Measurement 89 (2016) 166–170

Contents lists available at ScienceDirect

Measurement
journal homepage: www.elsevier.com/locate/measurement

Analysis of magnetic field generated by overhead cables

Tong Zi-Yuan a,b, Dong Zhao-Yang b, Tong Min-Ming a,⇑
a
Department of Information and Electrical Engineering, China University of Mining and Technology, Xuzhou, China
b
School of Electrical and Information Engineering, University of Sydney, NSW, Australia

a r t i c l e i n f o a b s t r a c t

Article history: Overhead cables are widely applied in power distribution and transmission networks due to their
Received 7 January 2016 financial and geographical merits. But since the environment issue has become a general concern,
Received in revised form 25 March 2016 negative impacts of overhead cables are no longer negligible. Although overhead transmission lines are
Accepted 1 April 2016
gradually being replaced by underground cables for the purpose of improving the security and stability
Available online 2 April 2016
of power transmission system, a large number of overhead cables are still in use. If buildings are built
below the high voltage cables, people who stay inside will be adversely affected by power frequency
Keywords:
magnetic fields. In order to solve this issue, our paper studies the distribution of magnetic field produced
Overhead cables
Magnetic field
by a 500 kV cable, and proposes a shielding method to reduce the indoor field intensity. In this research,
Shielding protection field distribution is analyzed through simulation, and the maximum indoor field intensity is calculated
and compared with safety limits in the guide rule of limitation. Shielding method proposed in this paper
provides good protection performance and halves the indoor intensity of magnetic field produced by the
500 kV overhead cable.
Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction countries have proposed strict national standard of field intensity.

Cables used in power systems are generally categorized as
overhead cable and underground cable. Although stability of
overhead cables can be greatly affected by meteorological changes,
which often causes electrical failures, they are commonly used in
power transmission systems for the cost effective advantage. As
overhead cable installation has no geographical limitation, it is
often used for long distance power transmission. In recent years,
with the development of power system, increased load on power
grid and concerns for environment result in gradual replacement
of overhead cables. Underground cables start to take place. But this
process is highly costly and requires time for system re-planning,
so there will be years until overhead cables are completely out of
the power system. Since a great number of overhead cables are still
in operation, the adverse impact needs to be dealt with.
Magnetic field is a public concerning pollution problem in
modern society. If the magnetic field (MF) has high intensity or
frequency, it could harm human health [1,2]. Changing power
frequency magnetic fields produced by power cable is usually
50 Hz or 60 Hz. They are not high frequency fields, but when the
power level of cable is high, magnetic field will have intense
strength. To provide protection from magnetic field, many
⇑ Corresponding author.
E-mail address: jctmm@163.com (M.-M. Tong).
Sweden presented a rule to limit the power level of transmission
lines passing uptowns to ensure the magnetic field intensity does
not exceed 20 lT. This limiting value is based on the research of
relations between magnetic field level and the childhood leukemia
[3]. This standard is also applied in the U.S. Several state and local
governments have adopted electric and magnetic field limits for
transmission lines, 10 kV/m within rights-of-way and 2 kV/m
at the edge of rights-of-way for electric fields and around 20 lT
for magnetic fields [4]. ICNIRP (International Commission on
Non-ionizing Radiation Protection) formally published the ‘‘Guide
rule of Limitation of varying of electric, magnetic and electromag-
netic field (under 300 GHz)” in April, 1998 [5]. This Guide rule has
introduced the limitation of direct and indirect effect of magnetic
field. Limitations for power frequency electromagnetic fields are
detailed in Table 1.
Currently, there are researches aiming at analyzing magnetic
field produced by high voltage transmission lines. Machado
evaluated the electromagnetic field of underground cables in
2010 [6]. He obtained the magnetic field distribution inside and
outside the cables. With these data, Machado proposed a Magnetic
Field Mitigation Shielding for power cables, and analyzed the mag-
netic shielding effects of 2 different materials for underground
three-phase cable in a flat configuration. A comparison is made
between the aluminum (conductive) and the steel (ferromagnetic)
shielding cases [7]. It turned out that the aluminum is more
efficient. Machado’s researches are focused on adding shielding

http://dx.doi.org/10.1016/j.measurement.2016.04.004
0263-2241/Ó 2016 Elsevier Ltd. All rights reserved.
 Z.-Y. Tong et al. / Measurement 89 (2016) 166–170 167

Table 1 2.2. Vortex field analysis

Guide rule of Limitation of varying of electric, magnetic and electromagnetic field.

Affected Basic limits EMF intensity Magnetic Current

! !
In 3D vortex field, both B (field density) and J (current
population (mA/m2) (kV/m) density (lT) (mA)
density) have three components. Then there are six unknowns in
50 Hz 60 Hz 50 Hz 60 Hz the function. If using the magnetic vector potential to solve the
Occupational 10 10 8.3 500 417 1 equation, scalar potential cannot be subtracted out, so the
population mathematical model for 3D vortex field is important.
General 2 5 4.2 100 83 0.5
In the analysis of 3D vortex field, the research field is usually
population
divided into two parts, the vortex area and non-vortex area. In
vortex area, both electric and magnetic field need to be described.
In non-vortex area, only magnetic field needs to be described. The
combination of vector potential and scalar potential is required in
cases on cables, which is highly effective on minimizing magnetic eddy current region. In the non-eddy zone, only vector or scalar
flux density generated by cables. But for overhead cables that have potential is required.
already been set up, it is not applicable to install extra component, !
since it’s not safe to let cables bearing extra weight. Based on Select vector magnetic potential A and A  u as base, followed
!
Machado’s research on shielding materials, aluminum is used as coulomb gauge r  A ¼ 0, classify source current in non-eddy
!
shielding for protection in this paper and shielding effect is further current region. Then uniqueness of A and u is guaranteed, and
analyzed regarding its thickness. stable numerical solution can be obtained for any harmonic field.
There are many other researches on power cables magnetic The vortex field function is,
field sensing, while very few aimed at the effects and shielding
1 ! 1 ! !
on residential buildings. In this paper, we focus on providing a r  r  A  r r  A þ rðjx A Þ ¼ 0 ð1Þ
feasible strategy for magnetic field protection. Before the method
l l
!
is proposed, magnetic intensity and distribution are analyzed. It r  rðjx A  ruÞ ¼ 0 ð2Þ
has been proved by other researches that finite element method
! !
(FEM), finite difference method, charge simulation method can B ¼ r A; E ¼ jx A  ru ð3Þ
all be applied in complicated 3-D magnetic field analysis [8–12].
The non-vortex field function is,
In this research we use Ansoft Maxwell to achieve the magnetic
field simulation around high buildings. To improve the computa- 1 ! 1 ! !
r r A r r A ¼ J ð4Þ
tional accuracy, we picked out encrypted and subdivided regions l l s

that need to be noted with adaptive grid encryption technology. ! !

2. Method for magnetic field simulation and computing
2.1. Principle of finite element method
FEM is an approximate discrete interpolation method based on
variation principle. This method is to divide target area into small
regions, and solve each region with linear calculation. Final results
for the whole area can be obtained by synthesizing solutions from
all divisions. Small regions in this method are called ‘‘unit” or
‘‘finite element”. After they are created, the model is simplified
and computations are easier, only left with algebraic operations.
Values of unknown area can be obtained by applying linear inter-
polation in finite elements, which means the regional integration is
transformed into summations of data in units.
The finite element method for solving boundary value problems
has three processes. The first is to use the variation principle to
transform the boundary value problem into a variation problem,
also known as functional extreme value problem. The second
process is using the field dispersion and grid division as well as
approximation of function interpolation on the unit to transfer
the variation problem into a general extreme value problem of
ordinary multiple functions. Finally, the problem can be listed as
a set of algebraic equations. Numerical solution of these equations
is the results of boundary value problems [13].
Applying FEM in a practical issue requires applicable boundary
conditions to describe a physical quantity of boundary, which is
also the initial step of computation. In practical problems, there
are a variety of boundary conditions, commonly used are: (1) the
natural boundary condition; (2) the Dirichlet boundary condition;
(3) the subordinate boundary conditions; (4) the symmetric
boundary conditions; (5) the impedance boundary condition; (6)
Zero Tangential H Field; (7) Radiation Boundary. In the simulation
in Section 3.2, radiation boundary condition is used.
B ¼r A ð5Þ
! ! !
In the functions, B is field density; J is current density; J s is
source current density; r is conductivity; l is permeability. The
advantage of this method is that, the interface condition is natural
boundary condition, source current term is easy to deal with and
the precision is high. When the geometric dimensioning, media
electromagnetic parameters and source current distribution are
given, spatial distribution and temporal variation of regional
magnetic field (B or H) and eddy current density (Je) can be calculated.
3. Indoor magnetic field distribution analysis of high voltage
cable
3.1. Model of high voltage cable and residential building
Assuming a 500 kV single circuit transmission cable is crossing
above a building. The building is assumed as standard cuboid-
shaped residential buildings, with dimensions of 30 ⁄ 10 ⁄ 35 m.
Middle conductor is right above the middle of the construction.
Power delivers in form of parallel transmission. The distance between
center line and side line D = 13.72 m. Model is shown in Fig. 1.
In practical project, transmission lines are bundled conductors.
500 kV cable consists of 4 partial conductors. Each conductor has
radius r of 0.0148 m, and bundling radius R0 is 0.323 m [13]. For
calculation convenience, we simplify the cable model, which is
shown in Fig. 2. The computational formula of equivalent radius is,
rffiffiffiffiffi
n nr
R ¼ R0 ð6Þ
R0
In the formula above, n is the quantity of partial conductors. To
pffiffiffi
cable with 4 partial conductors, R0 ¼ d= 2, in which d is the space
between each sub conductor. In high voltage cable, the space is
generally in the range of 400–450 mm. The equivalent radius of
three phase cables A, B, C can be calculated with Eq. (6), and the
168 Z.-Y. Tong et al. / Measurement 89 (2016) 166–170

Fig. 3. Field intensity testing path.

Fig. 1. Relative location model of cable and construction.

same, and observe maximum MF when the vertical distance
between cable and building changes from 12 m to 2 m. The
maximum values are collected along the testing path, and the data
2r is shown in Fig. 6. We can see from Fig. 6 that the magnetic field
Rÿ intensity is beyond the standard limit when the distance is shorter
than 3 m [14,15].

4. Shielding effect study

R d
In order to keep the maximum field intensity within the safe
limits when the building is 3 m away from the cable, a shielding
method is proposed in the following section.

4.1. Principle of MF shielding

Fig. 2. Cross-section of conductor. MF shielding is to suppress or eliminate the interference caused

result is R = 0.2 m. When there is a current flow in the conductor,
the flow will produce magnetic field. The MF has different charac-
teristic with electric field: (1) Because of the quasi-static properties
of MF, MF of circuit is generated by current only and has nothing to
do with voltage; (2) MF at 50 Hz can easily transmitting through
most objects without being interfered.
In our research, we assume that the building material is brick-
concrete, and the thicknesses of wall and roof are 0.25 m and
0.2 m respectively. We also see the MF as 0 T in the area far away
from the cable, as MF generated by 3-phase conductors can offset
each other from certain distance.
3.2. Finite element model of magnetic field processing
In FEM analysis, we set simulation frequency as 50 Hz, phase
current amplitude as 1500 A, A phase angle as 0°, B phase angle
as 120°, C phase angle as 120°, and boundary conditions as
radiation boundary condition. Then we ran eddy current field
solver in Ansoft Maxwell 3D. The computational domain is 300%
size of basic simulation model. Regardless of the eddying effect
of overhead current-carrying cables and generatrix, the system
adaptively increases 30% of grid size after each step, and the
maximum energy error in computational domain is smaller than
5%. All current and voltage used in calculation are peak value to
calculate the highest risks of overhead cable.
by MF coupling. In this section, the static magnetic field is
discussed first. Both MF generated by magnet and conductive coil
scatter MF lines in space. Lines of magnetic force mainly go
through magnetic circuit with low magnetic resistance. In order
to achieve desirable shielding effect, high magnetic permeability
materials are often used. These materials have low magnetic
resistance, so that the lines of magnetic force are sealed in the
material. The shielding effect is achieved accordingly. For low
frequency alternating magnetic field, the shielding principle is to
use the high permeability of ferromagnetic materials to shunt
disturbing magnetic field. According to Machado’s research [7,8],
aluminum has good shielding performance.
Based on the magnetic flux continuity principle, the magnetic
field line is a continuous closed curve, so that the flux of closed
loop can be called magnetic circuit. Magnetic circuit theory
formula is,
U m ¼ Rm  /m ð7Þ

3.3. Magnetic field calculation results

We simulated the magnetic field at the plane 10 cm vertical to

the ceiling, shown in Fig. 3, in conditions with and without the
existing of the building, and the results are shown in Figs. 4 and
5. From the figures, we can see that MF can transmitting through
the walls easily, and concrete structure has very little effect on
MF shielding. In order to figure out when does the maximum field
intensity exceeds safe limits if the distance between cable and is
continuously being shorted, we let other conditions remain the Fig. 4. Magnetic field distribution at 10 cm vertical to the ceiling.
 Z.-Y. Tong et al. / Measurement 89 (2016) 166–170 169

Fig. 7. Shielding effect of metal shielding plate with variable thickness.

4.2. Shielding effect of metal shielding plate and rebar grid on

Fig. 5. Magnetic field at same plane without existing building.
magnetic field distribution

To decrease the MF effect on indoor environment produced by

500 kV voltage cable, we established aluminum plate on top
surface of the building with vertical height H of 6 m and relative
permeability lr of 1000, and analyzed the shielding effect with
different shielding thicknesses of 0 mm, 4 mm, 6 mm, 8 mm and
10 mm. Fig. 7 shows the effect of shielding with different thick-
nesses. If there is no shielding for MF protection, the maximum
magnetic intensity is around 44 lT. With shielding provided, the
MF intensity decreases exponentially with the increase of shielding
thickness, and slowly reaches for a steady value. And when a
10 mm aluminum plate is installed, the intensity can be minimized
to 30 lT. 30% of the field intensity is reduced. This indicates that
the aluminum plate has a great performance on MF shielding.
Fig. 6. Shielding effect of metal shielding plate with variable.
But metal plate in 10 mm thickness is also heavily weighed, less
aesthetic and material consuming, so we proposed alternative
In the formula, U m is the magnetic potential difference between
structure, rebar grid as shielding. As both the grid size and diame-
2 points, and /m is magnetic flux through magnetic circuit, which
ter of rebar grid have influences on shielding effect, these factors
can be express as,
are taken into consideration when the performance is analyzed.
Z
! ! Because of the existence of concrete and the colligation form of
/m ¼ B dS ð8Þ rebar connection, we can assume that rebars have ideal electrical
s
contact. The first model is built with fixed grid size and variable
Rm is the magnetic resistance between point a and b, rebar diameters which shows in Table 2. The simulation result is
Rb ! ~ shown in Fig. 8. We can see from the figure that magnetic field
H  dl intensity declines with the increase of the grid size.
Rm ¼ R a! ! ð9Þ
s
B dS Then another model is built with fixed rebar diameter and vari-
able grid sizes which show in Table 3. The simulation result is
If the magnetic circuit cross section is uniform, and the MF is shown in Fig. 9. We can see from the figure that shielding effect
homogeneous, then Eq. (9) can be simplified as, is in inverse ratio with grid size.
Hl l
Rm ¼ ¼ ð10Þ Table 2
BS lS Rebar grid with variable rebar diameter.

In this formula l is the permeability (H/m) of ferromagnetic Model Grid size (length  width cm) Rebar diameter (mm)
material, S is the cross section area (m2), l is the length (m) of mag- A 10  10 3
netic circuit. From Eq. (10), we can see that Rm is low when l is B 10  10 6
large. Magnetic flow travels along the closed path with low C 10  10 9
resistance.
The shielding principle for high frequency magnetic field is
based on other theories. High frequency MF can produce vortex
on the surface of shielding shell, and its reflection can offset the
original MF and enlarge the MF aside the shielding, so the line of
magnetic force will go around the shielding. Larger the vortex is,
better the shielding effect will be. Therefore, good materials need
to be applied in high frequency MF protection. Vortex effect
increases with the rise of frequency, and the increase stopped
when the vortex-produced counter MF can completely reject the
disturbing MF. Under the influence of skin effect, thin metal mate-
rial is sufficient for high frequency MF shielding, but to build an
effective 50 Hz magnetic field shielding, thickness of metal needs
to be taken into consideration. Fig. 8. The influence of rebar diameter on shielding effect.
170 Z.-Y. Tong et al. / Measurement 89 (2016) 166–170
Table 3
Rebar grid with variable grid size.
Model Grid size (length  width cm) Rebar diameter (mm)
D 10  10 6  Grid size of shielding has a larger influence on shielding effect
E 20  20 6 than rebar diameter.
F 30  30 6  The calculation method can be applied in magnetic environ-
G 40  40 6 ment evaluation around residential constructions below high
Fig. 9. The influence of grid size on shielding effect.
Comparing the results in Figs. 8 and 9, we can see that model E
has twice grid size and half rebar diameter as those of model A,
while model A has a better effect than model E. This indicates that
grid size has a larger influence on shielding effect.
5. Conclusion
In this research we applied vortex field solver of Ansoft Maxwell
3D in magnetic field calculation, and achieved simulation and
numerical analysis of magnetic field effect of high voltage cable
on buildings. By analyzing maximum field intensity with changing
distances between cable and building and different parameters of
shielding, the conclusions can be drawn as follows,
 When no shielding strategy is applied and the vertical distance
between 500 kV high voltage cable and building is no less than
3 m, the maximum magnetic intensity indoor is within the
standard limit.
 With the aluminum plate on roof slab with relative permeabil-
ity of lr ¼ 1000, the magnetic field intensity decrease in index
law with a linear increase of shielding thickness.
voltage cables.
References
[1] J. Nadakuduti, M. Douglas, M. Capstick, S. Kühn, N. Kuster, Application of an
induced field sensor for assessment of electromagnetic exposure from compact
fluorescent lamps, Bioelectromagnetics 33 (2) (2012) 166–175.
[2] International Electrotechnical Commission (IEC), Assessment of electronic and
electrical equipment related to human exposure restrictions for
electromagnetic fields (0 Hz–300 GHz), IEC Technical Committee 106,
Geneva, Switzerland, IEC 62311, 2007.
[3] H.P. Schwan, K. Li, Hazards due to total body irradiation, Proc. IRE 44 (Nov)
(1956) 2058–2062.
[4] L.I. Kheifets, The precautionary principle and EMF, in: Meeting on EMF
Biological Effects + Standards Harmonization in Asia & Oceania, Seoul, South
Korea, 2001.
[5] IRPA/ICNIRP Guidelines for limiting exposure to time-varying electric,
Magnetic and electromagnetic fields (up to 300 GHz), Health Phys. 74(4)
(1998) 494-522.
[6] V.M. Machado, FEM/BEM hybrid method for magnetic field evaluation due to
underground power cables, IEEE Trans. Magn. 46 (8) (2010) 2876–2879.
[7] V.M. Machado, Magnetic field mitigation shielding of underground power
cables, IEEE Trans. Magn. 48 (2) (2012) 707–710.
[8] Q. Huang, W. Zhen, P.W.T. Pong, A novel approach for fault location of
overhead transmission line with noncontact magnetic-field measurement,
IEEE Trans. Power Delivery 27 (3) (2012) 1186–1195.
[9] X. Sun, Q. Huang, Y. Hou, L. Jiang, P.W.T. Pong, Noncontact operation-state
monitoring technology based on magnetic-field sensing for overhead high-
voltage transmission lines, IEEE Trans. Power Delivery 28 (4) (2013) 2145–
2153.
[10] H.D. Bruns, C. Schuster, H. Singer, Numerical electromagnetic field analysis for
EMC problems, IEEE Trans. Electromagn. Compat. 49 (2) (2007) 253–262.
[11] G. Ni, J. Qian, et al., Numerical Analysis of Electromagnetic Field, first ed.,
Higher Education Press, 1996.
[12] G. Kamiński, P. Góralski, The electromagnetic calculations of complex motion
common magnetic circuit electromagnetic converter, Arch. Electr. Eng. 63 (1)
(2014) 125–133.
[13] R. Otin, Regularized Maxwell equations and nodal finite elements for
electromagnetic field computations, Electromagnetics 30 (1) (2010) 190–204.
[14] W.N. Fu, X. Zhang, S.L. Ho, A fast algorithm for frequency-domain solutions of
electromagnetic field computation of electric devices using time-domain
finite-element method, IEEE Trans. Magn. 49 (1) (2013) 530–535.
[15] W.K. Daily, Measurement and computation of EMF in electric power
substations’, IEEE Trans. Power Delivery 9 (1) (1994) 324–333.