0% found this document useful (0 votes)
14 views

2015-Refinement of Underground Power Cable Ampacity by Multiphysics FEA Simulation

Uploaded by

Rajat Sharma
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
14 views

2015-Refinement of Underground Power Cable Ampacity by Multiphysics FEA Simulation

Uploaded by

Rajat Sharma
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 8

INTERNATIONAL JOURNAL OF ENERGY Volume 9, 2015

Refinement of Underground Power Cable


Ampacity by Multiphysics FEA Simulation
Simon Dubitsky1, George Greshnyakov2, and Nikolay Korovkin3

1)
Tor Ltd., St. Petersburg, Russia
2)
Research Institute "Sevkabel", St. Petersburg, Russia,
3)
St. Petersburg State Polytechnical University, Russia

assess the AC resistive losses in conductive cable elements:


Abstract— The maximum current in underground cable line is conductor, screen and armor.
limited by the rated temperature of conductor and insulation. The The classical method of ampacity calculation is given by the
standard method for ampacity calculation in accordance the IEC 60287 standard. Its theoretical background is a
IEC 60287 standard is quite accurate for individual power lines laid
Neher-McGrath model [2], which was generalized later by
in accordance to standard practice. However, when multiple power
lines should be placed together in a limited space, it is necessary to many authors, in particular G.J. Anders [3]. The
consider the electromagnetic and thermal influence of cables to each Neher-McGrath model relies on the thermal equivalent circuit
other. The increase of losses due to mutual influence of cables may technique. The parameters of the equivalent circuit are
be greater than is commonly believed. calculated by using a simplified 1D-model of the thermal field.
It is shown that the use of electromagnetic and thermal FEA Electromagnetic part of a calculation intended to assess the
simulation successfully solves this problem not only in the context of resistive and dielectric losses in the cable, is also based on a
research, but also in the engineering practice. The FEA-based
approach and the software tool is proposed for calculation ampacity simplified model of the skin effect and proximity effects.
of underground cable lines. Taking into account the details of cable laying in full extent
The electromagnetic FEA simulations can also assess the lead to the fact that manual calculation of the ampacity
electromagnetic environment near power lines and quantitatively according to IEC 60287 standard is a challenging and
evaluate the effect of magnetic shielding. cumbersome task. Therefore a number of software tools are
The proposed method can be used in cases where the standard proposed for both professional use (like CymCap) and for
IEC 2087 calculation gives unreliable results due to unusual cable
education [16].
line formation, inhomogeneous soil, presence of metallic or concrete
supports and other difficulties. When the cables are located close to each other, it is
necessary to take into account their thermal and
Keywords—Cable ampacity, buried cable, finite element electromagnetic interference. Electromagnetic interference is
analysis, multiphysics, magnetic shielding, proximity effect, cable the proximity effect and the skin effect, and the fact that,
duct bank depending on the chosen grounding mode, the screens and
sheaths appear electrically connected into a closed loop.
I. INTRODUCTION Many authors noted insufficient accuracy of the standard
methods of evaluating the proximity effect. For example, the
T he rated current of the underground electric power cable
line is limited by the maximal allowable temperature of
cable conductor, given by the standard or the cable
paper [17] describes the special study of electromagnetic
interaction of subsea cables, which has the metallic armor
manufacturer. The temperature raise in turn depends on thicker than usual.
resistive and dielectric losses in cable as well as on thermal The thermal interference is that neighboring cables warm up
conductivity of cables materials and the ability of surrounding to each other and the surrounding soil. Accounting of the
media to conduct and dissipate the heat flux. mutual heating is especially complicated when cables are laid
To calculate the ampacity of the cable line one must first out in the open air or in restricted airspace - in a pipe or a
rectangular conduit. In such case, the multiphysics model
should be supplemented with fluid dynamics analysis.
S. D. Dubitsky is with Tor Ltd., St. Petersburg, Russia (phone: Today FEA software [12] allows combining into a single
+7 812 710 1659; e-mail: [email protected]).
G. V. Greshnyakov, is with Sevkabel plant, R&D Department, St.
model the AC electromagnetic analysis, grounding electric
Petersburg, Russia. He also works with the Department of Cable Engineering, circuit, and the thermal analysis. Because the material
St. Petersburg State Technical University, Russia (e-mail: properties, such as electric resistivity, depends on the
[email protected]). temperature, one have to repeat electromagnetic and thermal
N. V. Korovkin is a head of Electromagnetic Theory Department,
St. Petersburg State Technical University, Russia (e-mail: analyses iteratively until the solution converges. Complexity of
[email protected]).

ISSN: 1998-4316 12
INTERNATIONAL JOURNAL OF ENERGY Volume 9, 2015

the model, however, is quite acceptable for engineering cross-sectional area Ώ.


practice. Solving the equations (1) and (2) one obtains the
The advantages of FEA model is particularly evident when distribution of the current density in all conductive parts of the
power cable line has rather complex structure of the , i.e. model: conductor, shield, metallic sheath, and some metallic
includes soil layers with different properties, strong supporting structure.
electromagnetic interference between cables, metallic
B. Model Geometry
supporting structure, crossing pipelines e.t.c.
In this paper, we consider only steady-state cable ampacity With two dimensional electromagnetic FEA simulation the
calculation. Nevertheless, the FEA based approach, allows the model geometry contains the cross-sections of all cables,
ampacity calculation in transient conditions: the long-term buried into the soil on the given depth. The left and right side
transient, where the a priory known load curve allows a short- borders of the modelling area located far enough to assign on
term uprating due to the inertia of thermal processes, and it the no-field border condition.
short-term transient, such as the raise of cable temperature due Our experiments show that for a model containing one cable
to short circuits of different kinds. line increasing the model width over 15 m does not effect on
The history of FEA analysis for cable ampacity calculation the solution accuracy. The model allows taking into account
begins presumably with [4], where the transient heat transfer the electric conductivity of soil as well as supporting metallic
FEA analysis was used three-phase buried cable line. Later parts or pipes nearby.
many authors have contributed to application of the FEA
technique for accurate predicting the ampacity of a cable line.
Those include: clarification of the model geometry – the shape
and the size of modelling area, optimal mesh density [5], short-
term and long-term transient simulations [6], [7], taking into
account the effect of the temperature on the cable losses,
combining the heat transfer analysis with fluid dynamics [8],
[9], estimation of resistive AC losses using the electromagnetic
FEA model [10]. The accumulated engineering experience of
the FEA simulation of the temperature field of cable lines was
summarized in the IEC technical report [11].
The contribution of this paper is the combining of AC
magnetic FEA simulation, Kirchhoff's equations of the
grounding circuit , and steady state heat transfer FEA analysis
into a single model of the power cable line.
Fig. 1 The model geometry

II. ELECTROMAGNETIC MODEL The discretized cross section of the cable is shown on the
figure 2.
A. Equations of AC Magnetic Field In the real world, the conductive parts of the cable are made
The governing equations of quasi-stationary magnetic field from separate wires or strips. Constructing the FEA model one
in frequency domain are written with respect to the phasor of can include the detailed geometry of wires or replace them by
the vector magnetic potential A, which has in the 2D-domain a solid metal cylinder. In many cases, the conductor wire
only one nonzero component A = Az [11]: structure plays an important role and cannot be neglected, for
∂  1 ∂A  ∂  1 ∂A  example with modelling of a pulse mode, high frequency
 +   = − jextern + jωσ ⋅ A , (1)
∂x  µ ∂x  ∂y  µ ∂y 
losses and others. In our case – the steady state simulation by
the fundamental frequency – the exact representation of the
where μ – is the absolute permeability (H/m), σ – electric conductor’s structure does not increase the accuracy, but
resitivity (S/m), ω – cyclic frequency (rad/s), jextern – the requires much more resources. Moreover, the exact modelling
external current density (A/m2). of the wires is not an easy task because of some uncertainty of
The need of taking into account of the grounding circuit the shape of deformed wires and the contact resistivity
(with one end, with two ends or with transposition) requires between them.
combining the field equation (1) with the Kirchhoff’s equation
of the connected circuit. The equation of a circuit branch
containing a solid conductor in magnetic field looks like this:
U
I = = −σ ∫ iωA ⋅ ds , (2)
R Ω
where U – is the conductor voltage drop (V), R – the DC
resistance (Ohm), The integration is made over the conductor’s

ISSN: 1998-4316 13
INTERNATIONAL JOURNAL OF ENERGY Volume 9, 2015

Fig. 2 The cable cross section with the FEM mesh

A separate question is how to choose properly the cross


section and the conductivity of the solid cylinder representing
the stranded conductor. In our experience, the best results can Fig. 3 Grounding the cable with one side
be obtained by choosing the inner and outer diameters of the
conductor the same as in reality. Acting in this way we set the
total cross sectional area a bit more than the sum of cross
section area of all wires. To compensate that we propose
proportionally decrease the electric conductivity and the
thermal conductivity of the simplified conductors.
C. Single Point and Both Ends Grounding
The shield of a cable section can be grounded with one side
or with two sides. With two-side grounding the closed loop is
formed for circulating current. This current is induced by the
alternating magnetic field created by the cable conductor
current. The one-side grounding does not provide the loop for
induced currents. On the other hand on the unbounded end of
the cable shield the induced voltage is observed, that should be
limited for sake of safety. We have to note that even with one-
side grounding of the cable having both a screen and metallic
sheath, these two are always electrically connected with both
sides of the cable. This forms a closed loop for circulated Fig. 4 Grounding the cable with two sides
current even with one-side grounding.
Presence or absence of a closed loop significantly affects on The values of resistance in the grounding scheme are known
the amount of losses in the shield and sheath. To consider with some degree of uncertainty. Therefore, we evaluated the
those one have to couple field equations (1) with the circuit sensitivity of the FEA solution to the values of the resistances
equations (2). RgX and Rground. The study shows that the variation of
resistance Rg in the range from 1 to 10 Ohms has virtually no
effect on the integral value of losses. The earth resistance
Rground has almost no effect for our model until the three phase
cable loading is symmetric and zero sequence current is almost
zero.
D. Dielectric Losses
According to IEC 60287-1-1 the dielectric losses per unit
length of the cable can be calculated by the known value of the
dielectric loss factor tgδ:
Wd = ωCU 2 ⋅ tgδ , (3)

ISSN: 1998-4316 14
INTERNATIONAL JOURNAL OF ENERGY Volume 9, 2015

where ω = 2πf, С is the capacitance per unit length (F/m), standard gives not enough accuracy in thermal state due
Uo – is the voltage to earth (V). to high influence of cables to each other;
The capacitance of a cylindrical capacitor is calculated by: 4. Develop the software tool for gradual replacement IEC
2πεε 0 calculation by FEA modeling where it is applicable.
C=
( )
D
ln d i
c
(4) Here we discuss pp, 1 and 2, and some preliminary
considerations about p. 3.
As long as we remain in the class of cables and conductors
with cylindrical conductors and screen screens the refinement A. FEA Model of Common Cable Formations
of formulas (3) and (4) by means of FEA is not required. The The modern approach to field simulation in electrical
FEA model of dielectric losses may be needed for more equipment often is multidisciplinary [13] in order to catch the
complex geometry configurations such as cable joint and mutual interference of processes from different domains of
termination. physics.
The steady-state simulation loop begins with magnetic field
III. HEAT TRANFER MODEL simulation (1.) for obtaining the spatial distribution of the
restive losses. The calculated resistive losses are summed up
The thermal state of the loaded power cable line is defined
with the dielectric losses (2.) and transferred to the heat
by the partial differential equation of thermal conductivity.
transfer analysis (3.). The thermal simulation gives us the
With steady state analysis it is reduced to:
temperature field, which is used for adjusting the conductivity
∂  ∂T  ∂  ∂T 
 λx  +  λx  = −q , (5) of copper and aluminum (4.). Then the loop (1. – 4.) is
∂x  ∂x  ∂x  ∂x  repeated until the solution converges (normally 3-4 loops is
where T is the temperature (К), t – time (с), λ – the thermal sufficient).
conductivity (W/(m·K) ), q – the heat source density (W/m3). The simulated cases include the flat cable formation (fig. 5
The thermal conductivity equation (5) is solved numerically and 6) and the touching trefoil formation (fig. 7, 8).
on the same computational domain as the magnetic field
equation (1) (see fig. 1) with the difference that the air above
the ground surface is excluded from the domain. On the side
boundaries of the domain we define the boundary condition of
thermal insulation, on the bottom border – an isothermal
boundary condition with the value of 4 deg. C, which is almost
constant throughout the year. On the earth surface the
convective boundary condition is set with the ambient
temperature T0=25 deg C and the convection coefficient α.
The suitable value of the convection coefficient we choose by
the dimensionless empirical equation:
Nu = 0.54 ⋅ (Pr⋅ Gr )0.25 , (6),
where Nu is the Nusselt number, Pr is the Prandtl number,
and Gr is the Grashof number.
From (6) obtain the convection heat transfer coefficient α:
λ Fig. 5 Magnetic field and current density with line cable formation
α = Nu ⋅ , (7)
Lref
where Lref is a characteristic length of the model.
Using the equation (6) takes into account the average wind
speed if such data are available.

IV. SIMULATION RESULTS


The long-term research agenda in FEA modeling of coupled
electromagnetic and thermal fields of underground power
cable lines is as follows:
1. Formulate the mathematical model, evaluate the impact of
the assumptions made;
2. Obtain FEA solution for the standard practice of cable
laying – the flat and trefoil formations, compare results
of FEA model with IEC 60287 calculation and tune the
FEA-model;
3. Identify the cable formations where the IEC 60287

ISSN: 1998-4316 15
INTERNATIONAL JOURNAL OF ENERGY Volume 9, 2015

Fig. 6 Temperature field and heat flux vectors with line formation

Fig. 8 Temperature field and heat flux with triangle formation

The resistive losses in cable conductors, screens and sheaths


with two different formations are summarized in the table 1:
Table 1: Resistive losses in cable elements

B. Cable Duct Bank Analysis


Fig. 7 Magnetic field and current density with triangle formation
In urban areas is often needed to lay a number underground
power cable lines in a limited space. For example, in St.
Petersburg, Russia, the input of eight 110 kV cable lines and
two 330 kV lines to the Vasileostrovskaya substation should
be constructed in a well developed urban area under the street
with heavy traffic. Another complication is that the owners and
designers of different lines are different. The common part of
all ten lines is the 400 meter long corridor, which can be
implemented as a concrete duct bank shown in the fig. 9.
The individual cables in the duct bank affects to each other
in both electromagnetic and thermal ways. The
electromagnetic influence is that the alternative current in each
conductor induces eddy currents in conductors and shields of
other cables. The induced currents critically depend on screen
grounding with one side, both sides or with transposition.
Anyway, the induced eddy currents create stray losses in all
cables.
We have developed a script creating the geometric model of
virtually arbitrary underground cable system. The complete

ISSN: 1998-4316 16
INTERNATIONAL JOURNAL OF ENERGY Volume 9, 2015

installation is described in XML using the specially designed failure mode the transient thermal analysis is needed to know
XML-scheme. That provides good flexibility, semi-automatic for how long time each line can be overloaded. The significant
syntax checking, and allows easily create a suitable graphical amount of calculation requires some software tool or script to
user interface, if necessary. automate repetitive tasks.

Fig. 9 The concrete duct bank

The computational scheme remains mostly the same as


described above. We start with AC magnetic analysis coupled
with grounding electric circuit. Electromagnetics analysis
allows to estimate the increase of Ohmic losses due to
proximity and skin effect (fig 10).
Next, the calculated Ohmic losses are summed with
dielectric losses and transferred to the thermal analysis, either
steady-state or transient one.
The assumption that needs to be verified and validated is the
replacement the convective and radiative heat transfer inside
the duct to an equivalent conductive heat transfer. To do that
we choose some equivalent value of heat conductivity for the
air inside the duct, which is four time greater when the heat
conductivity of air by the same temperature and pressure. It is
considered that this replacement to certain extent is equivalent
to the convective and radiative heat transfer from the cable to
the internal surface of the duct.
For successful design of the duct bank the extensive series
of calculation should be done, with both normal operational
load and various failure modes, when one or two lines
switched off whereas some other are overloaded. In heavy

ISSN: 1998-4316 17
INTERNATIONAL JOURNAL OF ENERGY Volume 9, 2015

Fig. 10 Electromagnetic field in the cable duct bank located close to the center of the duct bank. The value of this
ration as high as 1.92…1.98 should be considered by planning

Table 2: Losses and temperature of the cables in the duct bank

the load of each individual line.

V. CONCLUSION
Proposed further development of prediction the ampacity of
underground cable line using multiphysics FEA simulation.
The main contribution is the detailed consideration of cable
grounding, taking into account more than one electromagnetic
screen (namely the copper shield and the aluminum sheath).
The proposed approach combines in a single model the AC
magnetic FEA simulation, the grounding circuit, and the heat
transfer FEA. The first two parts coupled by the strong link,
i.e. they produced a single matrix after discretization. The
magnetic and thermal parts of the model a coupled together by
a two-directional loose (consecutive) link.
The FEA based calculation gives almost the same result as
the standard IEC 60287 calculation when the construction of
the cable line is ordinary. The dedicate software gives the
answer almost as quickly as the IEC based software.
Benefits of the multiphysics FEA appears in situations more
complex than those described in the standard, such as
heterogeneous soil with thermal backfill, using of steel or
concrete supporting construction. An important case is a line
with two or more circuits.
Benefits of the FEA simulation also expected with very
rapid transient conditions, such a direct lightning stroke [14].
Fig. 11 Steady-state temperature field in the duct bank Moreover, the FEA simulation of magnetic field gives
exhaustive information about inductive interference of two or
The calculation result of one of normal operation mode of more circuits, both cable and overhead ones. In addition, the
the duct bank is show in the table 2. The table two gives the external magnetic and electric field profiles on the earth
data for further analysis. In particular, it is of interest the surface can be obtained. The minimization problem of external
increasing of Ohmic losses due to proximity effect. The electromagnetic field from the point of view of circuit theory
AC/DC loss ratio is, as expected, much higher for the cables

ISSN: 1998-4316 18
INTERNATIONAL JOURNAL OF ENERGY Volume 9, 2015

was investigated in very general way [18-21] as an inverse [19] Korovkin N.V, Adalev A.S., Hayakawa M. De-embedding microwave
fixtures with the genetic algorithm В сборнике: IEEE 6th International
problem of circuit theory. The FEA model also gives Symposium on Electromagnetic Compatibility and Electromagnetic
comprehensive information to fulfill the rules of Ecology, 2005, Proceedings IEEE 6th International Symposium on
electromagnetic ecology and designing magnetic shielding Electromagnetic Compatibility and Electromagnetic Ecology, 2005. St.
when needed. Petersburg, 2005. С. 190-194.
[20] Korovkin N., Kuchinsky G., Kozhevnikov M., Schilin O., Vekhoreva
L., Selina E., Potienko A. Compact power high-voltage pulse capacitor
REFERENCES for quasi-rectangular nanosecond impulses formation. - in: Digest of
Technical Papers-IEEE International Pulsed Power Conference 12th
[1] IEC Standard-Electric Cables – Calculation of the Current Rating – Part
IEEE International Pulsed Power Conference. Monterey, CA, USA,
2: Thermal Resistance – Section 1: Calculation of the Thermal
1999. С. 1449-1452
Resistance, IEC Standard 60287-2-1, 1994–12
[21] Korovkin N., Marthe E., Rachidi F., Selina E. Mitigation of
[2] J. H. Neher, M. H. McGrath, “Calculation of the temperature rise and
electromagnetic field radiated by PLC systems in indoor environment. -
load capability of cable systems,” AIEE Trans., Vol. 76, Part 3, 1957,
International Journal of Communication Systems. 2003. Т. 16. С. 417.
pp. 755-772.
[3] G. J. Anders Rating of electric power cables: ampacity computations
Simon D. Dubitsky is currently with Tor Ltd, St. Petersburg. He receives
for transmission, Distribution, and Industrial Applications. - McGraw
MsC degrees in electrical engineering in 1983 and in computer science in
Hill Professional, 1997, 428 c
2003 both from SPbSPU. His main area of activity is development of
[4] N. Flatabo Transient heat conduction problems in power cables solved
QuickField FEA software in cooperation with Tera Analysis company locates
by the finite element method. -IEEE Trans. on PAS. Jan, 1973 pp. 56-
in Svendborg, Denmark. Main research interest is implementing FEA as a
63
handy tool for everyday engineering practice, advanced postprocessing of
[5] [5] Aras F., Oysu C., Yilmaz G. An assessment of the methods for
electromagnetic field solution, multiphysics FEA analysis coupled with
calculating ampacity of underground power cables //Electric Power
circuit equations and surrogate models.
Components and Systems. – 2005. – Т. 33. – №. 12. – С. 1385-1402
[6] Liang Y. Transient temperature analysis and short-term ampacity George V. Greshniakov is currently a head of laboratory in the Sevcabel
calculation of power cables in tunnel using SUPG finite element method research institute. He receives the MsC and PhD degrees in electrical
//Industry Applications Society Annual Meeting, - IEEE, 2013. - С. 1-4. engineering from St. Petersburg technical university (SPbSPU) in 1983 and
[7] Haripersad P. Uprating of cable current capacity for Utilities where 1992 respectivelly. He is also a docent in SPbSPU teaching cable and
load cycle profiles are known - Cigre 2005 Regional Conference paper, insulation engineering. Author of more than 30 reviewed papers. His research
Capetown interest is electromagnetics ant thermal analysis of power cable installation
[8] Sedaghat A., de Leon F. Thermal Analysis of Power Cables in Free Air: and development of advanced cable accessories.
Evaluation and Improvement of the IEC Standard Ampacity Nikolay V. Korovkin, professor, is currently head of Electromagnetic Theory
Calculations. - IEEE Transactions on Power Delivery Department of St. Petersburg State Polytechnic University (SPBSPU). He
[9] Mahmoudi A., Kahourzade S., Lee D. S. S. Cable ampacity calculation received the M.S., Ph.D. and Doctor degrees in electrical engineering, all
in heterogeneous soil using Finite Element Analysis //Power from SPbSPU in 1977, 1984, and 1995 respectively, academician of the
Engineering and Optimization Conference (PEOCO), 2011 5th Academy of Electrotechnical of Russian Federation, (1996) Invited Professor,
International. – IEEE, 2011. – С. 416-421 Swiss Federal Institute of Technology (EPFL), Lausanne (1997), Professor,
[10] Labridis, D.; Hatziathanassiou, V. "Finite element computation of field, University of Electro-Communications, Department of Electronic
forces and inductances in underground SF6 insulated cables using a Engineering, Tokyo, Japan (1999-2000), Professor EPFL (2000-2001), Otto-
coupled magneto-thermal formulation", Magnetics, IEEE Transactions fon-Guericke University, Germany (2001-2004). Head of the Program
on, On page(s): 1407 - 1415 Volume: 30, Issue: 4, Jul 1994 Committee of the Int. Symp. on EMC and Electromagnetic Ecology in St.
[11] IEC Technical Report TR 62095, Electric Cables—Calculations for Petersburg, 2001-2011.
Current Ratings—Finite Element Method, 2003 His main research interests are in the inverse problems in electro-magnetics,
[12] Claycomb J. R. Applied Electromagnetics Using QuickField and optimization of power networks, transients in transmission line systems,
MATLAB. – Laxmi Publications, Ltd., 2010. impulse processes in linear and non-linear systems, "soft" methods of
[13] Dubitsky S. D., Korovkin N.V., Hayakawa, M.; Silin N.V., Thermal optimization, systems described by stiff equations, the problems of the
resistance of optical ground wire to direct lightning strike electromagnetic prediction of earthquakes and identification of the behavior
//Electromagnetic Theory (EMTS), Proceedings of 2013 URSI of the biological objects under the influence of the electromagnetic fields
International Symposium on. – IEEE, 2013. – С. 108-111.
[14] Korovkin N.V., Greshnyakov G.V., Dubitsky S.D. Multiphysics
Approach to the Boundary Problems of Power Engineering and Their
Application to the Analysis of Load-Carrying Capacity of Power Cable
Line - Electric Power Quality and Supply Reliability Conference
(PQ2014), 11-13 June 2014, Rakvere, Estonia
[15] Gómez F. Á. et al. Numerical study of the thermal behaviour of bare
overhead conductors in electrical power lines //Proceedings of the 10th
WSEAS international conference on communications, electrical &
computer engineering, and 9th WSEAS international conference on
Applied electromagnetics, wireless and optical communications. –
World Scientific and Engineering Academy and Society (WSEAS),
2011. – С. 149-153.
[16] Aras, F. A. R. U. K., Yunus Bicen, and M. E. L. I. H. Inal. "Computer
Assisted Teaching Of Underground Power Cables (Powcabgui) For
Graduate Students." WSEAS Transactions on Advances in Engineering
Education 3.8 (2006): 760.
[17] Chien, Chang Hsin, and Richard Bucknall. "Evaluation of the proximity
effect upon the impedance characteristics of subsea power transmission
cables." WSEAS Transactions on Power Systems 1.7 (2006): 1141.
[18] Adalev A.S., Hayakawa M., Korovkin N.V., Nitsch J.B., De-embedding
and unterminating microwave fixtures with the genetic algorithm IEEE
Transactions on Microwave Theory and Techniques. 2006. Т. 54. No. 7.
С. 3131-3139.

ISSN: 1998-4316 19

You might also like