Journal of Petroleum Science and Engineering 167 (2018) 300–315

Contents lists available at ScienceDirect

Journal of Petroleum Science and Engineering

journal homepage: www.elsevier.com/locate/petrol

Artificial intelligence techniques and their applications in drilling fluid T

engineering: A review
Okorie E. Agwua,∗, Julius U. Akpabioa, Sunday B. Alabia, Adewale Dosunmub
a
Department of Chemical and Petroleum Engineering, Faculty of Engineering, University of Uyo, P.M.B. 1017, Uyo, Nigeria
b
Department of Petroleum Engineering, University of Port Harcourt, Rivers State, Nigeria

A R T I C LE I N FO A B S T R A C T

Keywords: For an oil well to be said to have been successfully and conclusively drilled, the drilling fluid lies at the heart of
Drilling fluid the solution. Therefore, the guarantee to solving issues in oil well drilling is to contrive an optimal drilling fluid.
Artificial neural network However, there is usually a complex interplay of factors involved during drilling fluid formulation, property
Fuzzy logic determination, its performance in the well and its relationship with other wellbore drilling parameters. This is so
Case based reasoning
because drilling muds exhibit time dependent properties. This time dependency is the direct product of the
Hybrid intelligent systems
synergy among the various active additives that make up the mud and the characteristic of each additive
Genetic algorithm
Particle swarm algorithm especially at downhole conditions where the effects of temperature and pressure are well pronounced. These
additives are more often than not diverse in size, chemical activity, density and surface energy. Deriving
knowledge from the data from these parameters in order to develop a functional relationship between them is a
challenging task requiring advanced modelling techniques as well as human intuition and experience. The de-
pendence on human intuition and on the experiential knowledge of professional mud engineers lays bare the
shortcomings of traditional mud design techniques. Artificial intelligence techniques have been shown to alle-
viate this challenge. Exploiting the abundant literature on the various applications of artificial intelligence in oil
and gas operations, several works that show how and what artificial intelligence techniques are used in the
drilling fluid industry, and what have been achieved due to their use have been selected. In this paper, a review
of existing artificial intelligence techniques and their applications in drilling fluid engineering is given. This
paper also dug up and analyzed the strengths and pitfalls of each artificial intelligence technique. The ex-
amination of the strengths and deficiencies was done using the following virtues as the basic criteria: robustness
against noise, self-organization, generalization ability, data volume requirements and the convergence speed.
The artificial intelligence techniques presented in this paper include: artificial neural networks (ANNs), fuzzy
logic, support vector machines (SVM), hybrid intelligent systems (HIS), genetic algorithms (GA), case based
reasoning (CBR) and particle swarm algorithm (PSA). An overview of the applications of classical artificial
intelligence in drilling fluid engineering is also presented. From the review, it was gathered that the ANN
technique is the most widely applied in drilling fluid engineering accounting for over 54% of the papers re-
viewed; while lost circulation problem was the most predicted well problem related to drilling fluids accounting
for over 17% of the mud problems predicted. It was also observed that a blend of AI techniques performed better
than when each one of the AI techniques was used singly. Finally, judging the AI techniques on the criteria
mentioned above, ANN was found to meet all the listed criteria except for its slow speed of convergence while
ANN, GA, SVM and fuzzy logic were all found to be robust against noise.

1. Introduction
One of the most important advances in the scientific community
that has appealed to virtually all fields of human endeavour is the
concept of artificial intelligence (AI). In defining the term artificial
intelligence, Lea (2015) is of the opinion that “artificial” is very easy to
define. According to him, artificial can simply mean having no
occurrence in real life or not having the same configuration in nature.
However, the definition of intelligence is wide and varied (Legg and
Hutter, 2007). As a result, this then leaves the definition of AI as a
knotty problem. Little wonder Schatsky et al. (2014) opines that the
field of AI suffers from two extreme situations namely-too little and too
many definitions. To further buttress this point, Nilsson (2010), stated
that AI lacks a universally accepted definition. Whereas attempts to

∗
Corresponding author.
E-mail address: okorieagwu@uniuyo.edu.ng (O.E. Agwu).

https://doi.org/10.1016/j.petrol.2018.04.019
Received 8 November 2017; Received in revised form 27 February 2018; Accepted 9 April 2018
Available online 12 April 2018
0920-4105/ © 2018 Elsevier B.V. All rights reserved.
O.E. Agwu et al.
give an all-encompassing definition to the concept of AI are hindered by
explosive complexity, these complexities do not erode the fact that AI
holds great promise as tools for understanding the relationship between
complex entities. However, this work would align itself with the defi-
nition of AI put forward by McCarthy (2007), wherein he defines AI as
the science and engineering of developing computer programs which
when fed into a machine makes the machine exhibit the intelligence of
humans. AI is also referred to as “soft computing”, “virtual in-
telligence”, and “computational intelligence” (Mohaghegh, 2000). Ac-
cording to Rable (2017), the measurable values of AI include: the
possibility of making machines solve difficult problems at a faster speed
than what the human brain would take to solve the same problem;
analysis of big data in order to understand trends and make predictions
of future scenarios in the best possible manner with the least waste of
time and effort; and the consequence of which is huge savings in op-
erating cost. From the available literature, virtually all parts of the oil
and gas exploration and production chain have applied one artificial
intelligence technique or another in the course of their operations. For
instance, Field development: Doraisamy et al. (1998); Wellbore drilling
process: (Valipour et al., 2014; Skalle et al., 2000); Drilling platform
selection: (Wang et al., 2011); Well placement: (Naderi and Khamehchi,
2017; Isebor et al., 2014); Reservoir properties determination: (Lim,
2005; Baarimah et al., 2015; Ho and Ehara, 2007); Sand production
prediction: (Khamehchi et al., 2014); Gas-lift performance prediction
and long term gas-lift allocation optimization: (Rasouli et al., 2015);
History matching: (Firoozjaee and Khamehchi, 2014; Zaimy and Rasaei,
2013); Two-phase flow in pipes: (Ternyik et al., 1995a, 1995b); Iden-
tification of well test interpretation models: (Sung et al., 1995); For-
mation damage prediction: (Nikravesh et al., 1996); Completion ana-
lysis: (Shelley et al., 1998). The list still goes on and on. However, in the
oil and gas industry, drilling operations have been one of the most
troublesome parts wherein various problems such as wellbore in-
stability, lost circulation, kicks and blowouts and stuck pipe may be
encountered (Meschi et al., 2010). One major component of the oil well
drilling operation which is often referred to as “the blood of the drilling
process” is the drilling fluid. The drilling fluid is directly or indirectly
related to the problems or solutions of the problems enumerated above.
For instance, increasing mud weight controls wellbore pressures and
when it is too high, it fractures the formation; and too low a mud weight
leads to kick and in extreme cases-blowouts. Little wonder Agwu et al.
(2015) puts it that the drilling fluid is the “architect” which can make
the drilling operation either materialize or unrealizable. On the cost
implication axis, Bloys et al. (1994) reports that drilling muds represent
5%–15% of drilling costs but may cause 100% of drilling problems. One
way of preventing drilling problems is optimal selection of drilling
fluids, which is almost always based on the analysis of data obtained
from previously drilled wells in the areas. A typical drilling mud pro-
gramme provides information on the type of mud, mud constituents,
mud rheology, mud density, filtration and other mud properties as well
as the mud maintenance requirements (Schlumberger, 2017). However,
in order to design drilling fluids to meet a desired rheology which
would suit specific formation properties, combat wellbore challenges
and prevent fluid losses all require detailed analysis and decision
making. As Shadravan et al. (2015) put it, designing drilling muds and
other wellbore fluids are all done by trial and error in the laboratory;
with the initial guesses depending largely on the experience of the mud
engineer or the analyst in the laboratory. However, as Abdollahi (2007)
put it, the knowledge gained through experience could easily be lost
when the laboratory analyst or the mud engineer who possesses the
requisite experience leave the company. Therefore, there is need for a
system which would make use of all the data available to map out
trends and ultimately offer better understanding of drilling mud design
(Shadravan et al., 2015). One area that comes handy is artificial in-
telligence. The benefits of AI techniques in the present circumstance are
numerous. They include but are not limited to: ability to predict si-
tuations with complex relationships between input and output
301
Journal of Petroleum Science and Engineering 167 (2018) 300–315
parameters, it allows finding the best values of model properties
minimizing calculation errors in a given area etc. Of course, there are
phenomenal AI researches in wellbore drilling as well as in drilling fluid
engineering. As a result of this, many textbooks and scholarly papers
have been published in both local and international journals that ad-
dress the subject of using artificial intelligence techniques in oilfield
applications. Most of these publications give qualitative and quantita-
tive studies on what has been done in using them and the successes
recorded. However, due to the elaborate nature of the studies that have
been presented by diverse researchers, there is need to assess all the
studies and the research findings in one piece so as to isolate what has
been achieved and identify possible areas of interest where more stu-
dies should be directed. To achieve the objective of digging deep into
the application of AI in mud engineering, this paper makes a chronicle
of the available studies on the application of artificial intelligence in
drilling fluid engineering. The various research findings as presented by
diverse authors are highlighted and the future research needs on the
subject are presented. However, there is a limit to what can be ac-
complished in just one article, hence, considering the full list of all AI
techniques and their applications in drilling fluid engineering in one
review paper is in itself not wrong but it is deemed that their inclusion
in this work has an overloading effect on its contents. Therefore the
artificial intelligence applications in drilling fluid engineering was
narrowed down to artificial neural networks (ANNs), fuzzy logic, sup-
port vector machines (SVM), hybrid intelligent systems (HIS), genetic
algorithms, case based reasoning (CBR) and particle swarm algorithm
(PSA). This paper is organized as follows: first, a highlight on the
evolution of artificial intelligence in the exploration and production (E
& P) industry is made, an expatiation on the applications of artificial
intelligence in mud engineering follows, an overview of the strengths
and weaknesses of artificial intelligence techniques follows thereafter.
Selected applications of classical artificial intelligence and molecular
simulations for various drilling fluid related challenges are presented
thereafter. This is closely followed by a discussion of the learnings from
the review. This paper climaxes with some concluding remarks and
suggestions for further research are made.
2. Evolution of artificial intelligence in the exploration &
production (E & P) industry
Before delving into the history of AI applications in the E & P in-
dustry, it is pertinent to look at how the concept of AI came to being. i-
Scoop (2016) reports that AI was conceived in the 1950s. According to
Alegre (1991), the term artificial intelligence was coined as a result of a
confluence of ideas generated during a conference held on the campus
of Dartmouth College, USA by a group of scientists in the summer of
1956. In general, the timelines in the development of the AI techniques
and their developers arranged chronologically is shown in Table 1.
From the table, it is observed that artificial neural networks were the
earliest AI techniques to be developed while hybrid intelligent systems
are the latest.
Narrowing down the history of AI to its application in the E & P
industry reveals that research papers on AI application in the industry
as early as in the 1970s exist (Bravo et al., 2012). However, during the
Table 1
Evolutionary trend of various artificial intelligence techniques.
AI technique Year of development Developer
Artificial neural network 1943 McCulloch and Pitts
Fuzzy logic 1965 Lofti A. Zadeh
Genetic algorithm 1970 John Holland
Case based reasoning 1977 Schank and Abelson
Support vector machines 1995 Vapnik, V
Particle swarm algorithm 1995 Eberhart and Kennedy
Hybrid intelligent systems – –
O.E. Agwu et al.
early stages of AI development in the industry, Expert systems were the
main AI technique used. Their usage was mainly in management of
drilling operations and the treatment of wellbore drilling problems
(Courteille et al., 1986).
Popa and Cassidy (2012), however reports that AI techniques have
been used in the oil and gas industry since the early 1990s. This view is
in sync with the work of Bello et al. (2015) wherein he asserts that AI
techniques were first applied in the E &P industry in 1989.
In order to consolidate on the gains of using AI and to encourage
researches in this area, the Artificial Intelligence Applications Institute
(AIAI) was created in 1983. According to the Institute, AIAI is a tech-
nology transfer organisation that encourages the applications of AI re-
search to the advantage of individuals, corporate bodies, industries and
the government (Artificial Intelligence Applications Institute, 2014).
This move triggered the development of several AI based software
packages aimed at solving real life oilfield problems. Hence, in the
drilling aspect of the oil industry, softwares such as MUDMAN and
DrillEdge were developed (Barton and Sviokla, 1988; Boman, 2014).
On the one hand, the MUDMAN developed by Baroid Fluids Company is
built to analyse drilling fluids. On the other hand, the DrillEdge – a case
based reasoning software developed by Verdande Technologies is built
to identify potential drilling problems involving drilling equipment as
well as problems relating to the drilled well. In recent times, E & P
companies are actively integrating various AI techniques to build hy-
brid intelligent systems that have the capability of handling complex
phenomenon. For instance, Sheremetov et al. (2008) reports that a
hybrid of expert systems, fuzzy logic and web services, called Smart-
Drill, is useful in solving the problem of lost circulation. They further
added that the software is at field testing stage in a Mexican oil com-
pany. Recently, an AI strategy based on supervised learning (MLP
neural network with K∗) as reported by Cristofaro et al. (2017) was
developed to generalize lost circulation data collected over a period of
three years from five hundred events pertaining to lost circulation. The
AI strategy developed with a user friendly interface was applied to two
wells experiencing lost circulation in the Santos Basin in Brazil deep
water pre salt. This application was essentially able to predict and re-
commend treatment procedures for the lost circulation with the re-
sultant effect of saving six days wellbore drilling time.
3. Applications of artificial intelligence in drilling mud
engineering
In this section, a survey of the state-of-the-art researches which
involves the applications of AI techniques in drilling mud engineering is
presented. Firstly, the fundamentals of these artificial intelligence
techniques are highlighted; secondly, their applications in mud en-
gineering are discussed.
3.1. Artificial neural network (ANN)
Artificial neural networks (ANNs) are a form of artificial intelligence
that has proved to provide a high level of competency in solving many
complex engineering problems that are beyond the computational
capability of classical mathematics and traditional procedures (Shahin
et al., 1999). For the fundamental details about ANNs, a good read is
Petrus et al. (1995).
3.1.1. Cases where ANN were applied in drilling fluid engineering
Table 2 summarizes the research efforts made on the application of
ANNs to mud engineering. Though tracing the researches from the
earliest was difficult, the summary chronicled available research papers
from 2003 up until 2017. The areas in mud engineering where ANN
have been applied from the review include: (a) lost circulation pre-
diction (b) mud density prediction (c) mud rheological properties pre-
diction (d) filtration loss and mud cake permeability predictions (e)
mud flow pattern prediction in wellbore annulus (f) hole cleaning and
302
Journal of Petroleum Science and Engineering 167 (2018) 300–315
cuttings transport efficiency of mud (g) settling velocity of cuttings in
mud prediction and (h) frictional pressure loss prediction. The table
highlights the type of study conducted by each researcher, the ANN
method used, the ANN architecture, the input and output parameters as
well as the performance evaluation of the network developed in each
case (this is represented in terms of the correlation coefficient (R2), root
mean square error (RMSE), average absolute percent relative error
(AAPE), etc). From the summary, the following are observed:
1. Most of the researchers (about 90%) carried out their works with
multiple inputs and single output except for the authors Jeirani and
Mohebbi (2005); Shadravan et al. (2015) and Elkatatny et al.,
(2016) whose works involved 2, 7 and 4 outputs, respectively. This
shows that ANN can be used to model parameters with multiple
outputs.
2. The most widely used performance evaluation criteria is the R2
value. The average R2 value across the research works reviewed was
approximately 0.939 showing a good model performance across
board.
3. The aspects of mud engineering that ANN was mostly applied are
prediction of stuck pipe and lost circulation. These accounted for
about 18% and 15% of the researches, respectively. Since lost cir-
culation, according to Drilling Specialties Company (Online), is the
most costly mud related problem and stuck pipe problems according
to Siruvuri et al. (2006) could gulp over 40% of total well costs in
certain areas; this could probably be the main driver for numerous
prediction studies on them using ANN.
4. Most of the authors used data sets extracted from open literature to
create their network models.
Beyond the four areas highlighted above, one area worth men-
tioning is the AI software used by each author for estimating the drilling
fluid related challenge they studied. It must be said that there are nu-
merous ANN softwares for developing applications, and some of them
are designed for industrial use. The neural network toolbox in the
MATLAB software is one of these tools (Bravo et al., 2012). Other
softwares include neuro solutions, pythia, neuroph, neural designer etc.
It is seen that majority of the researchers in Table 2 used the neural
network toolbox in the MATLAB software for their work. Examples are
found in the works of Moazzeni et al. (2010); Rooki et al. (2012, 2014);
Al-Baiyat and Heinze (2012); Jahanbakhshi et al. (2014); Adesina et al.
(2015); Elkatatny (2017) etc. Additionally, many of these researchers
used the back propagation (BP) algorithm to train the networks.
However, just a handful used the neuro solutions software in their
work. This is seen in the cases of Miri et al. (2007) and Siruvuri et al.
(2006). From Table 2, since only two neural network softwares were
found to be used by the researchers in their studies, it is somewhat
inadequate to draw a conclusion on the basis of just those two softwares
as to which of them is recommended for any of the wellbore challenge
listed in this work. However, viewing the matter from the standpoint of
which of them is the most widely used by majority of the authors in this
work, the MATLAB software tops the recommendation chart. Though
the authors never gave any reason for choosing the software they used,
the choice by most authors to use the MATLAB software may not be
unconnected with the expert view of Schultz, a researcher at Halli-
burton who reports that the real results with using the software include:
facilitating ease of simulation, provision of an accurate, production-
standard algorithm and huge time savings (Mathworks, online). How-
ever, when using the neural network tool in MATLAB software, it must
be appreciated that several learning algorithms exist and the choice of
the algorithm by the user may likely affect the outcome of developed
models. For instance, in predicting stuck pipe, Miri et al. (2007) and
Siruvuri et al. (2006) used the same software-neuro solutions but got
different results. In this case, Miri et al. (2007) used MLP and RBF while
Siruvuri et al. (2006) used GFNN.
 Table 2
Summary of researches on mud engineering to which ANN was applied.
Authors Type of study conducted ANN ANN Input Parameters Output parameters Errors (PERFORMANCE EVALUATION CRITERIA)
method Architecture
O.E. Agwu et al.

RMSE AAPE R2

Osman and Aggour (2003) Mud density of WBM and BP 4–6 – 1 Initial mud density at surface conditions, Density 0.0056 0.367 0.9998
OBM temperature, pressure, type of drilling fluid
Jeirani and Mohebbi Filtration volume and mud BP 4–30 – 2 Time, pressure drop, water and NaCl weight percent Filtrate volumes and – – 0.9815 (Filter volume), 0.9433
(2005) cake permeability of WBM filter cake permeability (cake permeability)
Siruvuri et al. (2006) Prediction of differential pipe GFFNN – Water based sticking model: differential pressure, Differentially stuck or – – Stuck: MSE = 0.01369;
sticking hole depth, API fluid loss, MBT, chlorides, total No stuck NMSE = 1.04; MAE = 0.03459;
hardness, PV, YP, gels, inhibitor concentration, pH. r = 0.063
Oil based sticking model: PV, YP, gels, emulsion Non stuck: MSE = 0.01847;
ability, HPHT fluid loss, lime, chlorides, oil water NMSE = 1.40399; MAE = 0.0608;
ratio, hole depth r = 0.01619
Miri et al. (2007) Prediction of differential pipe MLP 9–40 – 2 Differential pressure, hole depth, API fluid loss, Stuck index – – MSE for MLP = 0.00547
sticking RBF solid percent, mud filtrate viscosity, plastic MSE for RBF = 0.00419
viscosity, yield point, 10 s and 10 min gel strength
Murillo et al. (2009) Prediction of pipe sticking – – Measured depth, TVD, bit flow rate, 10 s and 10 min Stuck pipe condition – – –
mechanism gels, mud weight, PV, YP, calcium filtrate, chloride
filtrate, torque, circulating pressure, WOB, drag, bit
size, BHA, ROP, RPM
Ozbayoglu and Ozbayoglu Estimating flow patterns (FP) BP – Liquid superficial velocity (ft/s) Pressure drop, dP/dL – – MSE = 0.006 [FPL with BP]
(2009) & frictional pressure loss Jordan/ and gas superficial velocity (ft/s) (psi/ft), type of flow MSE = 0.005 [FPL with J/E];
(FPL) of two phase fluids Elman pattern MSE = 0.005 [FP with BP]
(J/E) MSE = 0.005 [FP with J/E]
Shadizadeh et al. (2010) Predicting stuck pipe FFBPN 6–4 – 1 Differential pressure, geometric factor, pH, YP, PV Stuck pipe probability – – –
and

303
Moazzeni et al. (2010, Lost circulation prediction
2012)
Alireza et al. (2011) Lost circulation
Oladunni and Trafalis Predicting fluid flow patterns
(2011) in wellbore annulus
Rooki et al. (2012) Settling velocity of particles
in Newtonian &non-
Newtonian fluids
Al-Baiyat and Heinze Predicting stuck pipe before
(2012) its occurrence
Jahanbakhshi et al. (2012) Predicting stuck pipe before
its occurrence
gel strength.
FFBPN 18–30 – 1 Depth of well from ground surface, depth of well Square of mud loss – – 0.82
from sea level, drilled depth of well in the volume
considered day, drilling time, length of open hole
section, asmari formation top from ground surface,
northing and easting of well, bit size, average output
of pump, average pump pressure, mud weight, solid
percent of mud, mud viscosity, mud fluid loss,
amount of lost of circulation, amount of lost of
circulation in previous two days
FFBPN – Well depth from ground surface and from sea level, Amount of lost – – 0.76543
drilled depth, drilling time, length of open hole circulation
section, asmari formation top from ground surface,
northing, easting of well, bit size, average output of
pump, average pump pressure, mud weight, solid
percent of mud, mud fluid loss, amount of lost
circulation, amount of lost of circulation in two days
FFBPN 6–26 – 1 Flow rate, fluid density, plastic viscosity, borehole Flow pattern 0.9028
diameter of the pipe, drill collar OD, mud type
FFBPN 6–12 – 1 Particle density, particle diameter, fluid density, Terminal velocity 0.038 m/s – 0.986
fluid consistency, flow behaviour index and
acceleration due to gravity
FFBPN 18–19 – 2 Well direction characteristics, mud properties and Stuck index
drilling parameters
FFBPN 12–18 – 2 Differential pressure, hole depth, mud filtrate Stuck, Non stuck – – Accuracy = 82.8%
viscosity, fluid loss, solid content, BHA length, still
pipe time, hole size, 10 s and 10 min gels, PV, YP
(continued on next page)
Journal of Petroleum Science and Engineering 167 (2018) 300–315
 Table 2 (continued)

Authors Type of study conducted ANN ANN Input Parameters Output parameters Errors (PERFORMANCE EVALUATION CRITERIA)
method Architecture
O.E. Agwu et al.

RMSE AAPE R2

Zhu et al. (2013) Model of pipe sticking pre- BPNN – Sticking point depth, WOB, revolution, Stuck condition – – –
warning displacement, pump pressure, mud density, funnel
viscosity, sand content, thickness of mud cake,
water loss
Razi et al. (2013) Rheological properties of FFMLP 2–2 – 1 Temperature and concentration Plastic viscosity – – 0.965
WBM and 0.986
yield point
Razi et al. (2013) Apparent viscosity of WBM FFMLP 3–3 – 1 Shear rate, temperature and concentration Apparent – – 0.993
viscosity
Rooki et al. (2014) Hole cleaning FFBPN 6–10 – 1 Pressure, Temperature, pipe rotation, foam quality, Cuttings concentration – 5.93% 0.914
efficiency of foam fluid foam velocity, annulus eccentricity
Toreifi et al. (2014) Lost circulation prediction MNN – Geographic coordinates (east and north), the Loss circulation volume – – 0.94
current depth, depth of formation tip, penetration
rate, formation type, annulus volume, mud pressure,
flow rate of mud pump, mud pump pressure, filter
cake, viscosity, solid content, PV,YP, 10sec/10 min
gel
Li et al. (2014) Prediction FFBPN 6–12 – 1 Fluid density, particle density, particle diameter, Wall factor 0.0315 0.942
of the wall factor effect on 7–12 – 1 fluid consistency index, flow behaviour index, tube 0.0209 0.973
particle settling to sphere diameter ratio, diameter to length ratio
Maghrabi et al. (2014) Modelling shale erosion – – Shale mineralogy and fluid composition % recovery of cuttings 0.85–0.90
behaviour in drilling fluids from the shale-erosion
test
Jahanbakhshi et al. (2014) Lost circulation prediction MLP 11-13-1 Hole depth, porosity, formation permeability, Lost circulation – – 0.94

304
differential pressure, ROP, ECD, average pump
pressure, temperature in lost interval, PV, YP, 10 s/
10 min gel strength, Solid percent, Mud filtrate
viscosity, API fluid loss, minimum horizontal stress,
uniaxial compressive strength, tensile strength,
natural fracture orientation, Young modulus
Jahanbakhshi and Lost circulation prediction MLP 11-13-1 Hole depth, porosity, formation permeability, Lost circulation – – 0.8770
Keshavarzi (2015) differential pressure, ROP, ECD, average pump
pressure, Temperature in lost interval, PV, YP, 10 s/
10 min gel, Solid percent, mud filtrate viscosity, API
fluid loss, minimum horizontal stress, uniaxial
compressive strength, tensile strength, natural
fracture orientation, Young modulus
Adesina et al. (2015) Predicting downhole mud FFBPN 1–10 – 1 Temperature Density – – 0.99852 (Diesel OBM)
density of OBM 0.99414 (Jatropha OBM) 0.99675
(Canola OBM)
Rooki (2015) Pressure loss estimation of FFMLP 6–10 – 1 Diameter ratio, eccentricity of annulus, flow Pressure loss – 4.32% 0.999
Herschel Bulkley fluid behaviour index (n), consistency index (K), yield
stress, liquid flow rate (Q).
Shadravan et al. (2015) Mud design BP 3–6 – 7 Fluid density, ingredient A content, temperature 600, 300, 200, 100, 6 & – – –
3 RPM
Rooki and Rakhshkhorshid Cuttings transport modelling RBFN 6–60 – 1 Pressure, temperature, pipe rotation, foam quality, Cuttings concentration – 5.7% 0.922
(2016) of foam fluid foam velocity, annulus eccentricity
Behnoud far,and Hosseini Prediction of lost circulation FFBPN 4–3 – 1 Pump pressure, depth, mud flow rate, mud weight Amount of lost 0.1879 – 0.9991
(2016) volumes circulation
Tatar et al. (2016) Brine density prediction – – Temperature, pressure and concentration Brine density 0.999999
Ali et al. (2016) Frictional pressure loss RBF – Gas superficial velocity, ROP, liquid superficial, Pressure drop 0.008783 – 0.9965
prediction velocity, pipe rotation speed, hole inclination
(continued on next page)
Journal of Petroleum Science and Engineering 167 (2018) 300–315
 Table 2 (continued)

Authors Type of study conducted ANN ANN Input Parameters Output parameters Errors (PERFORMANCE EVALUATION CRITERIA)
method Architecture
O.E. Agwu et al.

RMSE AAPE R2

Elkatatny et al. (2016) Rheological properties of – 4–12 – 4 Mud density, marsh funnel viscosity and solid PV, YP, flow behaviour – – n = 0.954; PV = 0.917; K = 0.9205
invert emulsion percent index, and fluid
mud consistency
Rooki (2016) Pressure loss estimation of GRNN 6–260 – 1 Diameter ratio, eccentricity of annulus, flow Pressure loss 8.45 kPa 6.24% 0.98
Herschel Bulkley fluid behaviour index (n), consistency index (K), yield
stress, liquid flow rate
Ejimofor (2016) Prediction of the mass flow FFNN 5–15 – 1 Downstream level (LT18), Middle level (LT17), Coriolis mass flow rate MAE 4.14 MSE 0.993
rate of drilling muds Upstream level (LT15), Density, Pressure drop kg/min 30.05
(PDT12)
Kamyab et al. (2016) Friction factor of cuttings slip – 2–5 – 1 Logarithm of Reynolds number, sphericity of the Logarithm of friction – – –
velocity particle factor
Hoang (2016) Mud viscosity estimation FFBPN 2–10 – 1 Density, shear stress Mud viscosity MSE 74.95 MAPE 0.98
8.12%
Chhantyal et al. (2016) Mud flow rate estimation RNN Ultrasonic level measurements (LT-15, LT17 and LT- Mud flow rate – MAPE –
18) 5.6%
Elkatatny (2017) Rheological properties of – – Mud density, marsh funnel viscosity and solid PV, YP, n and K & – < 6% > 0.90
WBM percent apparent viscosity
Bispo et al. (2017) Estimating apparent viscosity FFMLP 4–6 – 1 Temperature, concentrations of barite, bentonite Apparent viscosity – – 0.9486
of WBMs and xanthan gum
Chhantyal et al. (2017a) Prediction of mud flow rate – 3–4 – 1 Ultrasonic level measurements (LT-1, LT-2 and LT- Mud outflow – – –
3)
a
Chhantyal et al. (2017b) Prediction of mud flow rate RBNN – Ultrasonic level measurements (LT-1, LT-2 and LT- Mud flow rate 7.10 , 8.43** – –
3) kg/min

305
=RMSE with filtered inputs; ** = RMSE with unfiltered inputs.
Journal of Petroleum Science and Engineering 167 (2018) 300–315
O.E. Agwu et al. Journal of Petroleum Science and Engineering 167 (2018) 300–315

Table 3
Summary of researches on mud engineering to which fuzzy logic was applied.
Authors Type of study conducted Input Parameters Output parameters Errors (Performance evaluation
criteria)

R2 MSE

Sheremetov et al. (2008) Lost circulation problem – – – –

Murillo et al. (2009) Prediction of pipe Measured depth, TVD, bit flow rate, 10 s and 10 min gels, mud Stuck pipe
sticking mechanism weight, PV, YP, calcium filtrate, chloride filtrate, torque, circulating condition
pressure, WOB, drag, bit size, BHA, ROP, RPM
Yunhu et al. (2011) Drilling fluid leak zone Formation pressure, well depth, mud density, mud viscosity, mud – – –
identification flow rate
Ataallahi and Blowout Impact, duration and flow rate Consequence – –
Shadizadeh (2015)
Ahmadi et al. (2016) Mud density prediction Initial mud density, pressure, temperature Density 0.7237 69.0907
Chhantyal et al.(2017) Mud flow rate prediction Ultrasonic level measurements (LT-1, LT-2 and LT-3) Mud flow rate 0.98 1.74% (MAPE)

3.2. Fuzzy logic than when used singly.

A very good read for the fundamentals and applications of fuzzy
logic is found in the work of Ross (2010).
3.2.1. Cases where fuzzy logic were applied in drilling fluid engineering
Research papers reviewed on the use of fuzzy logic in drilling fluid
engineering indicates that aspects of mud engineering such as lost cir-
culation, stuck pipe prediction, mud density prediction and the con-
sequences of kick were in the spotlight as seen in Table 3.
3.3. Hybrid intelligent systems
3.3.1. Cases where hybrid systems were applied in drilling fluid engineering
For a good overview of this technique, the work of Medsker (1995)
is useful. From Table 4, the application of hybrid intelligent systems to
solving problems in mud engineering is highlighted. Practical problems
tackled with the use of hybrid systems include: kicks, lost circulation
and effects of high temperature high pressure (HTHP) conditions on
mud densities. It is also noteworthy that the combination of AI tech-
niques to make a hybrid is such that two techniques with unique
characteristics are chosen. For instance, since the training of ANN have
slow speed of convergence but training of fuzzy intelligent systems
(FIS) have high speed of convergence, their combination to form ANFIS
by Deregeh et al. (2013) would prove to be far more reaching in its
performance than when just one of them is used. In this case, the effects
of slow convergence by ANN would be neutralized by the high speed of
convergence of FIS. Moreover, since PSO and GA are optimization al-
gorithms, hybrids of PSO and ANN to form PSO-ANN as well as GA and
FIS to form GA-FIS would essentially make their performance better off
3.4. Genetic algorithms (GA)
3.4.1. Cases where genetic algorithms were applied in drilling fluid
engineering
A very good read on the fundamentals of genetic algorithms can be
found in the review paper of Ab Wahab et al. (2015). A summary of
research papers on the use of genetic algorithms in drilling fluid en-
gineering indicates that aspects of mud engineering such as rheological
properties determination, particle settling velocity in drilling mud etc.
were in the spotlight as seen in Table 5.
In addition to the above, Barati et al. (2014) used multi gene genetic
programming, a branch of genetic algorithm to develop empirical re-
lations that can predict the drag coefficient of flow around a smooth
sphere. They examined the performance of the models by comparing it
with other models. Using the sum of squared of logarithmic deviations
as a criterion; they report that the model gave approximately 16% and
69% better results than the existing models.
3.5. Support vector machines (SVMs)
The work by Steinwart and Christmann (2008) is a good read on the
subject of support vector machines.
3.5.1. Cases where support vector machines were applied in drilling fluid
engineering
Table 6 highlights the research endeavours of various researchers on
the application SVMs to solving problems in mud engineering. Practical
problems tackled with the use of SVMs include: fluid flow pattern

Table 4
Summary of researches on mud engineering to which hybrid intelligent systems was applied.
Authors Type of study Type of HYBRID Number of Input Parameters Output parameters Errors (Performance evaluation criteria)
conducted data points
RMAE R2 MSE

Ghaffari Analysis of mud Self-Organizing 392 Mud Weight, Funnel viscosity, Solid Mud loss pattern – – –
et al. loss patterns Neuro-fuzzy content, Plastic Viscosity, Yield
(2009) Inference system Point, RPM loss (bbl): the rate of
(SONFIS) Mud loss & Pump Pressure
Deregeh Kick detection ANFIS 412 Depth d-exponent 15.9753% – –
et al.
(2013)
Ahmadi Prediction of mud PSO-ANN 664 Initial mud density, pressure, Density – 0.9964 0.0001374
et al. density GA – FIS temperature 0.9397 0.091
(2016)
Zhou et al. HTHP drilling PSO-BP Neural – Initial mud density, pressure, Density at input – – –
(2016) fluid density of network difference temperature difference, temperature &
WBM & OBM water volume fraction, oil volume pressure
fraction

306
O.E. Agwu et al. Journal of Petroleum Science and Engineering 167 (2018) 300–315

Table 5 4. Strengths and weaknesses of artificial intelligence techniques

Summary of researches on mud engineering to which genetic algorithm was
applied. Of the seven AI techniques highlighted in this work, it is pertinent to
Authors Type of study Parameters Errors (Performance ask if there is any one of them that can be said to be 100% flawless and
conducted evaluation criteria) suitable for use in all circumstances. In what seems to be an answer to
the above question, Luchian et al. (2015) asserts that it is more bene-
R2 SSE
ficial focusing on solving the problem rather than wasting time to find
Rooki et al. Rheological Yield stress, fluid 0.9972a 1.8305a the best method. However, a theorem propounded by Wolpert and
(2012) parameter consistency and flow Macready (1997) called the No Free Lunch Theorem for Optimization
determination for index (NFLTO) states as follows: Given that all problems are assumed to be
Herschel Bulkley
equal in strength and irrespective of the criteria used for judging its
drilling fluids
Goldstein Prediction of Nominal diameter of 0.97 0.26 performance, all methods used for solving the problem have the same
and particle settling the settling particle, (RMSE) performance. In support of the NFLTO, Anifowose et al. (2016) opines
Coco velocity kinematic viscosity that there is not a single all-encompassing AI approach that will ef-
(2014) of the fluid, and fectively address all challenges in all data and computing conditions
submerged specific
since each of the AI techniques is locked up in its unique strengths and
gravity of the
particle inevitable defects. Table 7 summarizes the strengths and weaknesses of
some of the AI techniques. The strengths and weaknesses of four
a
Average values. techniques namely ANN, fuzzy logic, SVM and GA were benchmarked
on six criteria of robustness against noise, convergence speed, sus-
prediction, stuck pipe prediction, frictional pressure loss prediction, ceptibility to overfitting, data volume requirements, ability to self-or-
density and viscosity predictions, lost circulation and mud rheology ganize and generalization ability. It is observed that ANN, fuzzy logic,
estimations. It is necessary to state here that the most common kernel SVM and GA are all robust against noise, while fuzzy logic has a better
function used by most researchers is the radial basis function. speed of convergence compared with ANN, SVM and GA. The duo of
ANN and SVM has the ability to generalize while ANN can self-organize
3.6. Particle swarm algorithm and requires huge data to predict complex phenomena well while SVM
requires small data volumes.
A review of the particle swarm algorithm technique can be found in To show the effectiveness or otherwise of these AI techniques, a few
the work by Eberhart and Shi (2001). researchers have attempted to compare one or two of the AI techniques.
Table 8 shows a few of these researches. It is noticed that the re-
3.6.1. Cases where particle swarm algorithm were applied in drilling fluid searchers mostly compared ANN against SVM. It is observed from the
engineering performance criteria of better performance with lower RMSE, MAPE,
As at the time of this review, the only recorded case of using particle MSE and MAE values and high R2 values, that SVM tends to perform
swarm algorithm in drilling fluid engineering was the work done by better in all of the cases cited. A unique comparison between fuzzy logic
Toreifi et al. (2014) where they used the method in conjunction with and two hybrid systems was done by Ahmadi et al. (2016). In their
modular neural network to improve the prediction of the lost circula- work, it is observed that the hybrid performed better that the fuzzy
tion problem. logic used alone.

3.7. Case based reasoning 5. Selected applications of classical artificial intelligence in

Case-based reasoning (CBR) as a subset of artificial intelligence (AI)
solves new problems by reusing the experience of a similar problem
that had been solved in the past (Aamodt and Plaza, 1994; Raja et al.,
2011). An overview of the CBR technique can be found in the work of
Aamodt and Plaza (1994). The following are the cases in which CBR
were applied in drilling mud engineering:
(a) Lost circulation: Skalle et al. (2000) used a CBR tool to solve the
problem of lost circulation by creating 50 base cases based on in-
formation obtained from fields in the North Sea. The cases were
narrowed down to two best fitting cases from which the new case
was solved. In addition, Rui et al. (2013) used CBR to predict the
risks associated with lost circulation. The result of their study shows
that using the technique of CBR can predict drilling risks such as
lost circulation in a simple manner which can practically be applied
in the field.
(b) Hole cleaning: Shokouhi and Skalle (2009) used CBR to show how
the root cause of poor hole cleaning could be effectively de-
termined. Their results were satisfactory though they recommended
that some improvements could still be done.
(c) Wellbore leakage: Abdollahi et al. (2008) used CBR to create 18
cases, 12 of which were solved and 6 were unsolved cases. They
categorized the cases into five groups according to the main cause
of leakage. To effectively determine the main cause of the leakages,
they assert that predefined rules can be used.
drilling fluid engineering
Also known as symbolic artificial intelligence, classical artificial
intelligence refers to the branch of artificial intelligence that deals with
representing the implicit nature of human knowledge in an explicit
form by using facts and rules (Reingold and Nightingale, 1999). Clas-
sical artificial intelligence often deals with numerical simulation. Nu-
merical simulation is becoming one of the leading techniques used to
study several engineering problems that are complex in nature (Ablat
and Qattawi, 2017). Carrying out numerical and molecular simulations
on drilling muds parameters and challenges related to them are very
important. For instance, wellbore challenges such as stuck pipes and
blowout where controlled experiments are difficult or impossible to
perform can easily be studied using numerical simulations. Several
drilling fluid related challenges have been studied through numerical
simulations. Hence, in this section, applications of numerical simulation
in drilling fluid engineering are discussed. The following are some of
the selected cases:
Feng and Gray (2017) developed a numerical model on the basis of
the finite-element method for simulating the problem of lost circula-
tion. The model combines dynamic circulation of mud in the borehole
and induced-fracture propagation into the formation. The model pro-
vides an approximate value for the wellbore pressure, rate of fluid loss,
and fracture profile during wellbore drilling. The results from their
work show that the viscous pressure losses in the annulus of the well-
bore resulting from the dynamic circulation of mud can lead to re-
markable increases in wellbore pressure and fluid loss.

307

Table 6
Summary of researches on mud engineering to which support vector machines was applied.
Authors Type of study conducted
Oladunni and Trafalis Predicting fluid flow patterns in
(2011) wellbore annulus
Wang et al. (2012) Predicting mud density at HTHP
conditions
Al-Baiyat and Heinze Predicting stuck pipe before its
(2012) occurrence
Jahanbakhshi et al. Predicting differentially stuck
(2012) pipe
Chamkalani et al. (2013) Predicting stuck pipe
Li et al. (2014) Predicting wall factor effect on
particle settling
Shahdi and Arabloo Frictional Pressure Loss
(2016) estimation
308
Sorgun et al. (2014) Pressure losses prediction of
Newtonian and non-Newtonian
fluids
Jahanbakhshi and Prediction of the amount of lost
Keshavarzi (2015) circulation
Chhantyal et al. (2016) Viscosity estimation
Ahmadi (2016) Predicting the density of drilling
muds
Tatar et al. (2016) Prediction of brine density
Hoang (2016) Mud viscosity estimation
Manshad et al. (2017) Lost circulation prediction
Chhantyal et al. (2017) Prediction of mud flow rate
Kamari et al. (2017) Estimating drilling mud density
O.E. Agwu et al.
SVM Kernel function used Input Parameters Output parameter Errors (Performance evaluation criteria)
RMSE MAPE R2
Polynomial and Gaussian Flow rate, fluid density, plastic viscosity, borehole diameter of the Flow pattern – – 0.9528
RBF pipe (in); drill collar OD (in); mud type, WBM or OBM
Radial Basis Function Type of drilling fluid, initial surface density, temperature, pressure Density 0.117 0.872 0.9994
Linear kernel & Well direction characteristics, mud properties and drilling Stuck index – – –
RBF parameters
Gaussian Kernel function Differential pressure, hole depth, mud filtrate viscosity, fluid loss, Stuck, non stuck – – Accuracy = 92.2%
solid content, BHA length, still pipe time, hole size, 10 s and 10 min
gels, PV,YP
Radial Basis function TVD, fluid loss, differential pressure, formation loss, cross section Differential, 0.94
of annulus, ROP, RPM, measured depth, angle of well, calcium mechanical, non stuck
concentration, solid percent, water oil ratio, oil water ratio, open
hole, formation pressure, drill collar (OD), drill collar length, 10 s/
10 min gels, PV, YP, flow rate, pH
Radial Basis Function Fluid density, particle density, particle diameter, fluid consistency Wall factor 0.0272 0.9844
index, flow behaviour index, length, width, height of prism
– Rate of penetration, pipe rotation speed, hole inclination, and in- Pressure drop – – > 0.99
situ flow rate
– – Pressure loss AAPE < 5.1% AAPE5.98% non-Newt.
Newt.
Gaussian Kernel and Hole depth, porosity, formation permeability, differential pressure, Amount of lost – – 0.9851
Polynomial Kernel ROP, ECD, pump pressure (avg), temperature in lost interval, PV, circulation
functions YP, 10 s/10 min gels, Solid percent, Mud filtrate viscosity, API fluid
loss, Minimum horizontal stress, Tensile & Uniaxial compressive
strength, Natural fracture orientation, Young modulus
– Density and shear stress Viscosity – 2.74%
– Initial mud density, pressure, temperature Mud density – – 0.9999
– Temperature, pressure and concentration Brine density – – 1.0
Radial basis function Mud density, shear stress Mud viscosity – 2.7% –
Radial basis function Present driller depth of well in the day of study, Present depth of Amount of lost – – 0.795
well from sea level in the day of study, Well trajectory, Drilling circulation
time, Length of open hole section at the end of the day of study,
Asmari formation top from ground surface, bit size, pump flow rate
(avg), pump discharge pressure (avg), Mud weight, solid percent,
viscosity, Mud filtrate loss, Mud volume lost the day before the day
of study, Mud volume lost 2 days before the day of study, Rock
physical properties, Rock type & permeability of the interval,
Minimum horizontal stress profile
Radial basis function Ultrasonic level measurements (LT-1, LT-2 and LT-3) Mud flow rate 1.61% 0.99
Initial density, Density at pressure – 0.999
pressure and temperature of OBM, WBM, Colloidal Gas Aphron and
(CGA) and synthetic drilling fluids temperature
Journal of Petroleum Science and Engineering 167 (2018) 300–315
O.E. Agwu et al. Journal of Petroleum Science and Engineering 167 (2018) 300–315

Table 7 volume simulation and analytical calculations to develop a compre-

Summary of the strengths and weaknesses of various AI techniques. hensive model of Herschel-Bulkley fluid flow through porous media.
Benchmark ANN FUZZY SVM GA The result of the model shows that yield stress has a remarkable effect
on the flow of the fluid through the porous media.
Robustness against High High High High Song et al. (2017) reports the numerical simulation findings of the
noise
stability of a wellbore in the presence of nanoparticle based mud. The
Speed of convergence Slow Fast – Slow
Prone to overfitting? Yes but depends on – No –
findings from their work show that the filtrate from water-based muds
how the training is had a high likelihood to invade the shale matrix and increase the in-
done stability in shales. They however posited that the transmission of pore
Data Requirements Huge data required Small data – fluid pressure could be reduced with the addition of silica nanoparticles
required
to the water based mud.
Self-organization Yes – – No
Ability to generalize Yes – Yes – Wu et al. (2004) developed a numerical algorithm to simulate the
physics behind the invasion of mud-filtrate in vertical and directional
wellbores. The sensitivity studies from the work showed that the pro-
Huang et al. (1998) carried out numerical simulation on the effect of cess of mud filtrate invasion is influenced by the geometry of the
water based mud interaction with shale on the formation pressure and permeable beds gravity segregation, effective porosity, capillary pres-
swelling of shale. The simulation results indicate that drilling mud with sure, relative permeability, hydraulic communication between adjacent
potassium chloride salt has the least impact on shale formation whereas layers and permeability anisotropy. They concluded that standard
the mud with calcium chloride salt has the most impact on shale for- piston-like representations of drilling fluid filtrate invasion frequently
mation. used in interpretation of well logs can lead to erroneous interpretations
Lichun et al. (2014) carried out numerical simulation on the effect of measurements from wireline tools.
of fracture roughness on lost circulation dynamics of a drilling mud
exhibiting pseudoplastic behaviour. The results show that the pseudo- 6. Selected applications of molecular simulation in drilling fluid
plastic characteristic of power-law drilling fluid can lead to high fluid engineering
loss rate at the inception of the mud loss event. The result of their work
indicates that as the tortuosity and stiffness of the fracture increases, The following are areas where molecular simulations were applied
the lower the mud loss would be. They also added that increases in the in drilling fluid engineering.
length, opening, dip of the fracture, the higher the rate of mud loss.
Meng et al. (2015) investigated the pressure variation in wellbore 6.1. Fluid chemistry
caused by gas kick/injection. The model was numerically solved by the
finite volume method and advection upstream splitting method V Wang et al. (2015) used molecular dynamics (MD) modelling to
(AUSMV) scheme. The simulation results indicate that the gas and li- study the interlayer microstructures of montmorillonite (bentonite)
quid mass flow rates and well depth have a strong influence on the fused with 1-hexadecyl-3-methylimidazolium chloride monohydrate
variation of the bottom pressure caused by gas kick/injection. In (C12mimCl). The results of their work showed that the insertion of
comparison with experimental data, the simulation results showed good C12minCl into montmorillonite (bentonite) interlayer space influenced
agreement. the rheology of the system and improved various properties. They
Morais et al. (2009) worked on numerically simulating the flow further added that the organic clay composite formed was found to be
behaviour of non- Newtonian fluids in porous media. From their work, environmentally friendly and could be used in drilling fluid systems.
they concluded that the structure of the pore, the inertia and the yield
stress would generate a condition in which the fluid flow in the media is 6.2. Chemical reaction
increased.
Ouyang et al. (2013) used a combination of three dimensional finite Jin et al. (2017) carried out molecular simulations to assist in the

Table 8
Comparative studies of various AI techniques done by previous researchers.
Researcher Study conducted No. of data AI techniques Performance criteria
points compared

Oladunni and Trafalis (2011) Predicting fluid flow patterns in wellbore 92 ANN vs. SVM R2
annulus ANN SVM
0.9028 0.9528
Wang et al. (2012) Predicting the drilling fluid density of 339 ANN vs. SVM MAPE RMSE R2
WBM, OBM and SBM ANN SVM ANN SVM ANN SVM
4.343 0.872 0.575 0.117 0.9889 0.9994
Toreifi et al. (2014) Lost circulation prediction 1630 MNN vs. MLP R2
MLP MNN
0.745 0.964
Jahanbakhshi and Keshavarzi Lost circulation prediction 260 ANN vs. SVM ANN SVM
(2015) MSE MAE R MSE MAE R
5815.38 65.28 0.8770 766.98 20.66 0.9851
Ahmadi et al. (2016) Prediction of the density of mud 664 FIS vs. PSO-ANN vs. R2
GA-FIS FIS PSO-ANN GA-FIS
0.7237 0.9964 0.9397
Chhantyal et al. (2016) Viscosity estimation – ANN vs. SVM R2
ANN SVM
23% 2.74%
Chhantyal et al. (2016) Mud flow rate estimation 1800 DANN vs.Static ANN vs. MAPE
SVR DANN Static ANN SVR
5.6% 8.5% 7.7%

309
O.E. Agwu et al.
comprehension of the hidden mechanisms of clay damage. From their
studies, they proposed that clay damage occurs through two ways
namely-swelling and instability of the clay structure. They further
added that clays swell when they come in contact with water and the
swelling occurs in a series of distinct stages, forming one or more layers.
Their work subsequently revealed that the simulation results were in
good agreement with the experimental data for the sodium montmor-
illonite but not for potassium montmorillonite.
6.3. Drilling fluid additives
Anderson et al. (2010) used molecular dynamics to undertake si-
mulations of clay swelling inhibitor systems. From their work, the
structure and interaction of three different swelling inhibitor polymers
namely: poly (propylene oxide) –diamine, poly (ethylene glycol) and
poly (ethylene oxide)-diacrylate inhibitor molecules with montmor-
illonite clay was studied. Their results allowed them an interpretation
of bulk experimental data of the mode of interaction of swelling in-
hibitors with clay minerals at an atomistic and molecular level.
7. IN-SITU monitoring of drilling mud flow track route using
artificial intelligence
Variations in the downhole temperature and pressure of formations,
rock heterogeneities, inputs into and out of the drilled well and a host of
other factors contribute to making the measurement of drilling and
drilling fluid parameters a challenging one. Therefore, to ensure success
in drilling, it is incumbent on all drilling personnel to ensure that the
drilling process is monitored constantly, accurately and in real – time
(Ba et al., 2009; Sui et al., 2011). One aspect of the drilling process that
requires monitoring is the behaviour of the drilling fluid downhole, its
parameters and its flow track route (Thorsrud et al., 2000). To achieve
this, a real-time model of the drilling fluid flow route especially in
fractures or conductive pores is necessary. Hardware measurement
devices such as measurement while drilling (MWD) devices may pro-
vide real time measurements but may fail under harsh downhole con-
ditions (Sui et al., 2011). Soft sensors are useful in overcoming this
challenge. Soft sensors are essentially data driven tools that require a
model that details the relation between the measured data and the
quantities to be estimated (Leskens et al., 2008). This makes it possible
to estimate parameters which were hitherto difficult to be estimated
and then serve as the virtual eyes of the rig crew that enables them
make informed decisions (Verikas, 2009). In the literature, several soft
sensing tools abound (Leskens et al., 2008). One soft sensing tool
commonly used is the Kalman filter (KF) (Kalman, 1960). The Kalman
filter by definition is an efficient recursive data fusion algorithm that
estimates the future state of a variable of interest (which hitherto was
difficult to predict) by using a series of historical data measurements
that contains noise and several other errors (Aanonsen et al., 2009). The
following are the selected cases of using the Kalman filter to estimate
drilling fluid flow in wellbores.
Agu (2014) used the numerical method of the 1-D saint venant
equation to predict the rate of flow of fluid in both the steady and
dynamic states using a rectangular open channel venturi. The results
from the work indicate that it is only possible to determine the rate of
flow in the steady state condition. Using an ensemble Kalman filter in
combination with the 1-D saint venant equation, the work shows that it
is possible to predict in real-time the flow rate within an accuracy level
of 0.5%.
Agu and Lie (2014) in predicting the flow rate of fluids in real time
using a venturi channel applied the ensemble Kalman filter. In the
work, they report that this filter can be applied to a non-Newtonian
fluid with high flow velocity upstream within the throat where the flow
has slower velocities. The results indicates that the algorithms for real-
time predictions are able to predict the wavy characteristic of the flow
within ± 0.7% accuracy for the rate of flow and ± 1% accuracy for the
310
Journal of Petroleum Science and Engineering 167 (2018) 300–315
free surface level.
Aliouane et al. (2015) used two artificial intelligence techniques
namely: artificial neural network (multilayer perceptron) and fuzzy
logic to predict the pore pressure in two horizontal wells that were
drilled in the lower Barnett shale formation. The data from one of the
wells was used for the training of the fuzzy set and the neural network
while the data of another horizontal well was used for the general-
ization. They used four input parameters namely: the neutron porosity,
natural gamma ray, the slowness of the compression and shear wave.
The output parameter was the estimated pore pressure obtained when
Eaton's model was used. From the results they obtained, they reported
that the fuzzy logic was better than the neural network in predicting the
pore pressure of the shale gas reservoirs.
Ba et al. (2009) used marginalized particle filters to detect faults in
real time. The process model was developed in order to get a part which
represents the faulty case and another that represents the safety one.
The distribution of these two process states was updated through a
Kalman filter and the concerned state was involved by using a particle
filter. By using simulated data and real field data to carry out a per-
formance analysis indicate that the approach is efficient in terms for
fault detection and isolation tasks.
Bektas et al. (2015) studied how to design and apply a steady-state
Kalman filter to optimally predict in real-time the pore pressure of a
formation. This they did by combining outputs from both Eaton's model
and from logging while drilling (LWD). Based on their results, they
concluded that the Kalman filter can be a very effective tool for de-
tecting over-pressured zones before penetrating such zones. They also
reported that the developed algorithm can be used as a design tool in
both onshore and offshore drilling operations along with real-time LWD
data to optimally estimate the pore pressure of the next interval ahead
of the bit during the drilling operation.
Gu and Oliver (2007) in predicting multiphase flow through a
medium containing voids, used an ensemble Kalman filter. From their
work, the iterative filter – ensemble randomized maximum likelihood
filter (EnRML) gave accurate results for problems that were small,
linear and static in nature, and showed good agreement with the en-
semble Kalman filter (EnKF) for a problem that is linear and dynamic in
characteristic. Focusing on the problem of two-phase flow in a porous
medium, their work reveals that the trio of ensemble randomized
maximum likelihood filter (EnRML), ensemble Kalman filter (EnKF)
and conforming ensemble Kalman filter gave results that were similar
for the permeability and porosity (model variables) while the prediction
of saturation (the state variable) was more satisfactory from EnRML.
Keshavarzi and Jahanbakhshi (2013) used an artificial neural net-
work (feed-forward neural network with back-propagation algorithm-
FFBP and a generalized regression neural network -GRNN) to estimate
pore pressure gradient in a formation in Iran. The input parameters they
used include: porosity, permeability, rock density and true vertical
depth. The output parameter was the pore pressure gradient. The
models developed were compared with Eaton's method and were found
to be better than the calculated values by Eaton's method. The pore
pressure gradient values predicted by developed FFBP when compared
with GRNN show the best similarity and agreement with the measured
values of pore pressure gradient for the test data.
Ljones (2013) used unscented Kalman filters to estimate two drilling
fluid properties namely: drilling mud density and viscosity. He further
combined the filter results for density and viscosity with an unscented
Kalman filter estimating the mud pump pressure, choke pressure, flow
through the bit, and a geometry parameter. In each case, test data was
generated by Simulink and WeMod simulation. It was observed that
while the unscented Kalman filter estimating density and plastic visc-
osity had an acceptable performance for Simulink model generated test-
data, the unscented Kalman filters estimating density and plastic visc-
osity did not perform as desired in the case of the WeMod simulation.
Magalhães et al. (2016) designed a device that was able to con-
tinuously acquire experimental data of the electrical stability of oil
O.E. Agwu et al.
based mud and then compared the result against that obtained from
bench tests according to API. The device was installed and tested in an
automated drilling fluid production plant. In this plant, not only elec-
trical stability was measured on line, but also pressure, flow rate,
temperature, viscosity, density, solids content and size and water-in-oil
ratio. From their work, they found out that the on line and bench test
data showed good agreement, as well as with other authors results.
Magalhaes et al. (2014) evaluated the ability of a commercial and
built in property sensor to predict the rheological parameters of mud,
mud weight, water-oil content, emulsion electrical stability, fluid con-
ductivity and particle size distribution. They used the multi-layer per-
ceptron type of neural networks to connect the on-line equipment result
to increase the reliability of the mud properties determination. Their
findings indicate that the online measurements were in agreement with
conventional measurements.
Mahdianfar et al. (2013) used joint unscented Kalman filter to es-
timate some unknown model parameters and the bottom-hole pressure
(BHP) in managed pressure drilling. The results obtained from the study
show that reliable estimation of the BHP and uncertain parameters can
be made, even in circumstances when there are no measurements for
the BHP especially when connections are to be made.
8. Learnings from the review
Among the findings in relation to artificial intelligence applications
in the wellbore drilling fluid programme are:
• The numbers of research papers are indicative of the relevance and
growth of artificial intelligence in mud engineering. Focusing on the
contents of the available literature with information about various
applications of AI in mud engineering, we can roughly estimate that
about 54% papers are related to ANN, 22% are connected to support
vector machines, about 6% – to hybrid intelligent systems, a little
over 8% were dedicated to fuzzy logic, 3% to genetic algorithms and
about 6% to CBR while about 1% was dedicated to particle swarm
algorithm.
• Lost circulation and downhole mud density predictions were the
most studied aspects of mud engineering by the researchers whose
works were reviewed; showing that these are critical areas to the
wellbore drilling.
• Artificial intelligence techniques are unique in the ways in which
they obtain knowledge, depict, keep, apply the knowledge gained,
adapt to new problems, deal with variability, etc. Hence, it is always
vital to consider what technique would be best for predicting dif-
ferent phenomena of the wellbore drilling fluid program.
• The most commonly used artificial intelligence technique in mud
engineering is artificial neural network with multiple inputs and
single outputs; however, sometimes the problem under considera-
tion may involve multiple inputs and multiple outputs.
• With a plethora of data, there is need to choose carefully the most
important parameters in the data in order to decrease the size of the
datasets which in turn increases the predictive performance of the
models. Additionally, the commanding importance of data pre-
processing steps such as data normalization cannot be over-
emphasized given the boosting effect they impart to the perfor-
mance of AI models.
• While reviewing the various limitations of the existing AI techni-
ques, it was found that hybrid systems were proposed by some re-
searchers as robust ways to handle and overcome some of the lim-
itations of the other AI techniques.
9. Conclusions
This paper reviewed the artificial intelligence technologies in dril-
ling fluid engineering with recent research results and applications
presented. It started with the evolution of AI in the exploration and
311
Journal of Petroleum Science and Engineering 167 (2018) 300–315
production industry before delving into the many applications of AI in
mud engineering. On the basis of the above findings, the following
conclusions are drawn:
1. Mud engineering like other aspects of wellbore drilling engineering
presents an area occasioned by complex optimization and decision-
making.
2. Presently, AI techniques are at different stages of application in mud
engineering with artificial neural networks being the most used
while case based reasoning and particle swarm algorithm are the
least applied. This however does not preclude the fact that these two
techniques may be effectively applied in other areas of drilling en-
gineering.
3. The discussion presented in this paper shows that no single AI
technique can solve all problems; hence, the results achieved by
means of one tool can be better in certain areas than those obtained
using other methods. This was shown in the strength and weak-
nesses section of this work.
4. AI techniques when carefully chosen and blended can be com-
plementary in their function resulting in predictive capabilities that
are sometimes better than when they are used singly and the results
added together. For instance, merging the computational power of
ANNs with the symbolic level of processing associated with fuzzy
logic gives rise to a hybrid system that improves the speed of con-
vergence speed, improved ability to deal with faults and general
adaptability to varying conditions.
10. Recommendations for future research
1. From the review, multi-gene genetic programming-a variant form of
genetic programming was found to be unique in that it produced an
explicit model unlike other AI techniques. Hence, for comparative
purposes, it is recommended that its predictive capability be tested
vis a vis other AI techniques by using it to model other complex
phenomena relating to wellbore drilling.
2. A number of artificial intelligence techniques have been identified
that are most likely suitable for predicting and/or optimizing dril-
ling mud parameters and mud related problems with acceptable
levels of uncertainty. One of them is case based reasoning. To get the
most out of this technique requires a database of previous scenarios
and how they were solved. Thus, in order to benefit the industry, the
industry operators should make available historical data to CBR
researchers to enable them create a comprehensive database that
can diagnose more than one wellbore challenge at a time.
3. It is hoped that the review this paper presents would provide for the
reader a broad view of what artificial intelligence has accomplished
in drilling fluid engineering without necessarily limiting the reader's
imaginative thoughts about the existence of future research oppor-
tunities which are, literally speaking, limitless.
Acknowledgement
The authors would like to appreciate the management of the
University of Uyo for providing an enabling environment to carry out
this research.
Abbreviations
1-D One Dimensional
2-D Two Dimensional
AAPE Average Absolute Percentage Error
AI Artificial Intelligence
ANFIS Artificial Neuro Fuzzy Intelligent System
ANN Artificial Neural Network
AUSMV Advection Upstream Splitting Method V
BHP Bottom Hole Pressure; BP Back Propagation
O.E. Agwu et al. Journal of Petroleum Science and Engineering 167 (2018) 300–315

BPNN Back Propagation Neural Network Abdollahi, J., Carlsen, I.M., Randhol, P., Tenold, E., Haga, H.B., Jakobsen, T., 2008. A
CBR Case Based Reasoning case based approach to understand the complexity of causal connections related to
well integrity problems. In: IADC/SPE 111129–MS, Presented at the IADC/SPE
DANN Dynamic Artificial Neural Network Drilling Conference, Orlando, Florida, USA.
ELM Extreme Learning Machine Ablat, M.A., Qattawi, A., 2017. Numerical simulation of sheet metal forming: a review.
EnKF Ensemble Kalman Filter Int. J. Adv. Manuf. Technolology 89 (Issue 1 – 4), 1235–1250. https://doi.org/10.
1007/s00170-016-9103-5.
EnRML Ensemble Randomized Maximum Likelihood filter Ab Wahab, M.N., Nefti-Meziani, S., Atyabi, A., 2015. A comprehensive review of swarm
E&P Exploration and Production optimization algorithms. PLoS ONE 10 (Number 5). http://dx.doi.org/10.1371/
FFBPN Feed Forward Back Propagation Network journal.pone.0122827.
Adesina, F.A.S., Abiodun, A., Anthony, A., Olugbenga, F., 2015. Modelling the effect of
FFMLP Feed Forward Multilayer Perceptron temperature on environmentally safe oil based drilling mud using artificial neural
FFNN Feed Forward Neural Network network algorithm. Petroleum Coal J. 57 (Number 1), 60–70.
FIS Fuzzy Intelligent Systems Aanonsen, S.I., Naevdal, G., Oliver, D.S., Reynolds, A.C., Valles, B., 2009. The ensemble
Kalman filter in reservoir engineering – a review. Paper SPE 117274. SPE J. 393–412.
GA Genetic Algorithm
Agu, C.E., 2014. Model Based Estimation of Drilling Mud Flow Using a Venturi Channel.
GA – FIS Genetic Algorithm Fuzzy Intelligent System TUC, Porsgrunn.
GA – FIS Genetic Algorithm Fuzzy Intelligent System Agu, C.E., Lie, B., 2014. Smart sensors for measuring fluid flow using a venturi channel.
GA-RBF Genetic Algorithm Radial Basis Function In: Proceedings from the 55th Conference on Simulation and Modelling (SIMS 55),
Aalborg, Denmark, pp. 229–240.
GA-RBF Genetic Algorithm Radial Basis Function Agwu, O.E., Okon, A.N., Udoh, F.D., 2015. A comparative study of diesel oil and soybean
GFNN Generalized Feed Forward Neural Network oil as oil-based drilling mud. J. Petroleum Eng Hindawi Publishing Corporation.
HIS Hybrid Intelligent Systems https://doi.org/10.1155/2015/828451.
Ahmadi, M.A., 2016. Toward reliable model for prediction of drilling fluid density at
HTHP High Temperature High Pressure wellbore conditions: a LSSVM model. Neurocomputing J. 211, 143–149. https://doi.
KF Kalman Filter org/10.1016/j.neucom.2016.01.106.
LWD Logging While Drilling Ahmadi, M.A., Shadizadeh, S.R., Shah, K., Bahadori, A., 2016. An accurate model to
predict drilling fluid density at wellbore conditions. Egypt. J. Petroleum. https://doi.
MSE Mean Square Error org/10.1016/j.ejpe.2016.12.002.
MAE Maximum Absolute Error Al-Baiyat, I., Heinze, L., 2012. Implementing artificial neural networks and support vector
MAPE Maximum Absolute Percent Relative Error machines in stuck pipe prediction. In: Paper SPE 163370 Presented at the SPE Kuwait
International Petroleum Conference and Exhibition, Kuwait City, Kuwait.
MLP Multilayer Perceptron Alegre, L., 1991. Potential applications for artificial intelligence in the petroleum in-
MNN Modular Neural Network dustry. J. Petroleum Technol. 1306–1309 Society of Petroleum Engineers.
MWD Measurement While Drilling Aliouane, L., Ouadfeul, S., Boudella, A., 2015. Pore Pressure prediction in shale gas re-
servoirs using neural network and fuzzy logic with an application to Barnett Shale.
NFLTO No Free Lunch Theorem for Optimization
Geophys. Res. Abstr. 17, 2015–2723 EGU.
NMSE Normalized Mean Squared Error Ali, B.H., Adel, N.M., Afshin, T., Milad, A., Kim, P.L.T., Moonyong, L., Alireza, B., 2016.
NPT Non-Productive Time Prediction of frictional pressure loss for multiphase flow in inclined annuli during
OBM Oil Based Mud Underbalanced Drilling operations. Nat. Gas. Ind. B 3, 275–282. Elsevier Publishers.
https://doi.org/10.1016/j.ngib.2016.12.002.
PSA Particle Swarm Algorithm Alireza, M., Mohammad, N., Ahmad, A., 2011. Reducing consumed energy while drilling
PSO Particle Swarm Optimization an oil well through a deep rig time analysis. Adv. Petroleum Explor. Dev. CS 1
PSO-ANN Particle Swarm Optimization Artificial Neural Network (Number 1), 22–31 Canada Publishers.
Anderson, R.L., Greenwell, H.C., Suter, J.L., Jarvis, R.M., Coveney, P.V., 2010. Towards
PSO-ANN Particle Swarm Optimization Artificial Neural Network the design of new and improved drilling fluid additives using molecular dynamics
PSO-BP Particle Swarm Optimization Back Propagation simulations. Ann. Braz. Acad. Sci. 82 (Number 1), 43–60.
PSO-BP Particle Swarm Optimization Back Propagation Anifowose, F.A., Labadin, J., Abdulraheem, A., 2016. Hybrid intelligent systems in pet-
roleum reservoir characterization and modelling: the journey so far and the chal-
R2 Correlation Coefficient lenges ahead. J. Petroleum Explor. Prod. Technol. 7 (Issue 1), 251–263. Springer link
RBF Radial Basis Function Publishers. https://doi.org/10.1007/s13202-016-0257-3.
RBNN Radial Basis Neural Network Artificial Intelligence Applications Institute, 2014. Centre for Intelligent Systems and
Their Applications. School of Informatics. The University of Edinburgh. http://www.
RMSE Root Mean Square Error
aiai.ed.ac.uk/, Accessed date: 5 June 2017.
RNN Recurrent Neural Network Ataallahi, E., Shadizadeh, S.R., 2015. Fuzzy consequence modelling of blowouts in
SONFIS Self -Organizing Neuro Fuzzy Inference Systems Iranian drilling operations; HSE consideration. Saf. Sci. 77, 152–159. Elsevier
Publishers. https://doi.org/10.1016/j.ssci.2015.03.014.
SPE Society of Petroleum Engineers
Ba, A., Hbaieb, S., Mechbal, N., Vergé, M., 2009. Faults detection by marginalized particle
SVM Support Vector Machines; filters: application to a drilling process. In: Proceedings of the 7th IFAC Symposium
SVR Support Vector Regression on Fault Detection, Supervision and Safety of Technical Processes Barcelona, Spain.
SBM Synthetic Based Mud Baarimah, S.O., Gawish, A.A., BinMerdhah, A.B., 2015. Artificial intelligence techniques
for predicting the reservoir fluid properties of crude oil systems. Int. Res. J. Eng.
WBM Water Based Mud Technol. 2 (Issue 7).
Barati, R., Neyshabouri, S.A.A.S., Ahmadi, G., 2014. Development of empirical models
Funding with high accuracy for estimation of drag coefficient of flow around a smooth sphere:
an evolutionary approach. Powder Technol. 257, 11–19. Elsevier Publishers. https://
doi.org/10.1016/j.powtec.2014.02.045.
No funding was provided for this research by any individual or Barton, D.L., Sviokla, J., 1988. Putting Expert Systems to Work. Havard Business Review.
corporate organisation. https://hbr.org/1988/03/putting-expert-systems-to-work, Accessed date: 7 June
2017.
Behnoud far, P., Hosseini, P., 2016. Estimation of lost circulation amount occurs during
Conflicts of interest under balanced drilling using drilling data and neural network. Egypt. J. Petroleum
Elsevier Publishers. https://doi.org/10.1016/j.ejpe.2016.09.004.
Bektas, E., Miska, S.Z., Ozbayoglu, E.M., Yu, M., Takach, N., Velazquez-Cruz, D., Shahri,
No conflicts of interest exist.
M.P., 2015. Application of Kalman filter to predictions of pore pressure while drilling.
In: Paper SPE -174908-MS Presented at the SPE Annual Technical Conference and
References Exhibition, Houston, Texas, USA, . https://doi.org/10.2118/174908-MS.
Bello, O., Holzmann, J., Yaqoob, T., Teodoriu, C., 2015. Application of Artificial
Intelligence Methods in Drilling System Design and Operations: a Review of the State
Aamodt, A., Plaza, E., 1994. Case-based Reasoning: Foundational Issues, Methodological of the Art, vol. 5. De Gruyter International, pp. 121–139 Number 2.
Variations, and System Approaches. AI Communications, vol. 7. IOS Press, pp. 39–59 Bispo, V.D.d. S., Scheid, C.M., Calçada, L.A., da Cruz Meleiro, L.A., 2017. Development of
Number 1. an ANN-based soft-sensor to estimate the apparent viscosity of water-based drilling
Abdollahi, J., 2007. Analysing Complex Oil Well Problems through Case-based fluids. J. Petroleum Sci. Eng. 150, 69–73. https://doi.org/10.1016/j.petrol.2016.11.
Reasoning. Doctoral Thesis Submitted to the Department of Petroleum Engineering 030.
and Applied Geophysics. Norwegian University of Science and Technology, Bloys, B., Davis, N., Smolen, B., Bailey, L., Fraser, L., Hodder, M., 1994. Designing and
Trondheim. Managing Drilling Fluid. Schlumberger Resources. Oilfield Review. pp. 33–43.

312
O.E. Agwu et al.
Boman, K., 2014. Artificial Intelligence Software Aids Decision-making in Onshore
Drilling. Rigzone Incoporated. http://www.rigzone.com/news/oil_gas/a/133973/
Artificial_Intelligence_Software_Aids_DecisionMaking_in_Onshore_Drilling/?all=
HG2, Accessed date: 7 June 2017.
Bravo, C., Saputelli, L., Rivas, F., Pérez, A.G., Nikolaou, M., Zangl, G., de Guzman, Neil,
Mohaghegh, S., Nunez, G., 2012. State of the art of artificial intelligence and pre-
dictive analytics in the E&P industry: a technology survey. In: Paper SPE 150314
Presented at the SPE Western Regional Meeting Held in Bakersfield, California, USA, .
https://doi.org/10.2118/150314-PA.
Chamkalani, A., Shahri, M.P., Poordad, S., 2013. Support vector machine model: a new
methodology for stuck pipe prediction. In: Paper SPE 164003 Presented at the SPE
Middle East Unconventional Gas Conference and Exhibition, Muscat, Oman.
Chhantyal, K., Viumdal, H., Mylvaganam, S., 2017a. Soft sensing of non-Newtonian fluid
flow in open venturi channel using an array of ultrasonic level sensors—AI models
and their validations. Sensors 17 (Issue 11). http://dx.doi.org/10.3390/s17112458.
Chhantyal, K., Viumdal, H., Mylvaganam, S., 2017b. Ultrasonic level scanning for mon-
itoring mass flow of complex fluids in open channels — a novel sensor fusion ap-
proach using AI techniques. Sensors, IEEE. http://dx.doi.org/10.1109/ICSENS.2017.
8234010.
Chhantyal, K., Viumdal, H., Hoang, M., Mylvaganam, S., Elseth, G., 2016. Flowrate es-
timation using ultrasonic level sensors using dynamic artificial neural networks with
real time recurrent learning – a comparative study of models and practical im-
plementation. In: Paper Presented at the 9th Eurosim Congress on Modelling and
Simulation, Finland.
Courteille, J.M., Fabre, M., Hollander, C.R., 1986. An advanced solution: the drilling
adviser. J. Petroleum Technol.
Cristofaro, R.A.L., Longhin, G.A., Waldmann, A.A., de Sá, C.H.M., Vadinal, R.B., Gonzaga,
K.A., Martins, A.L., 2017. Artificial intelligence strategy minimizes lost circulation
non-productive time in Brazilian deep water pre-salt. In: Paper OTC-28034-MS
Presented at the Offshore Technology Conference. OTC Brasil, Rio de Janeiro, Brazil.
https://doi.org/10.4043/28034-MS.
Deregeh, F., Karimian, M., Nezamabadi-Pour, H., 2013. A new method of earlier kick
assessment using ANFIS. Iran. J. Oil Gas Sci. Technol. 2 (Number 1), 33–41.
Doraisamy, H., Ertekin, T., Grader, A., 1998. Key parameters controlling the performance
of neuro simulation applications in field development. In: SPE 51079, Proceedings,
SPE Eastern Regional Conference, Pittsburgh, PA.
Drilling Specialties Company (Online) The engineered lost circulation materials of drilling
specialties company. http://www.cpchem.com/bl/drilling/en-us/Documents/
THEENGINEEREDLOSTCIRCULATIONMATERIALSOF.pdf. (Accessed 18 September
2017).
Eberhart, R.C., Shi, Y., 2001. Particle swarm optimization: developments, applications
and resources. In: Proceedings of the IEEE Congress on Evolutionary Computation
(CEC), Seoul, South Korea. vol. 1. pp. 81–86. http://dx.doi.org/10.1109/CEC.2001.
934374.
Ejimofor, J.C., 2016. Open Venturi-channel Flow Metering of Non-Newtonian Fluids: a
Case Study with Emulated Drilling Fluids. Masters Degree Thesis Submitted to the
Department of System and Control Engineering, Faculty of Technology. University
College of Southeast Norway.
Elkatatny, S., 2017. Real-time prediction of rheological parameters of KCl water-based
drilling fluid using artificial neural networks. Arab J. Sci. Eng. 42, 1655–1665.
https://doi.org/10.1007/s13369-016-2409-7.
Elkatatny, S., Tariq, Z., Mahmoud, M., 2016. Real time prediction of drilling fluid rheo-
logical properties using artificial neural networks visible mathematical model (white
box). J. Petroleum Sci. Eng. 146, 1202–1210. https://doi.org/10.1016/j.petrol.2016.
08.021.
Feng, Y., Gray, K.E, K.E., 2017. Modelling lost circulation through drilling-induced
fractures. In: Paper SPE 187945 –PA. SPE Journal. Society of Petroleum Engineers.
https://doi.org/10.2118/187945-PA.
Firoozjaee, R.A., Khamehchi, E., 2014. A novel approach to assist history matching using
artificial intelligence. J. Chem. Eng. Commun. 202 (Issue 4) Taylor and Francis
Online. https://doi.org/10.1080/00986445.2013.852977.
Ghaffari, H., Shakeri, H., Sharifzadeh, M., 2009. Developing of new facets of indirect
modelling in the geosciences. In: Proceedings of the International Conference on Rock
Joints and Jointed Rock Masses, Tucson, Arizona, pp. 1–8. https://espace.curtin.edu.
au/handle/20.500.11937/33454.
Goldstein, E.B., Coco, G., 2014. A machine learning approach for the prediction of settling
velocity. Water Resour. Res. 50, 3595–3601. http://dx.doi.org/10.1002/
2013WR015116.
Gu, Y., Oliver, D.S., 2007. An iterative ensemble Kalman filter for multiphase fluid flow
data assimilation. In: Paper SPE 108438, SPE Journal. Society of Petroleum
Engineers, pp. 438–446.
Ho, T.L., Ehara, S., 2007. Oil Reservoir Properties Estimation by Fuzzy-neural Networks.
Memoirs of the Faculty of Engineering, vol. 67 Kyushu University Number 3.
Hoang, M., 2016. Tuning of Viscosity and Density of Non-Newtonian Fluids through
Mixing Process Using Multimodal Sensors, Sensor Fusion and Models. Masters Degree
Thesis Submitted to the Faculty of Electrical, it and Cybernetics. University College of
Southeast Norway.
Huang, H., Azar, J.J., Hale, A.H., 1998. Numerical simulation and experimental studies of
shale interaction with water based drilling fluid. In: Paper IADC/SPE 47796
Presented at the 1998 IADC/SPE Asia pacific Drilling Conference, Jakarta, Indonesia,
. https://doi.org/10.2118/47796-MS.
i-SCOOP, 2016. Artificial intelligence (AI) and cognitive computing: what, why and
where. https://www.i-scoop.eu/artificial-intelligence-cognitive-computing/,
Accessed date: 5 June 2017.
Isebor, O.J., Echeverría, C.D., Durlofsky, L., 2014. Generalized field-development opti-
mization with derivative-free procedures. SPE J. 19 (Number 5) (SPE-163631-PA).
313
Journal of Petroleum Science and Engineering 167 (2018) 300–315
https://doi.org/10.2118/163631-PA.
Jahanbakhshi, R., Keshavarzi, R., 2015. Quantitative and qualitative analysis of lost
circulation in natural and induced fractured formations: the integration of opera-
tional conditions and geomechanical parameters. Eur. J. Environ. Civ. Eng. 19
(Number 4), 418–444. https://doi.org/10.1080/19648189.2014.949872.
Jahanbakhshi, R., Keshavarzi, R., Jalili, S., 2014. Artificial neural network-based pre-
diction and geomechanical analysis of lost circulation in naturally fractured re-
servoirs: a case study. Eur. J. Environ. Civ. Eng. 18 (Number 3), 320–335. https://doi.
org/10.1080/19648189.2013.860924.
Jahanbakhshi, R., Keshavarzi, R., Shoorehdeli, M.A., Emamzadeh, A., 2012. Intelligent
prediction of differential pipe sticking by support vector machine compared with
conventional artificial neural networks: an example of Iranian offshore oil fields. SPE
Drill. Complet. https://doi.org/10.2118/163062-PA.
Jeirani, Z., Mohebbi, A., 2005. Artificial neural networks approach for estimating filtra-
tion properties of drilling fluids. J. Jpn. Petroleum Inst. 49 (Number 2), 65–70.
https://doi.org/10.1627/jpi.49.65.
Jin, L., Jamili, A., Huang, L., Perez, F., 2017. Modelling the mechanisms of clay damage
by molecular dynamic simulation. J. Geofluids Hindawi Publishing Corporation.
https://doi.org/10.1155/2017/1747068.
Kalman, R.E., 1960. A new approach to linear filtering and prediction problems. Trans.
ASME-Journal Basic Eng. 82 (Series D), 35–45. https://doi.org/10.1115/1.3662552.
Kamari, A., Gharagheizi, F., Shokrollahi, A., Arabloo, M., Mohammadi, A.H., 2017.
Estimating the drilling fluid density in the mud technology: application in high
temperature and high pressure petroleum wells. In: Mohammadi, A.H. (Ed.), Heavy
Oil. Nova Science Publishers, Inc, pp. 285–295.
Kamyab, M., Dawson, R., Farmanbar, P., Corval, Al, 2016. A new method to determine
friction factor of cuttings slip velocity calculation in vertical wells using Neural
Networks. In: Paper SPE – 182359- MS Presented at the SPE Asia Pacific Oil and Gas
Conference and Exhibition Held in Perth Australia, . https://doi.org/10.2118/
182359-MS.
Keshavarzi, R., Jahanbakhshi, R., 2013. Real-time prediction of pore pressure gradient
through an artificial intelligence approach: a case study from one of middle east oil
fields. Eur. J. Environ. Civ. Eng. 17 (Issue 8), 675–686. https://doi.org/10.1080/
19648189.2013.811614.
Khamehchi, E., Rahimzadeh Kivi, I., Akbari, M., 2014. A novel approach to sand pro-
duction prediction using artificial intelligence. J. Petroleum Sci. Eng. 123, 147–154.
https://doi.org/10.1016/j.petrol.2014.07.033.
Lea, G., 2015. The Struggle to Define what Artificial Intelligence Actually Means. Popular
Science. Bonnier Corporation. http://www.popsci.com/why-we-need-legal-
definition-artificial-intelligence, Accessed date: 12 June 2017.
Legg, S., Hutter, M., 2007. A collection of definitions of artificial intelligence. In:
Proceedings of the 2007 Conference on Advances in Artificial General Intelligence:
Concepts, Architectures and Algorithms. IOS Press, Amsterdam, pp. 17–24.
Leskens, M., de Kruif, B., Belfroid, S., Smeulers, J., Gryzlov, A., 2008. Downhole multi-
phase metering in wells by means of soft-sensing. In: Paper SPE 112046 Presented at
the 2008 SPE Intelligent Energy Conference and Exhibition, Amsterdam, Netherland.
Lichun, J., Mian, C., Bing, H., Zhen, S., Yan, J., 2014. Drilling Fluid Loss Model and Loss
Dynamic Behaviour in Fractured Formations. Research Institute of Petroleum
Exploration and Development, PetroChina, vol. 41. Elsevier BV Publishers, pp.
105–112. Issue 1. https://doi.org/10.1016/S1876-3804(14)60012-4.
Ljones, T., 2013. Drilling Mud Property Estimator. Masters Degree Thesis Submitted to
the Department of Engineering Cybernetics. Norwegian University of science and
Technology.
Lim, J., 2005. Reservoir properties determination using fuzzy logic and neural networks.
J. Petroleum Sci. Eng. 49 (Issues 3–4), 182–192. https://doi.org/10.1016/j.petrol.
2005.05.005.
Li, M., Zhang, G., Xue, J., Li, Y., Tang, S., 2014. Prediction of the wall factor of arbitrary
particle settling through various fluid media in a cylindrical tube using artificial in-
telligence. Sci. World J Hindawi Publishing Corporation. https://doi.org/10.1155/
2014/438782.
Luchian, H., Breaban, M.E., Bautu, A., 2015. On meta-heuristics in optimization and data
analysis. application to geosciences. In: Cranganu (Ed.), Artificial Intelligent
Approaches in Petroleum Geosciences. Springer International Publishing.
Magalhaes, S., Scheid, C.M., Calcada, L.A., Folsta, M., Martins, A.L., Marques deSa, C.H.,
2014. Development of on-line sensors for automated measurement of drilling fluid
properties. In: Paper SPE-167978-MS Presented at the IADC/SPE Drilling Conference
and Exhibition, Fort Worth, Texas, USA, . https://doi.org/10.2118/167978-MS.
Magalhães, S.C., Calçada, L.A., Scheid, C.M., Almeida, H., Waldmann, A.T.A., 2016.
Improving drilling performance with continuous online measurements of electrical
stability and conductivity in oil based drilling fluids. J. Petroleum Sci. Eng. 146,
369–379. https://doi.org/10.1016/j.petrol.2016.05.045.
Maghrabi, S., Kulkarni, D., Teke, K., Kulkarni, S.D., Jamison, D., 2014. Modelling of shale-
erosion behaviour in aqueous drilling fluids. In: Paper Presented at the SPE/EAGE
European Unconventional Resources Conference and Exhibition.
Mahdianfar, H., Pavlov, A., Aamo, O.M., 2013. Joint unscented Kalman filter for state and
parameter estimation in managed pressure drilling. In: European Control Conference
(ECC), Zürich, Switzerland.
Manshad, A.K., Rostami, H., Niknafs, H., Mohammadi, A.H., 2017. In: Heavy, Oil,
Mohammadi, Amir H. (Eds.), Integrated Lost Circulation Prediction in Drilling
Operation. Nova Science Publishers, Inc.
Mathworks (online) User Stories: Halliburton Makes Oil Exploration Safer Using MATLAB
and Neural Networks. https://www.mathworks.com/company/user_stories/
halliburton-makes-oil-exploration-safer-using-matlab-and-neural-networks.html.
(Accessed 26 February 2018).
McCarthy, J., 2007. What Is Artificial Intelligence? Computer Science Department,
Stanford University, Stanford, CA 94305. http://www-formal.stanford.edu/jmc/
O.E. Agwu et al.
whatisai.pdf, Accessed date: 12 June 2017.
Medsker, L.R., 1995. Hybrid Intelligent Systems. Kluwer Academic Publishers.
Meng, Y., Xu, C., Wei, N., Li, G., Li, H., Duan, M., 2015. Numerical simulation and ex-
periment of the annular pressure variation caused by gas kick/injection in wells. J.
Nat. gas Sci. Eng. 22, 646–655. Elsevier BV publishers. https://doi.org/10.1016/j.
jngse.2015.01.013.
Meschi, M.R., Shahbazi, K., Shahri, M.P., 2010. A new criteria to predict stuck occur-
rence. In: Paper SPE 128376 Presented at the SPE North Africa Conference and
Exhibition, Cairo, Egypt, . https://doi.org/10.2118/128376-MS.
Miri, R., Sampaio, J.H.B., Afshar, M., 2007. Development of artificial neural networks to
predict differential pipe sticking in Iranian offshore oil fields. In: Paper SPE-108500-
MS Presented at the International Oil Conference and Exhibition in Mexico, 27-30
June, Veracruz, Mexico, . https://www.onepetro.org/search?q=affiliation
%3A%28%22Petroleum+U.+of+Tech+Iran%22%29Lourenco, A. https://doi.
org/10.2118/108500-MS.
Moazzeni, A.R., Nabaei, M., Jegarluei, S.G., 2010. Prediction of lost circulation using
virtual intelligence in one of Iranian oilfields. In: Paper SPE 136992 Presented at the
34th Annual SPE International Conference and Exhibition Held in Tinapa – Calabar,
Nigeria, . https://doi.org/10.2118/136992-MS.
Moazzeni, A., Nabaei, M., Jegarluei, S.G., 2012. Decision making for reduction of non-
productive time through an integrated lost circulation prediction. Petroleum Sci.
Technol. 30 (Number 20), 2097–2107. https://doi.org/10.1080/10916466.2010.
495961.
Mohaghegh, D.S., 2000. Virtual- intelligence applications in petroleum engineering: part
1 – artificial neural networks. SPE 58046. J. Petrol. Technol.
Morais, A.F., Seybold, H., Herrmann, H.J., Andrade Jr., J.S., 2009. Non-Newtonian fluid
flow through three-dimensional disordered porous media. Phys. Rev. Lett. 103 (Issue
19). https://doi.org/10.1103/PhysRevLett.103.194502.
Murillo, A., Johanna Neuman, J., Samuel, R., 2009. Pipe sticking prediction and avoid-
ance using adaptive fuzzy logic modelling. In: Paper SPE-120128-MS Presented at the
SPE Production and Operations Symposium, Oklahoma City, Oklahoma, . https://
doi.org/10.2118/120128-MS.
Naderi, M., Khamehchi, E., 2017. Well placement optimization using metaheuristic bat
algorithm. J. Petroleum Sci. Eng. 150, 348–354. https://doi.org/10.1016/j.petrol.
2016.12.028.
Nikravesh, M., Kovscek, A.R., Jonston, R.M., Patsek, T.W., 1996. Prediction of formation
damage during the fluid injection into fractured low permeability reservoirs via
neural networks. In: SPE 31103, Proceedings, SPE Formation Damage Symposium,
Lafayette, LA, . https://doi.org/10.2118/31103-MS.
Nilsson, N., 2010. The Quest for Artificial Intelligence. Cambridge University Press,
Cambridge, pp. 13.
Oladunni, O.O., Trafalis, T.B., 2011. Single-phase fluid flow classification via learning
models. Int. J. General Syst. 40 (Number 5), 561–576.
Osman, E.A., Aggour, M.A., 2003. Determination of drilling mud density change with
pressure and temperature made simple and accurate by ANN. In: Paper SPE 81422
Presented at the SPE 13th Middle East Oil Show and Conference, Bahrain, . https://
doi.org/10.2118/81422-MS.
Ouyang, L., Hill, A.D., Zhu, D., 2013. Theoretical and numerical simulation of Herschel-
Bulkley fluid flow in propped fractures. In: Paper SPE-163841-MS Presented at the
SPE Hydraulic Fracturing Technology Conference, the Woodlands, Texas, USA, .
https://doi.org/10.2118/163841-MS.
Ozbayoglu, E.M., Ozbayoglu, M.A., 2009. Estimating flow patterns and frictional pressure
losses of two-phase fluids in horizontal wellbores using artificial neural networks.
Petroleum Sci. Technol. 27, 135–149. Taylor and Francis Group, LLC. https://doi.
org/10.1080/10916460701700203.
Petrus, J.B., Thuijsman, F., Weijters, A.J., 1995. Artificial Neural Networks: an
Introduction to ANN Theory and Practice. Springer,, UK.
Popa, A.S., Cassidy, S., 2012. Artificial Intelligence for heavy oil assets: the evolution of
solutions and organization capability. In: Paper SPE 159504 Presented at the SPE
Annual Technical Conference and Exhibition, San Antonio, Texas, USA, . https://doi.
org/10.2118/159504-MS.
Rable, B., 2017. The future is here: 3 ways AI roots itself in O&G in the surge Magazine.
http://thesurge.com/stories/future-artificial-intelligence-roots-oil-gas-industry,
Accessed date: 5 June 2017.
Raja, H.Z., Sormo, F., Vinther, M.L., 2011. Case-based reasoning: predicting real-time
drilling problems and improving drilling performance. In: Paper SPE-141598-MS
Presented at the SPE Middle East Oil and Gas Show and Conference, Manama,
Bahrain, . https://doi.org/10.2118/141598-MS.
Rasouli, H., Rashidi, F., Karimi, B., Khamehchi, E., 2015. A surrogate integrated pro-
duction modelling approach to long-term gas-lift allocation optimization. Chem. Eng.
Commun. 202 (Number 5), 647–654.
Razi, M.M., Mazidi, M., Razi, F.M., Aligolzadeh, H., Niazi, S., 2013. Artificial neural
network modelling of plastic viscosity, yield point, and apparent viscosity for water-
based drilling fluids. J. Dispersion Sci. Technol. 34 (Number 6), 822–827. http://dx.
doi.org/10.1080/01932691.2012.704746.
Reingold, E., Nightingale, J., 1999. Artificial Intelligence Tutorial Review (PSY371)-
Symbolic AI. http://www.psych.utoronto.ca/users/reingold/courses/ai/symbolic.
html, Accessed date: 31 December 2017.
Rooki, R., 2015. Estimation of pressure loss of Herschel–Bulkley drilling fluids during
horizontal annulus using artificial neural network. J. Dispersion Sci. Technol. 36,
161–169. Taylor and Francis Group, LLC. https://doi.org/10.1080/01932691.2014.
904793.
Rooki, R., 2016. Application of general regression neural network (GRNN) for indirect
measuring pressure loss of Herschel–Bulkley drilling fluids in oil drilling. J. Meas. 85,
184–191. https://doi.org/10.1016/j.measurement.2016.02.037.
Rooki, R., Rakhshkhorshid, M., 2016. Cuttings transport modelling in underbalanced oil
314
Journal of Petroleum Science and Engineering 167 (2018) 300–315
drilling operation using radial basis neural network. Egypt. J. Petroleum Elsevier
Publishers. https://doi.org/10.1016/j.ejpe.2016.08.001.
Rooki, R., Ardejani, F.D., Moradzadeh, A., Kelessidis, V.C., Nourozi, M., 2012a. Prediction
of terminal velocity of solid spheres falling through Newtonian and non-Newtonian
pseudoplastic power law fluid using artificial neural network. Int. J. Mineral Process.
110–111, 53–61. Elsevier Publishers. https://doi.org/10.1016/j.minpro.2012.03.
012.
Rooki, R., Ardejani, F.D., Moradzadeh, A., 2014. Hole cleaning prediction in foam drilling
using artificial neural network and multiple linear regression. Geomaterials 4, 47–53.
https://doi.org/10.4236/gm.2014.41005.
Rooki, R., Ardejani, F.D., Moradzadeh, A., Mirzaei, H., Kelessidis, V., Maglione, R.,
Norouzi, M., 2012b. Optimal determination of rheological parameters for
Herschel–bulkley drilling fluids using genetic algorithms (GAs). Korea-Australia
Rheology J. 24 (Issue 3), 163–170. https://doi.org/10.1007/s13367-012-0020-3.
Ross, T.J., 2010. Fuzzy Logic with Engineering Applications, third ed. John Wiley and
Sons Ltd.
Rui, C., Qian, L., Hu, Y., 2013. Application of case-based reasoning technology in drilling
risk prediction. CNKI J (China Academic Journals Electronic Publishers).
Shahin, M.A., Jaksa, M.B., Maier, H.R., 2009. Recent advances and future challenges for
artificial neural systems in geotechnical engineering applications. Adv. Artif. Neural
Syst. http://dx.doi.org/10.1155/2009/308239. Hindawi Publishing Corporation.
Schatsky, D., Muraskin, C., Gurumurthy, R., 2014. Demystifying Artificial Intelligence.
What Business Leaders Need to Know about Cognitive Technologies. Delloite
University Press. https://dupress.deloitte.com/dup-us-en/focus/cognitive-
technologies/what-is-cognitive-technology.html?id=
gx:2el:3dc:dup1030:eng:dup:cogtech#endnote-11, Accessed date: 12 June 2017.
Schlumberger, 2017. Mud Program. http://www.glossary.oilfield.slb.com/Terms/m/
mud_program.aspx, Accessed date: 20 May 2017.
Shadizadeh, S.R., Karimi, F., Zoveidavianpoor, M., 2010. Drilling stuck pipe prediction in
Iranian oil fields: an artificial neural network approach. Iran. J. Chem. Eng. 7
(Number 4).
Shadravan, A., Tarrahi, M., Amani, M., 2015. Intelligent tool to design fracturing, drilling,
spacer and cement slurry fluids using machine learning algorithms. In: Paper SPE
175238 Presented at the SPE Kuwait Oil and Gas Show and Conference, Kuwait, .
https://doi.org/10.2118/175238-MS.
Shahdi, A., Arabloo, M., 2016. Application of SVM algorithm for frictional pressure loss
calculation of three phase flow in inclined annuli. J. Petroleum Environ. Biotechnol. 5
(Number 3).
Shelley, R., Massengill, D., Scheuerman, P., McRill, P., Hamilton, R., 1998. Granite wash
completion optimization with the aid of artificial neural networks. In: SPE 39814,
Proceedings, Gas Technology Symposium, Calgary, Alberta, . https://doi.org/10.
2118/39814-MS.
Sheremetov, L., Batyrshin, I.Z., Filatov, D., Martinez, J., Rodriguez, H., 2008. Fuzzy ex-
pert system for solving lost circulation problem. J. Appl. Soft Comput. 14–29. Elsevier
Publishers. https://doi.org/10.1016/j.asoc.2006.11.003.
Shokouhi, S.V., Skalle, P., 2009. Enhancing decision making in critical drilling operations.
In: Paper SPE 120290, Prepared for Presentation at the SPE Middle East Oil & Gas
Show and Conference Held in the Bahrain, . https://doi.org/10.2118/120290-MS.
Siruvuri, C., Nagarakanti, S., Samuel, R., 2006. Stuck pipe prediction and avoidance: a
convolutional neural network approach. In: Paper IADC/SPE 98378 Presented at the
IADC/SPE Drilling Conference, Miami, Florida, U.S.A.
Skalle, P., Sveen, J., Aamodt, A., 2000. Improved efficiency of oil well drilling through
case-based reasoning. In: Proceedings of PRICAI 2000, Sixth Pacific Rim International
Conference on Artificial Intelligence. Melbourne.
Song, J., Yuan, Y., Gu, S., Yang, X., Yue, Y., Cai, J., Jiang, G., 2017. 2D Numerical
Simulation of Improving Wellbore Stability in Shale Using Nanoparticles Based
Drilling Fluid. Energies, vol. 10http://dx.doi.org/10.3390/en10050651.
Sorgun, M., Ozbayoglu, A.M., Ozbayoglu, M.E., 2014. Support vector regression and
computational fluid dynamics modelling of Newtonian and non-Newtonian fluids in
annulus with pipe rotation. J. Energy Resour. Technol. 137 (Number 3).
Steinwart, I., Christmann, A., 2008. Support Vector Machines. Springer Publishers, New
York.
Sui, D., Nybø, R., Gola, G., Roverso, D., Hoffmann, M., 2011. Ensemble methods for
process monitoring in oil and gas industry operations. J. Nat. Gas Sci. Eng. 3 (Issue 6),
748–753. https://doi.org/10.1016/j.jngse.2011.05.004.
Sung, W., Hanyang, U., Yoo, I., 1995. Development of HT-BP neural network system for
the identification of well test interpretation model. In: SPE 30974, Proceedings, SPE
Eastern Regional Conference and Exhibition, Morgantown, West Virginia.
Tatar, A., Halali, M.A., Mohammadi, A.H., 2016. On the estimation of the density of brine
with an extensive range of different salts composditions and concentrations. J.
Thermodyn. Catal. 7 (Number 167). http://dx.doi.org/10.4172/2160-7544.
1000167.
Ternyik, J., Bilgesu, I., Mohaghegh, S., Rose, D., 1995a. Virtual measurement in pipes,
part 1: flowing bottomhole pressure under multi-phase flow and inclined wellbore
conditions. In: Paper SPE 30975, Proceedings, SPE Eastern Regional Conference and
Exhibition, Morgantown, West Virginia.
Ternyik, J., Bilgesu, I., Mohaghegh, S., 1995b. Virtual measurement in pipes, Part 2: li-
quid holdup and flow pattern correlation. In: SPE 30976,.Proceedings, SPE Eastern
Regional Conference and Exhibition, Morgantown, West Virginia.
Thorsrud, A.K., Ekeli, Ø., Hilbig, N.C.C., Bergsvik, O., Zamora, M., 2000. Application of
novel downhole hydraulics software to drill safely and economically a north sea high-
temperature/high-pressure exploration well. In: Paper SPE-59189-MS Presented at
the IADC/SPE Drilling Conference, New Orleans, Louisiana, . https://doi.org/10.
2118/59189-MS.
Toreifi, H., Rostami, H., Manshad, A.K., 2014. New method for prediction and solving the
problem of drilling fluid loss using modular neural network and particle swarm
O.E. Agwu et al.
optimization algorithm. J. Petroleum Explor. Prod. Technol. 4, 371–379. https://doi.
org/10.1007/s13202-014-0102-5.
Valipour, S., Shokouhi, S., Skalle, P., Aamodt, A., 2014. An overview of case-based rea-
soning applications in drilling engineering. Artif. Intell. Rev. 41 (Issue 3), 317–329
Springer Link Publishers.
Verikas, A., 2009. Soft Sensors for Monitoring. Intelligent Systems Lab, Halmstad
University, Halmstad, Sweden.
Wang, G., Pu, X.L., Tao, H.Z., 2012. A support vector machine approach for the prediction
of drilling fluid density at high temperature and high pressure. J. Petroleum Sci.
Technol. 30 (Issue 5) Taylor and Francis Online. https://doi.org/10.1080/
10916466.2011.578095.
Wang, X., Liu, B., Yu, P., 2015. Research on the preparation and mechanism of the organic
montmorillonite and its application in drilling fluid. J. Nanomater Hindawi
Publishing Corporation. https://doi.org/10.1155/2015/514604.
Wang, Y., Duan, M., Wang, D., Liu, J., Dong, Y., 2011. A model for deepwater floating
platform selection based on BP artificial neural networks. Int. Soc. Offshore Polar
Eng.
Wolpert, D.H., Macready, W.G., 1997. No free lunch theorems for optimization. IEEE
315
Journal of Petroleum Science and Engineering 167 (2018) 300–315
Trans. Evol. Comput. 1 (Issue 1), 67–82. http://dx.doi.org/10.1109/4235.585893.
Wu, J., Torres- Verdin, C., Sepehrnoori, K., Delshad, M., 2004. Numerical simulation of
mud-filtrate invasion in deviated wells. In: Paper (SPE 87919) Revised for Publication
from Paper SPE 71739, First Presented at the 2001 SPE Annual Technical Conference
and Exhibition. SPE Reservoir Evaluation & Engineering, New Orleans, pp. 143–154.
Yunhu, L., Mian, C., Yan, J., Bing, H., Lichun, J., Hui, H., 2011. Identification of Leak
Zone Pre-drilling Based on Fuzzy Control. Trans Century Training programme
Foundation. IEEE, pp. 353–356 pp.
Zaimy, S.S., Rasaei, M.R., 2013. Reconstruction of porosity distribution for history
matching using genetic algorithm. J. Petroleum Sci. Technol. 31 (Issue 11) (Taylor
and Francis Online).
Zhou, H., Niu, X., Fan, H., Wang, G., 2016. Effective calculation model of drilling fluids
density and ESD for HTHP well while drilling. In: Paper IADC/SPE-180573-MS
Presented at the 2016 IADC/SPE Asia Pacific Drilling Technology Conference,
Singapore, . https://doi.org/10.2118/180573-MS.
Zhu, D., Liu, G., Zhang, Q., 2013. Research of pre-warning pipe-sticking based on neural
network. Appl. Mech. Mater. 325–326, 1734–1737. https://www.scientific.net/
AMM.325-326.1734.