
fluctuating values on the right hand sides in equation 4. Since = ui = 0 and uj = 0 ,
inserting relations 4 into Equations 1 and 2 and time averaging, one obtains
relations 6 and 7. By time averaging the velocity component results in expression
5 (Mazumder et al, 2003).

5
 (5)

(6)

(7)

Equations 6 and 7 are called Reynolds-averaged Navier-Stokes equations. They

have the same general form as the instantaneous Navier-Stokes equations, with
the velocities and other solution variables now representing ensemble-averaged
(or time-averaged) values. Additional terms now appear that represent the effects
of turbulence. These Reynolds stresses, − ρuiuj , must be modelled in order to
close Equation 7 (Kim and Yoo, 2006).

2.1 Stages in Computer Modeling using CFD

a. Formulate the Flow Problem

The first step of the analysis process is to formulate the flow problem by seeking
answers to the following questions:

o what is the objective of the analysis?

o what is the easiest way to obtain those objective?
o what geometry should be included?
o what are the freestream and/or operating conditions?
o what dimensionality of the spatial model is required? (1D, quasi-1D,
2D, axisymmetric, 3D)
o what should the flow domain look like?

6
 o what temporal modeling is appropriate? (steady or unsteady)
o what is the nature of the viscous flow? (inviscid, laminar, turbulent)
o how should the gas be modeled? (Mazumder, 2003)

b. Model the Geometry and Flow Domain

The body about which flow is to be analyzed requires modeling. This generally
involves modeling the geometry with a CAD software package. Approximations of
the geometry and simplifications may be required to allow an analysis with
reasonable effort. Concurrently, decisions are made as to the extent of the finite
flow domain in which the flow is to be simulated. Portions of the boundary of the
flow domain conicide with the surfaces of the body geometry. Other surfaces are
free boundaries over which flow enters or leaves. The geometry and flow domain
are modeled in such a manner as to provide input for the grid generation. Thus,
the modeling often takes into account the structure and topology of the grid
generation (Mazumder, 2003).

c. Establish the Boundary and Initial Conditions

Since a finite flow domain is specified, physical conditions are required on the
boundaries of the flow domain. The simulation generally starts from an initial
solution and uses an iterative method to reach a final flow field solution (Bakker,
2002).

d. Generate the Grid

The flow domain is discretized into a grid. The grid generation involves defining
the structure and topology and then generating a grid on that topology. Currently

7
all cases involve multi-block, structured grids; however, the grid blocks may be
abbuting, contiguous, non-contiguous, and overlapping. The grid should exhibit
some minimal grid quality as defined by measures of orthogonality (especially at
the boundaries), relative grid spacing (15% to 20% stretching is considered a
maximum value), grid skewness, etc... Further the maximum spacings should be
consistent with the desired resolution of important features. The resolution of
boundary layers requires the grid to be clustered in the direction normal to the
surface with the spacing of the first grid point off the wall to be well within the
laminar sublayer of the boundary layer. For turbulent flows, the first point off the
wall should exhibit a y+ value of less than 1.0 (Bakker, 2002).

e. Establish the Simulation Strategy

The strategy for performing the simulation involves determining such things as
the use of space-marching or time-marching, the choice of turbulence or
chemistry model, and the choice of algorithms.

f. Establish the Input Parameters and Files

A CFD codes generally requires that an input data file be created listing the values
of the input parameters consisted with the desired strategy. Further, a grid file
containing the grid and boundary condition information is generally required. The
files for the grid and initial flow solution need to be generated.

g. Perform the Simulation

The simulation is performed with various possible with options for interactive or
batch processing and distributed processing.

8
h. Monitor the Simulation for Completion

As the simulation proceeds, the solution is monitored to determine if a

"converged" solution has been obtained, which is iterative convergence.

i. Post-Process the Simulation to get the Results

Post-Processing involves extracting the desired flow properties (thrust, lift, drag,
etc...) from the computed flowfield.

j. Make Comparisons of the Results

The computed flow properties are then compared to results from analytic,
computational, or experimental studies to establish the validity of the computed
results.

k. Repeat the Process to Examine Sensitivities

The sensitivity of the computed results should be examined to understand the

possible differences in the accuracy of results and / or performance of the
computation with respect to such things as:

o dimensionality
o flow conditions
o initial conditions
o marching strategy
o algorithms
o grid topology and density
o turbulence model

9
 o chemistry model
o flux model
o artificial viscosity
o boundary conditions
o computer system (Bakker, 2002)

l. Document

Documenting the findings of an analysis involves describing each of these steps in

the process.

3.0 RESULTS AND DISCUSSIONS

In this case study, the computational mesh represents a virtual flow system,
formed by the configurations of the well wall inner diameter of 380 mm and the
drill pipe with a diameter of 200 mm. Apart from the simulation that will be
carried out to observe the effect of shear thinning behavior of the fluid in cuttings
transportation, the drilling pipe will not be rotating and is positioned at the center
of the well bore. The total depth of the well is 2000 m. GAMBIT or IcemCFD
software can be used for the annulus modeling and discretization.

10
Plate 1: Left: Resulted model of the discretized annulus. Right: Cell distribution in
the annulus.

A one meter section of the geometry has a total mesh layout of 76,457
hexahedric cells that provides the system with a structured mesh. In the
simulation, Fluid model 1 (Fluid1) will be used which has density of 1003 kg m-3 ,
consistency index, K = 0.174 and the flow behaviour index , n = 0.631 as obtained
from the laboratory tests. The mud feeding rate will be varied within a range of
100 ft min-1 to 350 ft min-1 (0.508 m/s – 1.778 m/s ). The cuttings are assumed as
Sandstone rock source which has density of 2300 kg m-3 and penetration rate of
0.125 kg sec-1. All these values are assumed from a laboratory test on cuttings
transport conducted in University of Stavenger Multiphase flow loop test
(Mazumder, 2003). In order to calculate the velocity components, the ruling
equations will be integrated in each computational cell throughout the domain,
being discretized following the finite volumes approach. They will be linearized
and solved numerically. The calculations will be done using the pressure

11
discretization scheme following the PRESTO routine. For the pressure-velocity
coupling the SIMPLEC algorithm will be applied. Since the drilling fluid charging
rate is varied to involve the laminar and the turbulent regions, the k − ε model of
Lander and Spalding will be selected for the turbulence simulation. The equations
of mass, momentum, turbulent kinetic energy and turbulent dissipation rate will
be discretized using the QUICK routine scheme due to its better adaptation to
hexahedric meshes. The commercial code recommended to conduct the
simulation is ANSYS (FLUENT) (Kim and Yoo, 2006).

After the mesh is imported into the flow solver, ANSYS Fluent, flow physics is set
up and simulations run. The velocity profile is confirmed to be consistent.

Sand particles of various sizes will be injected at selected portions of the annulus
at various angles.

Contours of sand distribution in the annulus will reveal the movement of cuttings
in the annulus and the performance of various sand sizes.

Further simulations can be made to study the following scenarios:

3.1 Hole cleaning performance at various well inclinations

The extent to which the wellbore is cleansed is understood by the fraction of

cuttings concentration at the surface of the well at different operational flow
velocity. This concentration is produced from the particles transported
successfully from the well bottom to the surface. We assumed that there is no
energy or mass transfer between particles during the upward motion in the
annulus (Hussain, 2010).

12
Three different orientations can be simulated with 3 mm particles diameter and
0.9 sphericity. Results at 5, 10 and 15° diverted well orientations can be plotted
and predictions made (Hussain, 2010).

3.2 Effect of cuttings size on

Three cutting sizes of equivalent diameters of 3, 5 and 7 mm for small, medium

and large respectively will be selected to study their contribution on the wellbore
cleaning performance. The discrete phase model simulation solver in FLUENT
which allows the user to inject drilled cuttings with various sizes and shapes into
the fluid flow was chosen for this analysis. The simulation results of the
concentration at the annulus outlet can be plotted. It could be concluded that the
particles with diameter of 3 mm have better cleaning performance compared
with the other larger sizes. This implies that fine particles are the easiest one to
clean out (Hussain, 2010).

3.3 Effect of cuttings shape

In this analysis, drilled cutting shapes can be standardized and compared to the
shape of the sphere. This concept is widely known as the shape factor. The shape
factor is simply a measure of the sphericity of a certain cutting particles. A cubic
shape is known to have a sphericity of 85% of that of the sphere, while the sphere
has a shape factor of 1. Three cutting shape factors of 1, 0.9 and 0.85 can be
selected to investigate its effect on the cleaning performance. Cuttings are
produced in different sizes and shapes while drilling depending on the rate of
penetration, type of formation and drilling fluid properties. The simulation results
at various shape factors can as well be plotted (Hussain, 2010).

13
4.0 CONCLUSION

The CFD technique emerged as a result of the current increase in computer

processing speed and available memory. Currently, this branch of fluid dynamics
complements experimental and theoretical work, providing economically
interesting alternatives through the simulation of real flows and allowing an
alternative form for theoretical advances under conditions unavailable
experimentally.

This paper has proved that CFD can be used to cuttings transport in wellbores and
also evaluate the performance of drill bits.

REFERENCES

Abouzar M.P, Bandar D.A (2008): “Using Nanoparticles to Decrease Differential

Pipe Sticking and its feasibility in Iranian Oil Fields”. Oil and Gas Business.

Bakker, A. (2002) ‘Lecture 5-Solution Methods Applied Computational Fluid

Dynamics’, Lectures of Fluent Instructor, pp. 41–43

Belavadi, M.N. and G.A. Chukwu (2014). Experimental study of the parameter
saffecting cutting transportation in a vertical wellbore annulus. SPE Paper
27880. DOI: 10.2118727880-MS

Cho, H., Shah, S.N. and Osisanya, S. O. (2002): “A three-segment hydraulic

model for cuttings transport in coiled tubing horizontal and deviated drilling”.
J. Can. Petroleum Technol., 41: 32-39. DOI: 10.2118/02-06-03.

Hussain, H. A. K. (2010) "Simulation of the Cuttings Cleaning During the Drilling

Operation." American Journal of Applied Science, P 7 (2010).

14
Kim, C. and Yoo, B (2006): “Rheological properties of rice starch-xanthan gum
mixtures”. Journal of Food Engineering, 75:120-128.
DOI:10.1016/J.FOODENG.2005.04.002.

Mazumder, Q. H., Santos, G., Shirazi, S. A. and McLaury, B. S. (2003) ‘Effect of

Sand Distribution on Erosion in Annular Three Phase Flow’, in Eighth
International Symposium on Liquid-Solid Flow, ASME Summer Meeting.

Mme, Uduak, and P. al Skalle. (2012)"CFD calculations of cuttings transport

through drilling annuli at various angles." International Journal of Petroleum
Science and Technology 6.2: 129-141.

15