Electrical Power Systems

High Voltage Engineering and Insulation

Paper ID 1071
Ampacity Calculation of the Underground Power
Cables in Voluntary Conditions
by Finite Element Method
s. Kahourzade
1
, A. Mahmoudi
1
,
2
, B. Nim Taj
1
and o. Palizban
1
[email protected], [email protected], [email protected], omid [email protected]
Electrical Engineering Department, University of Malaya, 50603 Kuala Lumpur, Malaysia.
Engineering Department, Help College of Arts and Technology, Klang, 41050 Selangor, Malaysia.
Abstract- Exceeding the ampacity of the underground power
cables according to the depth of soil and other parameters is a
problem in literature because of the risk of temperature rise and
subsequently the probable material failure. This paper
introduces the heat transfer mechanisms in the underground
cable installations and analyzes the available solution methods of
the diffusion equation by fnite element method. The cable
ampacity is a function of soil thermal resistively, and burial
depth. Cable ampacity relationship with these parameters is
investigated and the manner that they affect the current of the
cable is surveyed. To do so, the commercial ANSYS software has
been employed. Results show that ampacity decreases as the
depth of soil or its conduction thermal resistance increases. This
method gives us the criteria and a fexible approach to change the
strategy on those cases which have not been pointed in cable
routing standard documentary.
NOMENCLATURE
f
System fequency [Hz]
U Cable operating voltage (phase-to-phase) [V]
( Conductor temperature [e]
Tamb Ambient temperature [ e]
H Burial depth below the earth surface [m]
1 Curent [Amp]
T Operating temperature [ e]
p
Thermal resistivity [OKmlw]
S Cross-sectional area of conductor [mm
2
]
We
Losses in conductor per unit length [wlm]
WI Total R2 power loss of each cable [wlm]
Wd
Dielectric losses per unit length per phase [wlm]
WI Total power dissipated in the cable per unit [wlm]
Relative permittivity of insulation
tan i Loss factor of insulation
Ys
Skin effect factor
Yp
Proximity effect factor
D; Diameter over insulation [m]
de
Exteral diameter of conductor [m]
,1
Ratio of the total losses in metallic sheath
,
2
Ratio of the total losses i metallic armor
020
temperature coeffcient at 20 degrees of
Celsius
P20
resistance at 20 degrees of Celsius
The 8th Electrical Engineering/ Electronics, Computer,
Telecommunications and Information Technology (ECTI)
Association of Thailand - Conference 2011
R' dc conductor resistance [Om]
R ac conductor resistance [Om]
I. INTRODUCTION
Calculation of the curent-carrying capability or ampacity of
power cables has been extensively discussed in the literature
and is the subject of several interational and national
standards. Ampacity calculation techniques are as old as the
cables themselves. Anders has summarized the history of
ampacity calculations i his 1997 book [1]. There are
analytical and numerical approaches to calculate cable
ampacity. The two major interational standard associations,
the IEEE and the IEC, adopted the analytical methods as the
basis for their standards [2], [3], [4]. The calculation
procedures in both standards are the same in principle and are
based on the model proposed by Neher and McGrath [5]. The
major difference between them is the use of units and because
of this, the same equations look completely different. The
nwnerical approaches are mainly based on [mite-difference or
[mite-element techniques. The [mite-element technique is
better suited for cable ampacity determination owing to
physical condition of this problem.
Ampacity in an underground cable system is determined by
the capacity of the installation to extract heat fom the cable
and dissipate it in the surounding soil and atmosphere [6].
The maximum operating temperature of a cable is a fnction
of the damage that the insulation can suffer as a consequence
of high operating temperatures. The insulation withstands
different temperatures as fnction of the duration of the
current circulating in the conductors. There are three
standardized ampacity ratings: steady state, transient (or
emergency) and short-circuit. Only steady state ampacity
ratings are discussed in this paper. As providing analytical
solution for complex situation is very indirect, in many cases
the experimental method may preferred. In these condition
nwnerical methods such as fnite element method (FEM) is
more pragmatic and cost effective. This paper focuses on the
numerical techniques for the computation of cable ampacity in
steady-state through the use of assumptions that simplif the
problem.
FEM is chosen to calculate the heat transfer feld and heat
Page 657
generation with in the cable and heat dissipation in the
surrounding soil. ANSYS sofware is used to generate and
solve the problem [7]. A two dimensional model of the cable
in fll size and surrounding area are designed in ANSYS.
Results are then compared to values presented in standard
tables. Also the skin effect in the isolation is considered as a
reduction in the main current. Both three-phase and single
phase cable systems installed in standard condition is
simulated and the temperature contour of them is presented.
To investigate the effects of cable depth and soil theral
resistivity; these parameters are changed step by step and the
ampacity of the cables in each case is calculated.
II. MODELING
The cable laying example shown in Fig. 1 is used to
develop the mathematical model. It is a 400kV, paper
polypropylene-paper cable with 2000mm
2
copper segmental
conductor and aluminum corrugated sheath. The outer
covering is a PE jacket. The cable cross section is shown in
Fig 2. The cables are laid in a fat formation without
transposition, directly in the soil, with theral resistivity of 1
K.mlw and ambient temperature equal to 25 C. The sheaths
are cross bounded with unknown minor section length. The
centers of the cables are 1.8m below the ground and phases are
O.5m apart. This laying condition is called the standard
installation. The cable parameters are provided in table I.
r T . mb
=25
Fig. 1. Standard installation of the cable system
To consider the effect of infnite area around the cable and
heat dissipation in this area to be reasonable minimum 100
meter of soil in every direction is considered, and isotheral
boundary condition is then loaded as far beyond condition. As
the soil conductivity is a great amount compare to the
convection resistance of the air on earth surface, convection
resistance of the air is neglected and the soil surface is
considered as an isothermal surface. ANSYS analysis relies on
its element and choosing the element has an important effect
on the results. PLANE35 (a 6-node triangular element) is
selected for simulations. The triangular shape element is well
suited to model irregular meshes. The element has one degree
of feedom, temperature at each node. The 6-node theral
element is applicable to a 2-D, steady-state or transient
thermal analysis.
The 8th Electrical Engineering/ Electronics, Computer,
Telecommunications and Information Technology (ECTI)
Association of Thailand - Conference 2011
f
f
Dj
Electrical Power Systems
High Voltage Engineering and Insulation
C~m.cop tf
tDM !
vndu<t|d
1,ovm
Hwonpqr
8,o
0crrugtedsmw|n
,o1+w
mw~rensm~
P.0~
Fig. 2. Cable cross section
TABLE I
SOME PARAMETERS OF THE SIMULATED CABLE
General Data
85c 2.8
60 (Hz) tanS 0.001
de 0.058
Calculated data for 3-phase cable system in standard installation
y,
0.132
y
0.005
AI 0.150 A2 0
R 0.0126 (Qlkm) We 23.40 (wlm)
WJ 26.92(wlm) Wd 6.53 (!Im)
I 1365 (Amp) Wr 33.45 (!Im)
Boundary conditions are isothennal (25 degrees of
Centigrade) and heat generation in the cable conductor enters
as a load. An iterative method is used to calculate ampacity of
the cable. Also, the skin effect in the isolation is considered as
reduction in the main current. Maximum temperature of power
cables is deterined by the cable insulation. This temperature
is produced by the energy which is generated through Ohmic
losses. dc conductor resistance R' (Qlm) is frst obtained from
the following equation:
R
'
=
1.02 ^ 10
6
P20
[1 +
a
20(f - 20)
1
S
(1)
R
'
is then modifed to take into account skin and proximity
effects. The resistance of a conductor when carrying
alterating curent is higher than that when carrying direct
current. The principal reasons for the increase are skin effect,
proximity effect, hysteresis and eddy curent losses in nearby
ferromagnetic materials, and induced losses in short-circuited
non-ferromagnetic materials. For conductor losses, only skin
and proximity effects, and in some cases approximation of
metallic sheath effect, are considered, except in very high
voltage cables consisting of large segmental conductors. The
relevant expressions are:
R=R
'
(l+ys+Y
p
)
(2)
We = R J
2
(3)
where, R(Qlm) and Wlwlm) are the conductor's ac
resistance and losses, respectively. Sheath losses are curent
and bonding dependent, and can be divided into two
categories. The losses are due to current circulating in single-
Page 658
core-cable and divided in to eddy curent circulating radially
(skin effect) and azimuthally (proximity effect). Regardless of
bonding, eddy current losses occur in both three-core and
single-core cables. Eddy curent losses in solidly bonded
single-core-cable sheaths are negligible. So, they are ignored
except for cables with large segmental conductors. Cable
losses Wj can thus be expressed as:
w, =wc+ws+wa=Wc(I+AJ+A
2
)
(4)
Al is called sheath loss factor, and is equal to the ratio of
total losses in the metallic sheath to total losses. A2 is armor
loss factor, and is equal to the ratio of total losses in the
metallic armor to total losses of conductor. When paper and
solid dielectric insulations are subjected to alterating voltage,
they act as large capacitors. Each time voltage direction
changes, realignment of electrons produce heat. It results in
loss of real power called dielectric loss. For a unit length of
cable, magnitude of the required charging curent is a fnction
of dielectric insulation constant, dimensions of the cable, and
operating voltage. For some cable arrangements (notably for
high-voltage, paper insulated) this loss can signifcantly affect
cable rating. Dielectric losses are computed fom the
following expression:
Wd=27jC.UJtao (5)
Where, the electrical capacitance C and the phase to ground
voltage Va are obtained fom:
c =

.10-
9
181n( J
U
Uo
=
.
(6)
(7)
It is convenient to express all heat fows due to power
losses in the cable in term of the loss per meter of the
conductor. From Eqs. (4) and (5) the total loss WI is computed
from:
w, =w, +Wd =Wc(l+AJ +A2
)+Wd
(8)
Finally, heat generation applied to the conductor surface in
the simulations is obtained fom [8]:
H.G=
S
III. SIMULATION
(9)
A three-phase cable in standard installation is simulated by
the ANSYS sofware and then a single-phase cable that is
installed in its standard condition is simulated. Simulation
method in both three-phases and single-phase is the same; the
only difference is number of the cables. The temperature
contour is presented for both three-phase and single-phase
cable installation in Figs. 3 and 4 respectively.
In three-phase cable achieving 85 restricts the current by
1363 Amp. In single-phase cable installation, the accepted
curent is limited by 1739 Amp. Current of three-phase and
single-phase cable systems in standard installation are 1365
Amp and 1735 Amp respectively. The standard values indicate
The 8th Electrical Engineering/ Electronics, Computer,
Telecommunications and Information Technology (ECTI)
Association of Thailand - Conference 2011
Electrical Power Systems
High Voltage Engineering and Insulation
less than 1 percent difference compare to the results of
simulation. Table II compares the simulated results with
standad values. The results are very close to the ones
presented in the standard that shows the validity of present
simulation approach. Also, simulation results show that there
is a signifcant cable ampacity reduction when the cable is
used in three-phase installation (around 28 percent).
Fig. 3. Temperatures counter of the three-phase cable installation
Fig. 4. Temperatures counter of the single-phase cable installation
TABLE II
SIMULATION AND STANDARD VALUES
Deviation }om
ANSYS Simulation Standard
standard
Three-phase 1363 1365 0.00146734
Single-phase 1739 1735 -0.00230017
IV. ANALYSES OF RESUL TS AND PARAMETER EFFECT
Effect of burial depth is shown in Figs. 5 and 6 for three
phase and single-phase cable installation, respectively. they
show that with any increase in depth of the soil, its
conductivity is reduced, less heat dissipation occurs and
ampacity drops. The rate of cable ampacity variation is much
more itense when the cable is buried near the earth surface
than it is buried in depth.
Soil conductivity varies in soils with different properties
Page 659
and its variation depends on moisture and substance content of
the soil. Figs. 7 and 8 show the amount of cable ampacity
against the soil resistivity variation for both three-phase and
single-phase cable installation respectively. Cable ampacity is
proportionate to soil conductivity; as soil conductivity raise,
cable ampacity increases because more heat dissipation
occurs.
Z>
Z0
>
0
0 Z
Buried Depth Iml
Fig. 5. Effect of buried depth variation of cable ampacity
(Three-phases cable installation)
2M
W
c

R
'5
.
C
L

w !0

.
W
0 _ _
Buried Depth 1m)
Fig. 6. Effect of buried depth variation on cable ampacity
(single-phase cable installation)
Z
!&
!6
'4

!Z
" !0
b
N
.

6
.
4
Z
0
0 0 !.b Zb J.b
Therm.1 Resistivity I K.m/w)

4.
Fig. 7. Efect of soil thermal resistivity variation on cable ampacity
The 8th Electrical Engineering/ Electronics, Computer,
Telecommunications and Information Technology (ECTI)
Association of Thailand - Conference 2011
2N
Z
I
1N
.
C
10

.
. N
Electrical Power Systems
High Voltage Engineering and Insulation
0 __
0 .6 I6 2.5 3b
Thermal Resistiity ( K.m/w)
4.5
Fig. 8. Effect of soil thermal resistivity variation on cable ampacity
(Single-phase cable)
V. CONCLUSION
To determine more closely the relevant temperature rises
including the effects of soil conductivity and buried depth of
cable is possible by the use of fnite element method. The
proposed model was used to determine fairly well the
maximum allowable load currents for the cable under
consideration. Such a model and analysis can be applied for
assessing the ampacity of cables in various installation
conditions. Results showed the amount of cable ampacity
depends on the cable geometry, cable construction, burial
depth and thermal conditions of the soil. Cable ampacity
decreased as the depth of soil or its conduction thermal
resistance increased. Also, cable ampacity reduces
considerably when the cable is used in three-phase installation
compared to single-phase installation. This method gives us
the criteria and a fexible approach to change the strategy on
those cases which have not been pointed in cable routing
standad documentary .
VI. REFERENCES
GJ. Anders(1997), "Rating of Electric Power Cables-Ampacity Calculations
for Transmission, Distribution and Industrial application," IEEE Press,
New York, McGraw-H ill( 1998).
IEEE Standard Power Cable Ampacity Tables, 1994. IEEE Std. 835-
1994, NY.
IEC Standard 60287 (1969,1982,1994), "Calculation of the continues current
rating of cables (100% load factor)" I" edition 1969, 2nd edition 1982,
3rd edition 1994-1995.
IEC Standard 60287, part 2-1 (1994), "Calculation of thermal resistances. "
J. H. Neher and M. H. McGrath, "The calculation of the temperature riseand
load capability of cable systems," AlEE Trans. Power App. Syst, vo1.76,
pp. 752-772, Oct. 1957.
GJ. Anders, "Rating of Electric Power Cables in Unfavorable Thermal
Environments " IEEE Press, New York, (2005).
ANSYS Finite Element Simulation Software, ANSYS Inc., Canonsburg, PA.
A. Mahmoudi, "Ampacity Derating Factors for Cables in Diferent
Environmental Conditions by FEM," M.S. dissertation. Dept. Elect.
Eng., Amirkabir University of Technology, 2008 .
Page 660