See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/326479738

REVIEW ON CATHODIC PROTECTION SYSTEM MODELLING

Article · December 2017

CITATION READS

1 1,247

2 authors:

Adnan Mujezinovic Irfan Turkovic

University of Sarajevo University of Sarajevo
50 PUBLICATIONS   90 CITATIONS    40 PUBLICATIONS   93 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Research on the influence of increased voltage in the power system of Bosnia and Herzegovina on the formation of the stationary corona View project

Investigation of impulse characteristics of power line towers View project

All content following this page was uploaded by Adnan Mujezinovic on 19 July 2018.

The user has requested enhancement of the downloaded file.

 Volume 11 January/December 2017 81

Review scientific paper/Pregledni naučni rad

REVIEW ON CATHODIC PROTECTION

SYSTEM MODELLING

OSVRT NA MODELIRANJE SISTEMA

KATODNE ZAŠTITE
Adnan Mujezinović1, Irfan Turković1

Abstract: Cathodic protection is most frequently used technique for protection of underground or underwater metallic infrastructures from
corrosion. Design of the cathodic protection system implies determination of the electric potential and current density distribution on the
electrode surfaces. These values must meet criterions defined by relevant standards. For the calculation of the electrical potential and current
density distribution, different analytical and numerical methods can be used. In this paper, a review of the methods that are used for calculate
of the cathodic protection system parameters is given. Also, due to frequent use, a mathematical model based on the boundary element
method for the calculation of the cathodic protection system parameters is presented in more detail. Presented mathematical model was
used for calculation of the cathodic protection system parameters of one geometry simple example.

Keywords: Cathodic protection system, current density distribution, electric potential, boundary conditions

Sažetak: Katodna zaštita je najčešće korištena tehnika za zaštitu podzemnih i podmorskih metalnih instalacija od korozije. Projektiran-
je sistema katodne zaštite podrazumjeva određivanje raspodjele električnog potencijala i gustoće struje na elektrodnim površinama.
Ove vrijednosti moraju zadovoljiti kriterije definirane relevantnim standardima. Za proračun raspodjele električnog potencijala i gustoće
zaštitne struje, različite analitičke i numerčke metode mogu biti korištene. U ovom radu je dat pregled metoda koje se koriste za proračun
navedenih parametara sistema katodne zaštite. Također, zbog česte upotrebe, detaljnije je predstavljen matematički model baziran na
metodi graničnih elemenata za proračun parametara sistema katodne zaštite. Predstavljeni matematički model je iskorišten za proračun
parametara jednog geometrijski jednostavnog sistema katodne zaštite.

Ključne riječi: Sistem katodne zaštite, raspodjela gustoće zaštitne struje, električni potencijal, granični uslovi

INTRODUCTION reason for the adequate application of protection against

corrosion. Corrosion-induced costs affect industry, gov-
With the constant increase in energy demand as a general ernments and individuals, and their share depends on
civilization trend, the number of metal installations such the degree of development of society, the political system
as metallic pipelines for the transport and distribution of and the lifestyle [3]. Extensive researches conducted in the
gas and oil is growing. Also, a large number of indus- United States over the period from 1998 to 2001 showed
trial production branches use different types of metallic that annual costs due to the devastating effects of metal
installations such as of pipelines, reservoirs and other corrosion for 1998 amounted to about 276 billion US $,
underground installations, enabling demanding industrial which is about 3.1% of the gross national income. Studies
activities [1]. However, the application of underground or conducted in Japan and the European Community coun-
underwater metal installations has one problem. Namely, tries show that costs caused by corrosion are at a level of
non-noble metals or their alloys are mainly used as mate- 4.2 to 4.5% of gross national income [4], [5]. The present-
rial for the production of these installations. These types of ed data are incomplete because they do not take into the
materials are highly susceptible to corrosion destruction. account indirect corrosion-induced consequences such
The problem of corrosion of underground installations has as physical disasters, human health hazards, ecological
to be solved because of technical, but also economic and catastrophes, production losses, etc. It is assumed that
ecological reasons [2]. the costs of indirect effects can be estimated and their
amount is approximately equal to the costs of direct con-
The economic aspects of corrosion of underground or un- sequence of corrosion [4]. Recent research by the US
derwater metallic installations are certainly the dominant company G2MT Lab indicates that annual costs by cor-
rosion in the United States, taking into the account direct
Faculty of Electrical Engineering, University of Sarajevo,
1 and indirect consequences in 2013, amounted to more
Bosnia and Herzegovina than 1,000 billion US $, or about 6.2% of gross national
[email protected] income, and thus represents the highest individual cost
Paper submitted: October 2017 Paper accepted: December 2017
of the economy [5], [6]. Also, it is considered that almost
82 B&H Electrical Engineering Bosanskohercegovačka elektrotehnika

20% of the world's metallurgical production (production are used [14], [16], [17]. To date, a number of analytical
of metal from their ore) is only used to compensate for models and models based on numerical methods used
loss of lost metal in the corrosion process [7]. for calculations when designing and applying cathodic
protection systems have been developed.
Previous research has shown that the greatest percent-
age of damage caused by corrosion can be prevented by
The purpose of cathodic protection system modelling is to
the use of modern protection techniques. The metal can
determine the distribution of electrical potential and pro-
be protected from corrosion by using different methods.
However, technical practice has shown that one of the tection current density on the electrode surfaces as well
most effective methods of protecting metallic structures as at the corresponding points in/on electrolyte (usually at
from corrosion is the application of the cathodic protection the points where the reference electrode is set for the pur-
system in combination with passive protection (insulating pose of verifying the measured values) [7], [14], [16]. The
coatings based on polymers or concrete) [2], [7] – [13]. The calculation of electric potential and current density distri-
cost of installing the cathodic protection system is approx. bution comes down to solving the well-known Laplace
0.1 - 1% of the total investment value of the object being partial differential equation. Also, the correct calculation of
protected. the parameters of the cathodic protection system implies
the definition of the appropriate boundary conditions at
The application of the cathodic protection system to un- the overall boundary of the calculation domain. Bound-
derground metal pipelines results in extending the pro-
ary conditions in the cathodic protection system are
tected object’s life span for 10 to 25 years, depending on
most commonly defined through analytical expressions
the chosen solution of the cathodic protection system [2].
for functional relationships between the protection cur-
Modelling of corrosive processes implies the definition of rent density and the electrical potential on the electrodes
the simplest possible mathematical model with the invari- surfaces [18] – [20]. The establishment of boundary con-
ability of the basic characteristics of the electrochemical ditions on electrode surfaces in cathodic protection sys-
system. Practical engineering calculations can simplify tems is based on the physical and analytical definitions of
the mathematical model, depending on the level of the electrochemical mechanisms of electrode reactions. The
required accuracy of the calculation. In case of complex total anodic and cathodic current density is the result of
geometries of the system, calculation method can be electrochemical reactions, which occur simultaneously on
simplified only in cases of objects with a high degree of the electrodes surfaces. Partial electrochemical reactions
symmetry, which can be adequately modelled as one-di- occurring simultaneously on the electrode surfaces result
mensional or two-dimensional problems [7].
in a nonlinear relationship between the electrical potential
and the current density i.e. nonlinear polarization charac-
The design of the complex geometry objects installed in
different corrosive aggressive environments (electrolytes) teristics as boundary conditions.
results in the introduction of more sophisticated scientif-
ic methods of calculating the parameters of the cathod- 1.1. Analytical models
ic protection system. The development of underground
metal infrastructure increases the need for application of Analytical models used in designing of the cathodic pro-
cathodic protection systems [14]. More intensive applica- tection system can generally be divided into two types.
tion of cathodic protection systems results in the devel- The first type is based on basic principles of electrical engi-
opment of numerical methods for solving the distribution neering [9], [21] - Ohm's law and Kirchhoff's laws. In such
of electric potential and current density of these systems. models, anodes and protected objects are equivalented
by a combination of appropriate DC voltages sources and
Defining the spatial distribution of cathode protection resistors, while electrolytes resistance and resistance of
currents density is a preliminary assumption for the opti- the connecting wires are represented by resistors. These
mal design of electrochemical protection of underground
types of models are very simple for application, since only
metallic structures. The cathodic protection system must
basic laws of electrical engineering are applied. On the
be designed to ensure the highest possible homogenous
distribution of the protection current density on the sur- other hand, this type of models has a number of disad-
face of the protected object [15]. vantages, consisting in the fact that by representing the
electrodes simply by resistors, the current density at all
points on the electrode surface is obtained as a constant.
This is a rough approximation of most real cases of ca-
1. LITERATURE REVIEW ON CATHODIC PROTECTION
MODELLING METHODS thodically protected objects. Also, polarization effects of
the cathodic protection system cannot be taken into ac-
For the purpose of determining the relevant parameters of count with simple elements such as linear resistors. By
the cathodic protection system parameters, mathematical introducing nonlinear (variable) resistors, which represent
models based on calculation of distribution of electrical the polarization resistance of the electrodes, the previous
potential and current density on the electrode surfaces model can be improved [22], [23]. However, the deter-
 Volume 11 January/December 2017 83

mination of the value of these nonlinear resistors is only discretization of the entire domain (electrolyte) considered
possible for linearized polarization characteristics, which as a semi-infinite domain and due to a large difference in
make this model unacceptable for modelling realistic ca- the subdomain dimensions (electrolyte as a semi-infinite
thodic protection systems. subdomain and, for example, the pipeline wall thickness).
This results in a large number of finite elements and there-
Second type of analytical models is based on calculat- fore a large number of algebraic equations, and conse-
ing the stationary current field by solving Laplace partial quently requires a long time for calculation.
differential equation (LPD) using analytical methods such
as variable separation, obtaining solutions in the form of 1.2.3. Boundary element method (BEM)
infinite series, superposition method, confident mapping
method, etc. The application of this approach is limited One of the most efficient methods for solving problems
to simpler geometries with high the degree of symmetry in the field of application of cathodic protection systems
such as cathodic protection of the inner wall of the well, is the boundary element method (BEM). The basic ad-
assuming that the working point of the cathodic protec- vantage of this method is that discretization is required
tion system is on the linear part of polarization character- only at the boundary of the domain of interest and there
istics [14], [23]. is no need for discretization of infinite boundaries [30] –
[32]. For the modelling of the cathodic protection system
of underground or underwater metallic installations using
1.2. Numerical models the boundary element method, Laplace's partial differ-
ential equation (LPD), which is valid for the considered
1.2.1. Finite difference method (FDM) domain (electrolyte), is transformed into the integral field
equation that only treats the boundary surfaces [33]. The
In the early seventies of the last century, the finite different discretization of this equation as well as the application of
method was used to solve the problem of the distribu- boundary conditions leads to the formation of the system
tion of electrical potential and the current density of the of algebraic equations that need to be solved. For non-
cathodic protection system [24]. This method requires linear boundary conditions, such as the case of cathod-
discretization of the entire domain of interest with a net- ic protection system modelling, the system of equations
work. Solutions are located in network nodes. Although is nonlinear, and it is necessary to use iterative methods
by using this method results with satisfactory accuracy such as the Newton - Raphson method [3], [34] – [38].
are obtained, it is very rarely used when modelling realistic
cathodic protection systems because of the need for dis- As the most suitable method for calculating the param-
cretization of the entire domain. The disadvantages that eters of the cathodic protection system of underground
have eliminated this method from this field are long time metallic installations, this method has been improved
needed for calculations, especially in the case of complex during the last three decades through hybridization with
underground structures that need to be modelled in 3D other mathematical methods. As such, it has been used
space. Long time needed for calculation is caused by a for specific calculations in the application of the cathodic
large number of nodes required for calculation. protection system. Below are some interesting applica-
tions of these types of methods for specific calculations in
1.2.2. Finite element method (FEM) the application of the cathodic protection system:

The finite element method is one of the most frequently - for the analysis of the impact of the horizontal layered
used numerical methods in science and engineering. It soil on the level of polarization of the protected pipe-
is characterized by robustness, modularity and multidis- lines by the cathodic protection system as well as
ciplinary. In the finite element method, the entire domain the wells. Two approaches based on the boundary
of interest is divided into the volumetric elements of the element method were proposed. The first approach,
finite dimensions. For each finite element, a set of algebra- called the MRBEM (Multi Region Boundary Element
ic equations is written, by solving those yields the distri- Method), where all boundaries between neighbour-
bution of fields within each finite element. Unlike the finite ing soil layers are to be discretized by the network
difference method, the finite element method is algorith- of boundary elements [19], [39], [40] as well as the
mic and relatively simple to program for general equations MLBEM (Multi-Layer Boundary Element Method) in
of electromagnetic fields [14], [16]. As with the modelling which the layered soil is treated by applying the ap-
of most engineering problems, the finite element meth- propriate Green's functions [19], [30], [35], [38], [40],
od has been found to be applied in the field of cathodic [41];
protection systems [25] – [29]. However, this method has
one disadvantage when modelling the cathodic protec- - to determine the optimal anode location and output
tion system of underground pipelines due to the need for current of the cathodic protection systems of under-
84 B&H Electrical Engineering Bosanskohercegovačka elektrotehnika

ground pipeline or the ship’s hull, many authors used surface domains, so it is suitable for calculating distribu-
a hybrid method based on the boundary element tion of electric potential in the electrolyte. On the other
method for the calculation of the distribution of elec- hand, the finite element method is suitable for solving
tric potential and optimization methods such as gra- stationary current field of bounded domains; therefore it can
dient method, Gauss - Newton method, a simulated be applied for the field calculation within the electrodes. The
annealing method [42], a genetic algorithm [43], or potential difference on the electrode/electrolyte interface is
a combination of optimization methods [44], [45] for calculated using the hybrid boundary element/finite element
optimization of location and output current of anode; method, as described in literature [7], [14], [53], [55] – [59].

- for the measurement results interpretation of the ON 1.2.5. Meshless methods

potential (the measured pipeline potential in relation to
the reference electrode mounted on the ground sur- In order to accelerate the calculation of the distribution of
face directly above the pipeline during the operation electromagnetic fields, more attention has been paid to
of the cathodic protection system), the OFF poten- meshless numerical methods. W. J. Santos et al. [60] –
tial (the measured electrical potential of the pipeline [62] applied the method of fundamental solutions (MFS) in
relative to the reference electrode mounted on the combination with the genetic algorithm (GA) to solve the
ground surface directly above the pipeline, cathodic problem of distribution of the current density and electric
protection system excluded) and the current density, potential on the electrode surfaces of the cathodic protec-
in order to determine the condition of the coating and tion system. As well as the boundary elements method,
the estimation of the existence and size of defects the method of fundamental solutions is applicable when
on isolation coating based on the boundary element the fundamental solutions of the field partial differential
method and the nonlinear regression analysis. Due to equations are known. Advantage of method of funda-
the robustness of the electrode surface model, they mental solutions over the boundary element method is
are discretized by cylindrical boundary elements, the it does not require any special integration or any special
anode polarization curve is assumed to be linear, treatment of singular points. By this method, the approx-
while the linearized polarization characteristic is used imation of the solution is obtained as a linear superposi-
for the cathodic surface [46] – [48]; tion of fundamental solutions. The singularity points are
positioned outside the domain of interest. These singular
- for the calculation of the distribution of the electric points are called virtual sources or a "pseudo-boundary",
potential of the cathodic protection system with which physically has no touch points with the actual do-
dynamic polarizing characteristics based on poten- main boundary points. The basic disadvantage of this
tiometric measurements caused by deposition of method is to determine the position of virtual sources as
carbonate deposits on the surface of the protected well as their current intensity. Their position must be deter-
object, approaches based on the method of bound- mined in such a way as to meet the boundary conditions
ary elements [49] – [52] were proposed. In order to at the observation points that are set at the domain of
obtain electrical potential values on the surface of the interest. Due to the nonlinearity of boundary conditions, it
protected object, two approaches were proposed: is necessary to apply iterative techniques in combination
the fictitious time method and the fictitious potential with the method of fundamental solutions.
method combined with the boundary element meth-
od.
2. BASIC EQUATIONS FOR CATHODIC PROTECTION
1.2.4. Coupled boundary element/finite element method SYSTEM MODELLING

Although the boundary element method is one of the The distribution of electric potential in electrolyte and on
most frequently used methods for the modelling of the electrode surface of cathodic protection system, under
cathodic protection system, it does not give satisfactory assumption that electrolyte is homogeneous and iso-
results in the case of very long electrode surfaces such as tropic, is defined by the Laplace partial differential (LPD)
long underground pipelines protected by a cathodic pro- equation for the static current field [8]:
tection system. The reason for this lies in the fact that this
method treats the electrode systems as equipotential, i.e.
− ∇(γ ⋅ ∇ϕ ) = 0 (1)
does not take into account the potential drop caused by
the resistance of the material from which the electrodes
where: γ electric conductivity of electrolyte in which ca-
are made. To overcome this problem, hybrid boundary el-
thodic protection system is placed, φ electric potential, a
ement/finite element method is proposed in literature [7],
symbol ∇ is Nabla operator.
[14]. The boundary element method is suitable for solving
stationary current field problem in infinite and semi-sub-
 Volume 11 January/December 2017 85

By solving the previous partial differential equations (1), significantly different form in comparison to the remaining
the value of the electric potential at any point within do- two functional relationships. The reason for this lies in the
main of interest can be determined. It is well known that fact that the electrochemical reaction of oxygen reduction
the LPD equation (1) have infinite solutions. Therefore, is concentration controlled and subject to diffusion laws,
to obtain a unique solution, it is necessary to solve the while the remaining two electrochemical reactions are ac-
LPD along with the appropriate boundary conditions. The tively controlled and their functional dependencies can be
boundary that needs to be take into account when mod- described by the Butler-Volmer equation.
elling cathodic protection systems are the anode/elec-
trolyte boundary and cathode/electrolyte boundary [63]. Different electrochemical reactions take place on different
The boundary conditions applied at these boundaries electrodes. For anodic surfaces, it is characteristic that
are called polarization characteristics and they take into an electrochemical reaction occurs simultaneously are
account the electrochemical reactions that take place on anode metal dissolution and oxygen reduction. The total
the electrode (anode and cathode) surfaces. Polarization value of the polarization current density of galvanic anode
characteristics represent functional relationships between surface is defined by following equation [65]:
the current density and the electrical potential of the elec-
trode surfaces [64]. Due to the nature of the electrochem-  ϕ −ϕ A 
j A = jO2 ⋅ 10 β A − 1
ical reactions, these functional relationships are nonlinear.  
The mathematical form of these functional relationships   (3)
depends primarily on the number and type of partial elec-
trochemical reactions occurring simultaneously on the where φA is the corrosive potential of the galvanic anode,
electrode surfaces. Partial electrochemical reactions oc- and βA is the Tafel's coefficient (slope) of the galvanic
curring at the same time on the cathode surface are metal anode.
dissolution, oxygen reduction and hydrogen separation.
Therefore, the total value of the polarization current den- Previous equation (3) is suitable for the galvanic anode
sity of the cathode current can be represented by the fol- cathodic protection system modelling. In case of the im-
lowing equation [31], [59]: pressed current cathodic protection system boundary
condition on the impressed current anode surface have
ϕ −ϕ F e (
− ϕ −ϕ H
2
) the following form [14]:
βF e βH2
j c = j 0, F e ⋅ 10 − j lim,O2 − j 0, H 2 ⋅ 10
(2)  ϕ − ∆U r −ϕ A 
j A = jO2 ⋅ 10 β A − 1
 
where jc is total current densities for cathode surface, jc,Fe   (4)
current density corresponding to the metal dissolution re-
action, jlim,O is the limiting current density of oxygen reduc- where ΔUr is the voltage added by the rectifier.
2
tion jH2 is current density corresponding to the reaction of
hydrogen separation, φ is potential difference of interface
metal/electrolyte φFe, and φH2 are equilibrium potentials of 3. DIRECT BOUNDARY ELEMENT METHOD
corresponding electrochemical reactions and βFe, and βH2
are Tafel's slopes (coefficient) of metal dissolution and hy- The direct boundary element method used for the calcu-
drogen separation, respectively. lation of the electrical potential and current density dis-
tribution is based on solving the integral equation of the
static current field. The integral equation of the static field
From the previous equation (2) it is noticeable that the var-
can be obtained by applying Green’s symmetric identity to
ious electrochemical reactions do not result in the same the Laplace partial differential equation (1). Integral equa-
form of functional relationship between the electric poten- tion of the static current field has the following form [59],
tial and the partial current density. It can be noticed that [66], [67]:
the functional relationship between partial current densi-
ty and electrical potential characteristic for reactions of
the metal dissolution and hydrogen separation have the  
c(q )ϕ (q ) + ∫ ϕ ( p ) ⋅ (n ⋅ ∇G ( p, q )) dΓ = ∫ G ( p, q ) ⋅ (n ⋅ ∇ϕ ( p )) dΓ
same form but the opposite sign. The reason is that the Γ Γ (5)
electrochemical reaction of metal dissolution is an anode
reaction and according to the electrochemical convention where q is observation point, p is field source point, c(q) is
there is a positive sign while the electrochemical reaction constant, φ(q) is the electric potential of the observation
of hydrogen separation is a cathode reaction and to the point q, φ(p) is the potential of the field source point p,
electrochemical convention has a negative sign. It is also G(p,q) is the Green’s function, Γ is the boundary of the
noticeable that the functional relationship that represents 
domain and n is normal unit vector.
the electrochemical reaction of oxygen reduction in has
86 B&H Electrical Engineering Bosanskohercegovačka elektrotehnika

In order to calculate the normal component of the current integral equations of each element has the same form.
density on the electrode surfaces, the Ohms law needs be After the application of the collocation point method, the
introduced, as follows [54] ,[59], [68]: coordinate transformation from global coordinate system
to the local coordinate system and the Gaussian - Legen-
 j( p) dre numerical integration on each boundary element, the
n ⋅ ∇ϕ ( p ) = −
γ (6) current density and the electric potential at the each point
of all the boundary elements can be obtained by solving
where j(p) is normal component of the current density of the following matrix equation [14], [63], [68], [71]:
field source point p.
 H cc H ca − 1 ϕ c  Gc c Gc a 
H   j 
− 1 ⋅ ϕ a  = Ga c Ga a  ⋅  c 
By incorporating the Ohm's law (6) into the integral equa-
 ac H aa
tion of the static current field (5), an integral field equation is j
obtained that is suitable for the calculation of current densi-  0 0 0  ϕ ∞   Ac Aa   a 
(10)
ty and electrical potential distribution on the electrode sur-
faces of the cathodic protection system, as follows: where [H] and [G] matrix coefficients, {φc, φa, φ∞}T vector
 1 of unknown electric potentials of electrode surfaces and
c(q )ϕ (q ) + ∫ ϕ ( p ) ⋅ (n ⋅ ∇G ( p, q )) dΓ = − ∫ G ( p, q ) ⋅ j ( p ) dΓ {jc, ja}T vector of current density. Indexes in the previous
γ
Γ Γ (7) equation c and a denote the cathode and anode surfaces,
respectively.
The previous integral equation of the static current field
(7) is suitable for solving bounded problems such as the
Since the elements of the current density vector are given by
calculation of the current density and electric potential
functional relationships between the current density and the
distribution of the inner wall of the pipeline or tanks. In or-
electrical potential, which are nonlinear, the matrix equation
der to calculate the parameters of the cathodic protection
(10) is nonlinear. Therefore, to solve nonlinear matrix systems
systems of metallic installations that are placed in infinite
it is necessary to use iterative techniques such as Gauss's
and semi-infinite domains, such as soil and sea water, the
method, Gauss - Seidel method or Newton - Raphson
Gaussian boundary condition for infinite boundaries must
method [21,72,73,74]. For the solution of the nonlinear ma-
be satisfied [16], [25], [52], [59], [63], [69], [70]:
trix equation (10), Newton - Raphson 's method was used
in this paper.
∫ j( p ) dΓ = 0 (8)
Γ
3.1. Iterative Newton - Raphson method
This boundary condition can simply be met by imposing a
constant potential at the infinite bound φ∞ which is added The Newton - Raphson's iterative method, proposed in
to the right side of the integral field equation (7). Therefore, the papers [49], [50], [51], [52], was used in this paper to
integral field equation for the modelling the cathodic pro- solve the nonlinear matrix equation (10). The first step of
tection systems placed in infinite and semi-infinite domain the Newton - Raphson's iterative method is the expansion
can be written in the following form [14], [25], [49], [50], of the current density vector into Taylor series. Then the
[53], [54], [59], [68]: current density vector in each iteration can be determined
by applying the following equation:
 1
c(q )ϕ (q ) + ϕ ( p ) ⋅ (n ⋅ ∇G ( p, q )) dΓ = −
∫ G ( p , q ) ⋅ j ( p ) dΓ + ϕ ∞
∫
γ (9)
Γ Γ
{ j}n = { j}n−1 +  ∂j  {∆ϕ}n
 ∂ϕ  n −1
(11)
By introducing a constant potential at the infinite bound-
where n denotes the current iteration, n - 1 previous it-
ary φ∞ in equation (7) provides a conservative flow of cur-
eration, [∂j/∂φ] is Jacobian's matrix and {Δφ} is vector of
rent between anode and cathode surface [16].
electric potential increments between the two adjacent
iterations. The vector of electric potential increments be-
In the procedure of calculating the current density and
tween two adjacent iterations can be calculated by using
electrical potential distribution on the electrode surfaces
the following equation [68,71]:
of the cathodic protection system using a direct boundary
element method, the boundary surfaces must be discret-
ized to a certain number of boundary elements. After the ϕ c  ϕ c 
   
discretization of the boundary surfaces, equations (8) and {∆ϕ}n = ϕ a  − ϕ a 
(9) are solved over each boundary element. Boundary sur- ϕ  ϕ 
 ∞  n  ∞  n −1
(12)
faces are usually discretized to the boundary elements of
the same geometric shape and thus the solution of the
 Volume 11 January/December 2017 87

By adding the equations (11) and (12) in to the matrix tivity of γ = 0.01 S/cm. The system consists of a cylindri-
equation (10), the following matrix equation is obtained cal protected object (cathode) of radius rc = 500 cm and
[49] – [52], [63]: a cylindrical galvanic anode radius ra = 25 cm. Distance
between centres of anode and cathode is d = 625 cm.
  Polarization characteristics parameters of anodic and ca-
 [H ] − [G ] ∂j   ⋅ {∆ϕ } = [G ]{ j} − [H ]{ϕ }
 n n −1 n −1 thodic surfaces are given in Table I.
  ∂ϕ S  n −1  (13)
Table I: Parameters of polarization characteristics [14], [31]

From the previous matrix equation (13), a vector of electric Parameter Value
potential increments can be determined between nth and j0Fe 1.0 (µA/cm2)
(n-1)th iterations. Further, by using equations (11) and (12), φFe -700 (mV) vs. CSE
a vector of the current density and the vector of electrical βFe 60 (mV/dec)
potentials of all the boundary elements in the nth iteration
jlimO 1.1 (µA/cm2)
can be determined. The iterative cycle is repeated until 2
jH 1.0 (µA/cm2)
convergence is achieved i.e. until all elements of electric 2

potential increments vector is less than the set value (suf- φH -800 (mV) vs. CSE
2
ficiently close or equal to zero) [49] – [52], [68], [71]. βH 120 (mV/dec)
2
jO 1.0(µA/cm2)
2
In order to start the iterative cycle, it is necessary to guess φa -1100 (mV) vs. CSE
the initial value of electric potential on all points of the βa 60 (mV/dec)
electrode surfaces. Since the performance of Newton -
Raphson's iterative methods depends on the initial guess,
Below are given results of the calculation of the electrical po-
it is necessary to consider the choice of initial point on the
tential and current density distribution on the electrode sur-
polarization characteristics. In this paper the initial value
faces of the analysed cathodic protection system. Due to the
of the electrical potential of all the points on the electrode
existence of symmetry on both analysed electrode surfaces,
surfaces was calculated from the functional relationships
it is sufficient to give the distribution of the electrical potential
defined by equations (2) and (3) under assumption that
and current density in the range 0° to 180°. Figures 2 and 3
there is no current flow in the system.
show the distribution of electrical potential and current densi-
ty on the surface of the protected object (cathodes).
Since matrix [H] and [G] are full matrix, calculation of the
equation (13) can be time consuming. Therefore, to accel-
erate the calculation of equation (13), the GMRES iterative
method was used [68], [71].

4. BENCHMARK EXAMPLE

A previously presented mathematical model was used for

the parameter calculation of one benchmark example of
galvanic anode cathode protection system. The geometry
of considered problem is given in Figure 1.
Figure 2: Distribution of electrical potential on the cathode surface
90°

90°

0° d 0°
180° ra 180°
Anode

270° rc
Electrolyte

270° Cathode

Figure 1: Geometry of analysed benchmark problem

The analysed cathodic protection system is placed in a

homogeneous isotropic electrolyte with a electric conduc- Figure 3: Distribution of current density on the cathode surface
88 B&H Electrical Engineering Bosanskohercegovačka elektrotehnika

The results shown in Figures 2 and 3 indicates that the 5. CONCLUSION

highest negative value of the electrical potential on the
cathode surface that is closest to the anode (the most When designing and verifying the efficiency of modern
negative value of the electric potential on the cathode cathodic protection systems it is necessary to carry out
surface is for α = 180°). Based on the above, it can be very precise and accurate methods for the calculation of
concluded that the cathode surface that is closest to the cathodic protection system parameters. The paper pres-
anodes will be more protected from the other parts of the ents an overview of used mathematical models in litera-
cathode. Also, the cathode surface closest to the galvanic ture for the cathodic protection system modelling. Also, a
anode is the highest value of the protective current den- mathematical model based on the direct boundary element
sity. method for the calculation of parameters of the cathodic
protection system is presented in this paper. The presented
The distribution of electrical potential on the galvanic an- mathematical model takes into account the electrochem-
ode surface is shown in Figure 4. From the results given ical reactions that take place on the electrode surfaces
in Figure 4, it can be noted that the highest value of the through the corresponding polarizing characteristics. An
potentials is on the anode surface that is closest to the example of the application of the presented mathematical
cathode (the highest value of the anode electrical poten- model is demonstrated in a geometrically simple example
tial is for α = 0°). of the cathode protection system with galvanic anodes.

The results given in Figure 5 indicate that the value of the

current density of the anode surface is highest on the sur- REFERENCES
face closest to the cathode. Since the distribution of the
protection current density is nonlinear on the surface of [1] A. Bahadori: Cathodic Corrosion Protection Systems:
the galvanic anode, it will come to uneven dissolution of A Guide for Oil and Gas Industries, Gulf Professional
the galvanic anode over time. The previously presented Publishing, USA, Massachusetts, Waltham, 2014
results indicate that the galvanic anode surface that is [2] A. Muharemović: Contribution to the analysis of the
impact of polarization effects on the optimization and
closer to the cathode will faster dissolve.
dimensioning of the cathodic protection system (in
Bosnian), Ph.D. Thesis, Sarajevo 1988
[3] V. G. DeGiorgi: Corrosion basics and Computer Mod-
eling, Industrial applications of the boundary element
method, Computational Mechanics, pp. 47 – 79,
September 1993
[4] G. H. Koch, M. P.H. Brongers, N. G. Thompson, Y. P.
Virmani & J.H. Payer: Corrosion Costs and Preven-
tive Strategies in the United States, PUBLICATION
No. FHWA-RD-01-156, NACE International, 2002
[5] Cost of Corrosion to exceed $1 Trillion in the United
States in 2012-G2MT Labs - The Future of Materials
Conditions Assessment
[6] C. I. Ossai: Advances in Asset Management Tech-
Figure 4: Distribution of electrical potential on the anode surface niques: An Overview of Corrosion Mechanisms and
Mitigation Strategies for Oil and Gas Pipelines, ISRN
Corrosion, Article ID 570143, pp. 1 – 10, November
2012
[7] I. Turković: Contribution to the study of the influence of
nonlinear distribution of voltage and current on the opti-
mization of protective lengths in the domain of cathodic
protection (in Bosnian), Ph.D. Thesis, Sarajevo 2010
[8] L. Lazzari, P. Pedeferri: Cathodic Protection, Polip-
ress Milano, 1st edition, 2006
[9] J. H. Morgan: Cathodic Protection, 2nd Edition,
NACE, Houston, TX, 1993
[10] V. Cicek: Cathodic Protection - Industrial solutions for
protection against corrosion, Wiley, 2013
Figure 5: Distribution of the current density on the [11] A. W. Peabody: Control of pipeline corrosion, 2nd
anode surface Edition, NACE, Houston, TX, 2001
[12] W. Baeckmann, W. Schwenk, W. Prinz: Handbook of
cathodic corrosion protection, Gulf Publishing Com-
pani, Houston, TX, 1997
 Volume 11 January/December 2017 89

[13] B. Jarić, A. Rešetić: Corrosion: Elektrokemijske os- [28] P. Marcassoli, A. Bonetti, L. Lazzari, M. Ormellese:
nove and cathodic protection (in Croatian), Korex- Modeling of potential distribution of subsea pipeline
pert, Zagreb, 2003 under cathodic protection by finite element method,
[14] D. P. Riemer: Modeling Cathodic Protection for Pipe- Materials and Corrosion, Vol. 66, No. 7, pp. 619 –
line Networks, Ph.D. dissertation, University of Flori- 626, July 2015
da, Gainesville, FL, 2000 [29] R. Montoya, O. Rendon, J. Genesca: Mathematical
[15] A. Muharemović, H. Zildžo, I. Turković, H. Matoruga: simulation of a cathodic protection system by finite
Modeling of the Protection Current/Potential Distri- element method, Materials and Corrosion Vol. 56,
bution in the Cathodic Protection Systems with Gal- No. 6, 2005
vanic Anode by using Coupled BEM/FEM Method, [30] A. Mujezinović, A. Muharemović, I. Turković, A. Mu-
Journal of Energy, Zagreb Vol. 58, No. 4, 2009 haremović: Calculation of the Protective Current
[16] R. A. Adey, P. Y. Hang: Computer simulation as an aid Density Distribution of a Cathodic Protection System
to corrosion control and reduction, NACE Corrosion with Galvanic Anodes in Terms of Double-layer Elec-
Conference, USA, Texas, San Antonio, April 1999 trolyte, 34th International conference on Boundary
[17] K. Amaya: Mathematical modeling for corrosion anal- Elements Method and Other Reduction Methods,
ysis, Modelling of Cathodic Protection Systems, Vol. BEM/MRM, Croatia, Split, 2012
12, No. 1, pp. 1-12, November 2005 [31] S. Martinez: Evaluation of the Uniform Current Den-
[18] A. Muharemović, I. Turković, A. Muharemović, S. sity Assumption in Cathodic Protection System with
Tašaković, A. Mujezinović: Calculation methods of Close Anode to Cathode Arrangement, Materials and
cathodic protection system parameters with vertical Corrosion, Vol. 61, No. 4, pp. 338-342, April 2010
anode zinc strings, 20TH International expert meet- [32] R. A. Adey, S. M. Niku, C. A. Brebbia, J. Finnegan,
ing, Power Engineering, Slovenia, Maribor, May 2011 “Computer Aided Design of Cathodic Protection
[19] A. B. Peratta, J. M. W. Baynham, R. A. Adey: Com- System”, Applied Ocean Research, Vol. 8, No. 4, pp.
putational modeling of cathodic protection systems 209 – 222, October 1986
for pipelines in multi-layer soil, 3rd International Con- [33] N.G. Zamani, J. M. Chuang, J. F. Porter: BEM Sim-
ference on the Simulation of Electrochemical Pro- ulation of Cathodic Protection Systems Employed in
cesses (ELECTROCORR), Italy, Bologna, June 2009 Infinite Electrolytes, International Journal for Numerical
[20] J. F. Yan, S. N. R. Pakalapati, T. V. Nguyen, R. E. Methods in Engineering, Vol. 24, pp. 605 – 620, 1987
White, R. B. Griffin: Mathematical Modeling of Ca- [34] E. Chisholm, L. J. Gray, G. E. Giles: Solution of Non-
thodic Protection Using the Boundary Element Meth- linear Polarization Boundary Conditions, Electronic
od with a Nonlinear Polarization Curve, Journal of Journal of Boundary Elements, Vol. 1, No. 3, pp. 418
The Electrochemical Society, Vol. 139, No.7, pp. – 438, October 2007
1932-1936, July 1992 [35] A. B. Peratta, J. M. W. Baynham, R. A. Adey, J. Vit-
[21]R. A. Corbett: Cathodic protection as an equivalent tonato: Advances in Cathodic Protection Modeling of
electrical circuit, IEEE Transactions on Industry Appli- Deep Well Casing in Multi – Layered Media, NACE In-
cations, Vol. IA-21, No. 6, pp. 1533–1537, Novem- ternational Conference & Expo, CORROSION, USA,
ber/December 1985 Georgia, Atlanta, January 2009
[22] L. I. Freiman: On the Insulation Resistance of an Un- [36] W. Sun: Optimal control of impressed cathodic pro-
derground Pipeline Polarization Component and To- tection systems in ship building, Applied mathemati-
tal Resistance, Protection of Metals, Vol. 37., No. 3, cal modelling, Vol. 20, No. 11, pp. 823-828, Novem-
pp. 268 – 279., May 2001 ber 1996
[23] S. Martinez, I. Štern: A mathematical model for the [37] A. B. Perrata, T. Froome, R. A. Adey: Jacket SACP
internal cathodic protection of cylindrical structures system design and optimization using simulation,
by wire anodes, Journal of Applied Electrochemistry, NACE International Conference & Expo Corrosion,
Vol. 30, No. 9, pp. 1053-1060, January 2000 USA,Utah, Salt Lake City, March 2012
[24] J. P. Sale: The evolution of anode configuration for the [38] L. Q. Liu, H. T. Wang: A line model - based fast bound-
internal cathodic protection of the pipes, Conference ary element method for the cathodic protection analysis
on Corrosion and Protection of Pipes and Pipelines of pipelines in layered soils, Computer Modeling in En-
[25] M. Masud, M. Arita, L. E. Ju, K. Hanada, H. Minaga- gineering and Sciences (CMES), Vol. 90, No. 6, pp. 439
wa, K. Kawamata: The Application of FEM to Ca- - 462, February 2013
thodic Corrosion Protection of Steel Reinforcement [39] A. B. Peratta, J. M. W. Baynham, R. A. Adey, J. Vit-
in Concrete, Materials transactions, Vol. 45, No.12, tonato: Advances in Cathodic Protection Modeling of
pp. 3349-3355, Juny 2004 Deep Well Casing in Multi – Layered Media, NACE In-
[26] E. Kurgan, A. Wantuch: Impressed cathodic protec- ternational Conference & Expo, CORROSION, USA,
tion of underground structures, Przegląd Elektro- Georgia, Atlanta, January 2009
techniczny, Vol. 87, No. 5, pp. 96-99, January 2011 [40] A. B. Peratta, J. M.W. Baynham, R. A. Adey: Numer-
[27] L. Y. Xu, Y. F. Cheng: Experimental and numerical ical Modelling of Cathodic Protection Systems for
studies of effectiveness of cathodic protection at cor- Deep Well Casings, 3rd International Conference on
rosion defects on pipelines, Corrosion Science, Vol. the Simulation of Electrochemical Processes (ELEC-
78, pp. 162-171, January 2014 TROCORR), Italy, Bologna, June 2009
90 B&H Electrical Engineering Bosanskohercegovačka elektrotehnika

[41] A. Mujezinović, A. Muharemović, I. Turković, A. Mu- [55] A. Muharemović, H. Zildžo, I. Turković, H. Matoruga:
haremović: Calculation of the protective current den- Modeling of Protection Current/Potential Distribution in
sity distribution of a cathodic protection system with Cathodic Protection Systems with a Galvanic Anode by
galvanic anodes in terms of double-layer electrolyte, Using the Coupled BEM/FEM Method, Journal of Ener-
In Corrosion: Material Performance and Cathodic gy, Vol. 58, No. 6, pp. 496 – 517, October 2009
Protection, 99, 231, 2017 [56] A. Muharemović, H. Zildžo, N. Behlilović, I. Turković:
[42] H. Y. Hang, R. A. Adey: Design optimisation of ship Numerical model for calculation of parameters of ca-
ICCP systems, 21st International Series on Boundary thodic protection system with galvanic anodes, XXII
Element, pp. 195-206, 1999 International Symposium on Information, Commu-
[43] P. Miltiadou, L. C. Wrobel: Optimization of Cathodic nication and Automation Technologies (ICAT), IEEE,
Protection Systems Using Boundary Elements and Bosnia & Herzegovina, Sarajevo, October 2009
Genetic Algorithms, Corrosion, Vol. 58, No. 11, pp. [57] A. Muharemović, H. Zildžo, E. Letić: Modelling of pro-
912-921, November 2002. tective potential distribution in a cathodic protection
[44] S. Aoki, K. Amaya: Optimization of Cathodic Pro- system using coupled BEM/FEM method, 30th Inter-
tection System by BEM, Engineering Analysis with national conference on Boundary Elements Method
Boundary Elements, Vol. 19, No. 2, pp. 147–156, and Other Reduction Methods, BEM/MRM, Slove-
March 1997 nia, Maribor, July 2008
[45] K. Amaya, S. Aoki: Optimum Design of Cathodic Protec- [58] F. Brichau, J. Deconinck: A Numerical Model for Ca-
tion Using the Boundary Element Method, Simulation of thodic Protection of Buried Pipes, CORROSION, Vol.
Electrochemical Processes, Spain, Cadiz, May 2005 50, No. 1, pp. 39 – 49, January 1994
[46] C. Qiu: Model for interpretation of pipeline survey [59] A. Mujezinović, S. Martinez, A. Muharemović, I.
data, Ph.D. Theses, University of Florida, USA, Flori- Turković: Application of the Coupled BEM/FEM
da, 2003 Method for Calculation of Cathodic Protection Sys-
[47] C. Qiu, M. E. Orazem: A weighted nonlinear regres- tem Parameters, International Journal of Computa-
sion-based inverse model for interpretation of pipe- tional Methods and Experimental Measurements,
line survey data, Electrochimica Acta, Vol. 49, No. Vol. 5, No. 5, pp. 659 – 666, 2017
22, pp. 3965–3975, March 2004 [60] W. J. Santos, J. A. F. Santiago, J.C.F. Telles: Optimal
[48] A. B. Peratta, J. M.W. Baynham, R. A. Adey: A Positioning of Anodes and Virtual Sources in the De-
computational approach for assessing coating per- sign of Cathodic Protection Systems using the Meth-
formance in cathodically protected transmission od of Fundamental Solutions, Engineering Analysis
pipelines, NACE Internationa Conference & Expo, with Boundary Elements (EABE), Vol. 46, pp. 67 – 74,
CORROSION, USA, Georgia, Atlanta, January 2009 September 2014
[49] J. A. F. Santiago, J. C. F. Telles: On Boundary Ele- [61] W. J. Santos, J. A. F. Santiago, J.C.F. Telles: An Ap-
ments for Simulation of Cathodic Protection System plication of Genetic Algorithms and the Method of
with Dynamic Polarization Curves, International Jour- Fundamental Solutions to Simulate Cathodic Pro-
nal for Numerical Methods in Engineering, Vol. 40, tection Systems, Computer Modeling in Engineering
No. 14, pp. 2611-2627, July 1997 and Sciences (CMES), Vol. 87, No. 1, pp. 23 – 40,
[50] J. A. F. Santiago, J. C. F. Telles: A solution technique December 2012
for cathodic protection system with dynamic bound- [62] W. J. Santos, J. A. F. Santiago, J.C.F. Telles: Using
ary conditions by the boundary element method, Ad- the Gaussian Function to Simulate Constant Po-
vances in Engineering Software, Vol. 30, No. 9, pp. tential Anodes in Multiobjective Optimization of Ca-
663-671, September 1999 thodic Protection Systems, Engineering Analysis with
[51] J. A. F. Santiago, W. J. Mansur, J. C. F. Telles: Time Boundary Elements (EABE), Vol. 73, pp. 35 – 41,
Dependent Effects in Cathodic Protection Systems September 2016
Using Boundary Elements, Proc. 1st Int. Offshore and [63] A. Mujezinović, I. Turković, A. Muharemović, S. Martinez,
Polar Engineering Conf., UK, Edinburgh, August 1991 S. Milojković: Modeling of the Galvanic Anode Cathodic
[52] J. A. F. Santiago: Static and Dynamic Simulation of Protection System with Dynamic Polarization Character-
Cathodic Protection Systems with Boundary Ele- istics, International Journal of Chemistry and Chemical
ment Method (in Portuguese), Ph.D. Thesis, COPPE/ Engineering Systems, Vol. 1, pp. 95-100, 2016
UFRJ, Rio de Janeiro, RJ, 1991 [64] A. Muharemović, N. Behlilović, I. Turković, A. Muhare-
[53] C. Liu, A. Shankar, M. E. Orazem, P. D. Riemer: Nu- mović: Functional relationship between cathodic pro-
merical simulations for cathodic protection of pipe- tection current/potential and duration of system de-
lines, in: Underground pipeline corrosion - Detection, ployment in desert conditions, Advances in Engineering
analysis and prevention, Edited by M. E. Orazem, Software, Vol. 42, No. 6, pp. 359-367, June 2011
Woodhead Publishing, 2014 [65] D. P. Riemer, M. E. Orazem: A Mathematical Model
[54] D. P. Riemer, M. E. Orazem: Modeling coating flaws for the Cathodic Protection of Tank Bottoms, Corro-
with non-linear polarization curves for long pipelines, sion Science. Vol. 47, No. 3, pp. 849-868, 2005
Corrosion and Boundary Element Methods, Advanc- [66] C. A. Brebbia, J. C. F. Telles, L. C. Wrobel: Boundary
es in Boundary Elements, Vol. 12, No. 1, pp. 225- Element Techniques Theory and Applications in En-
259, November 2005 gineering, Springer – Verlag, 1984
 Volume 11 January/December 2017 91

[67] C. A. Brebbia, J. Dominguez: Boundary Elements: An BIOGRAPHY

Introductory Course, 2nd Edition, WIT press, 1994
[68] A. Mujezinović, S. Martinez, S. Milojković: On the use Adnan Mujezinović received his B.Sc., M.Sc. and Ph.D
of Boundary Element Method for Cathodic Protec- degrees in electrical engineering from University of Sara-
tion System Modeling, In Advanced Technologies, jevo (Bosnia and Herzegovina), in 2009, 2011 and 2017,
Systems, and Applications II, Springer International
respectively. Currently, he is an assistant Professor at the
Publishing, in press 2018
Faculty of Electrical Engineering, University of Sarajevo.
[69] J. C. F. Telles, W. J. Mansur, L. C. Wroble: On Bound-
ary Elements for External Potential Problems, Me- His topics of interest are numerical calculations of elec-
chanics Research Communications, Vol. 11, No. 6, tromagnetic fields, cathodic protection and grounding
pp. 373 – 377, November – December 1984 systems.
[70] J. C. F. Telles: A Self-Adaptive Co-ordinate Transfor-
mation for Efficient Numerical Evaluation of General Irfan Turković received his M.Sc. and Ph.D. degrees
Boundary Element Integrals, International Journal for in electrical engineering from the University of Sarajevo,
Numerical Methods in Engineering, Vol. 24, pp. 959 – Bosnia and Herzegovina, in 2003 and 2010, respectively.
973, 1987 Currently, he is an associate Professor at the University of
[71] A. Mujezinović: Modelling of the Galvanic Anode Ca- Sarajevo – Faculty of Electrical Engineering. His topics of
thodic Protection System with Dynamic Nonlinear interest include cathodic protection, electrical measure-
Polarization Characteristics (in Bosnian), Ph.D. The-
ments, safety in low voltage network and EMC in power
sis, Sarajevo 2017
systems.
[72] J. M. Chuang, N. G. Zamani, C. C. Hsiung: Some
Computational Aspects of BEM Simulation of Ca-
thodic Protection Systems, Applied Mathematical
Modelling, Vol. 11, No. 5, pp. 371 – 379, October
1987
[73] W. Sun, G. Yuan, Y. Ren: Iterative algorithms for im-
pressed cathodic protection systems, International
Journal for Numerical Methods in Engineering, Vol.
49, pp. 751 – 768, 2000
[74] W. Sun: Relaxation iterative algorithms for solving
cathodic protection systems with non-linear polariza-
tion curves, International Journal for Numerical Meth-
ods in Engineering, Vol. 55, pp. 401 – 412, 2002

View publication stats