EMTP Model of Anti-Earthquake Building Stroked by Lightning

Abstract

This paper models an anti earthquake building during lightning stroke to the building. The transient voltages due to lightning of modeled building at the connection point between the building and ground base are investigated to study the effect of the connection wire between the building and grounding base of the building. Lightning current was injected into the vertical pillar A top of the building. The transient voltages due to lightning at the top of the building at vertical pillar A with roof connection and vertical pillar A with first floor connection were observed and discussed. The modeling for different cases and discussion show good results, and agree with the physical meaning of real cases. The maximum voltage at first floor depends on pillars base grounding resistance and connection wires. The effect of varying connection wire inductance for different grounding resistance on first floor maximum voltage show that as the connection wire inductance increases, the first floor maximum voltage increases.

Full text

EMTP Model of Anti-Earthquake Building
Stroked by Lightning
M Nayel1,2, J. Zhao1, J. He2, Senior Member, IEEE, N. Nagaoka3, Member, IEEE
Z. Cai1, and Q. Wang1
Abstract--This paper models an anti earthquake
building during lightning stroke to the building. The
transient voltages due to lightning of modeled
building at the connection point between the building
and ground base are investigated to study the effect of
the connection wire between the building and
grounding base of the building. Lightning current was
injected into the vertical pillar A top of the building.
The transient voltages due to lightning at the top of the
building at vertical pillar A with roof connection and
vertical pillar A with first floor connection were
observed and discussed. The modeling for different
cases and discussion show good results, and agree
with the physical meaning of real cases. The
maximum voltage at first floor depends on pillars base
grounding resistance and connection wires. The effect
of varying connection wire inductance for different
grounding resistance on first floor maximum voltage
show that as the connection wire inductance increases,
the first floor maximum voltage increases.
This paper studies transient voltages in a building
due to lightning stroke. The studied building is an
anti-earthquake building. It has investigated the
transient voltages due to lightning stroke to modeled
building.
Index Terms—Lightning, Electromagnetic
transient analysis, Anti-earthquake building,
Grounding
I. INTRODUCTION
ightning frequently strike a tall structure such as
building, tower and etc.. Lightning stroke causes
serious problems and damage to human, electric
equipments, information system and the
telecommunication equipment by the lightning stroke.
Increase the safety and protection from lightning
strokes need a well modeling to the studied problems.
Buildings are one of the attractive sources of lightning
strokes. Isolation building from ground or bad
grounding of buildings may results to high potential of
buildings during lightning stroke. Anti earthquake
building is a building is designed to be separated from
building foundations by a rubber slat to absorb
earthquake waves. There are few papers are interested
in study of transient voltages in building due to
lightning strokes [1-3].
II. BUILDING DESCRIPTION
Fig. 1, shows the construction of the studied
building. The building has 3 floors with 4 rooms per
floor. A room is of 4×4 m2
with a height of 4 m. The
building is constructed from steel beam constructions.
The building is electrically isolated from ground due
to a rubber slat between building pillars and building
grounding bases.
III. EMTP SIMULATION
The system model consists of four parts :
1- Lightning surge
2- Building pillars
3- Building grounding bases
4- Connection wire
III.1 Lightning Surge
Lightning surge is modeled by a slope ramp
current source of 1 μs front time and 40 μs tale time
with 1kA amplitude. Paralleled with the current source
(a) Building pillars
(b) Building pillars bases
Fig. 1 Studied building construction
L
1 China Southern Power Grid Company, Guangzhou 510623,
China. (mnayel@csg.cn)
2 Sta t e K e y L a b of P o we r S ys t e ms , Dept. of Electrical
Engineering, Tsinghua University, B e ij i n g 1 0 00 8 4 , China.
(hejl@tsinghua.edu.cn )
3 Electric Engineering Ddepartment, Doshisha University, Kyoto,
Japan, nnagaoka@mail.doshisha.ac.jp

a resistance of 400 Ω. The lightning surge source is
connected to top of pillar A as shown in Fig. 2.
III.2 Building Modeling
Each pillar of the building is modeled as a
transmission line surge impedance to construct
impedances matrix which are presented as shown in,
Fig. 2. Pillars of the building have H shape for
horizontal pillar and square shape for vertical pillar as
shown in Fig. 3-a. A circular cylinder in the model as
in Fig. 3-b of which the inner and outer radii are
determined represents them so that the impedance
becomes nearly the same as the original one [2]. A
reinforced wall of the original building is not
considered in this model.
III.2.1 TL Model for the Building Pillars
It is necessary to obtain an equivalent cylindrical
conductor for an EMTP simulation of a transient
response. There exists an approach to determine the
radii of a cylindrical conductor equivalent to an
arbitrary cross-section conductor [4]. In this approach,
the equivalent radii are determined by the surface and
the cross-section areas being the same as those of the
arbitrary cross-section conductor. This approach has a
fairly good accuracy except the H-shape conductor
where the error reaches 20% which might not be
acceptable to carry an EMTP simulation.
The other approach has been proposed in Ref. [2].
The outer radius ri of an equivalent cylindrical
conductor are determined in the following steps.
(1) Obtain the radii r1 and r2 of circumscribed
and inscribed circles of a conductor as
illustrated in Fig. 3.
(2) Determine the outer radius r0 of an equivalent
cylindrical conductor r0=(r1+r2)/2.
(3) Determine the inner radius ri so that the
cross-section areas become the same in the
original and equivalent cylindrical
conductors.
Approach 2 gives a much better result in general for
an H-shape conductor, and also a better result in
general for other shapes (L, square , cylindrical ○).
A. Horizontal Pillars
After determine the equivalent outer/inner radius
of conductor, the surge will calculated by the LINE
CONSTANTS at pillar height. The horizontal pillar
transmission line model parameters are summarized in
Table 2.
B. Vertical Pillars
A vertical pillar parameters can not been obtained
by the EMTP LINE CONSTANTS, it is necessary to
obtain an equivalent horizontal pillar. It has been
experienced that an algebraic mean height gives a
greater surge impedance than a measured value. For
this reason, a geometrical mean height hm given in the
following equation is adopted [2], although the mean
height can not be applied to the case of the conductor
lowest height being nearly zero.
(1)
21m h
where h2 : highest conductor height, h1 : lowest height
The transmission line model parameters of vertical
and horizontal pillars are summarized in Table 2.
Zl : lightning model impedance,
Zvn ,Zhn : impedance of vertical and horizontal pillar at nth floor,
Lc : inductance of connected wire,
Rg : ground resistance / impedance
Fig. 2 EMTP model circuit of studied system
Table 1 Building pillars dimensions and equivalent conductor
dimensions
Pillar circumscribed and
inscribed circles radiuses
Equivalent
radiuses
r
1 [m] r2 [m] ro [m] ri [m]
□ shaped 0.566 0.523 0.5445 0.5168
H shaped 0.566 0.523 0.5445 0.5236
(a) □ and H shaped pillars
(b) Equivalent cylindrical conductor
Fig. 3 Pillars and their equivalent cylindrical conductor
Table 2 Pillars parameters
Horizontal pillars Characteristic imp. [Ω] Velocity [m/μs]
1st floor 97.84899 129.8
2nd floor 201.9765 251.2
3rd floor 229.1807 270.8
Roof 246.5827 279.5
Vertical pillars Characteristic imp. [Ω] Velocity [m/μs]
1st-2nd floor 154.2496 201.2
2nd-3rd floor 215.3969 261.8
3rd floor-roof 237.7602 275.4
hh =

(I) Lc = 1 μH
(II) Lc = 10 μH
(a) Case 1 Rg=10 Ω
(I) Lc = 1 μH
(II) Lc = 10 μH
(b) Case 3 Rg=100 Ω
Fig. 4 Calculated waveforms
III.3 Grounding models
The building bases in ground are modeled by a
pure resistance. There are three cases to study the
effect of varying connection wire inductance by
varying ground resistance of pillar base.
Case 1 : the ground resistance of a pillar base buried
in the ground is 10 Ω.
Case 2 : the ground resistance of a pillar base buried
in the ground is 50 Ω.
Case 3 : the ground resistance of a pillar base buried
in the ground is 100 Ω.
III.4 Connection wire
The building is connected electrically with
building bases of pillars by connection wires are
model by inductances The connection wires
inductance value is varied from 1 to 10 μH for each
wire connected between the building pillar and pillar
base in the ground.
IV. CALCULATED RESULTS AND DISCUSSIONS
IV.1 Voltage and current waveforms
Fig. 4 shows top and bottom voltages of pillar A
and injected current in the building for case 1 and case
3 when connection wire inductance is 1 and 10 μH.
Fig. 4-a, 4-b show the wave propagation incident
and reflection as a step stairs build the wavefront
voltage waveform. The time of incident and reflection
is about 0.2μ sec. This come from that wave
propagation of the building is about 246.13 m/μsec
and the building height is 12m then the time for wave
to propagate from top to bottom of the building is
about 0.07 μsec and this agree with delay of
waveform at bottom of pillar A.
To study the building and grounding
characteristics, the system is simply simulated as
shown in Fig. 5.
Fig. 5 Simple circuit for the studied building

(a) Max. voltage at the first floor (b) Max. voltage difference on connection wire
Fig. 6 Max. voltages vs. wire inductance for different cases
It is observed that input peak current in the
building top deceases from 997A to 971A with the
base ground resistance increases from Rg=10 Ω to100
Ω. This can be explained by the current distribution
between the building base ground resistance and
lightning impedance RL400 Ω, ib = i R
L/( RL+ Rgt)
where Rgt=Rg/9=1.1Ω,11.1Ω.
Fig 4-a shows that the top and bottom pillar
voltages oscillate when the pillar base ground
resistance is small but when it is high, the voltage
becomes steady and the characteristic is nearly
resistive as shown in Fig. 4-b. From this it can be
expected that the building and connection wire
inductance impedance is greater than 1.1Ω and less
than 11.1Ω. The building surge impedance Z=√(L/C)
when Lb=3.24μΗ, Rb=0.0586Ω, Cb=0.34μF then
Z=3.14~3.57 for Lct=0.11~1.1μH respectively.
The oscillation frequency in case of connection
wire 1μH shows 1.3 time that of 10μH where
f1/f2=√(L2/L1) = 1.138.
The wavefront of the voltage waveform at 10Ω
shows a square wave this is due to that the voltage is a
result of di/dt where current wavefront is a inclined
line. V=L di/dt = L 1000/1μs = 3350~4350 V which
agree reasonable with observed wavefront voltages at
Fig. 2-a (I and II) 3000 and 4500 V respectively.
IV.2 Connection wire effect
Fig. 6 shows the effect of varying connection wire
inductance with max voltage at the first floor and the
voltage difference on the connection wire.
The max. voltage at the first floor is shown in Fig.
6-a proportional with pillar base grounding resistance
as observed for Vmax/Rgt = 11029, 1031 and 10771 for
Rgt= 0.11, 5.56 and 11.11 Ω respectively shows a
nearly constant value. The change of max. voltage at
first floor for different cases is linearly proportional
with the variation of connection inductance.
The slope of the curve for three cases is equal to
di/dt = 158.5, 109.9 and 98.4 A/μs. the variation of
di/dt comes from the effect of building capacitance
where as the ground and connection wire impedance
increase the discharge current through building
capacitance increase and injection current through
pillar base to ground decrease. This explain that as the
pillar base ground resistance decrease, the current
injection in it increase and the voltage on the
connection wire increases as shown in Fig. 6-b. Also
same as the increasing on connection wire inductance
increases the voltage across the connection wire
increases.
V. CONCLUSIONS
From above calculation study and analyses of
results, one can obtained the following conclusions:
• The calculated results show a good agreement
with physical understanding of the studied system.
• The max. voltage at first floor and across the
connection wire depend on the base pillar ground
resistance and connection wire inductance.
• The voltage max. at first floor and across the
connection wire varies linearly with the base pillar
ground resistance and connection wire inductance.
• The voltage wave front shows resonance
oscillations at low pillar base ground resistance
which it is harmful to communication and
electronic devices.
• When pillar ground resistance is high the voltage
wave form is steady but it has a high value.
• If the ground pillar base resistance higher than the
building impedance, the transient voltage due to
lightning stroke has not a resonance oscillation.
• Decreasing connection wire inductance decreases
to the building voltages and oscillation.
VI. REFERENCES
[1] Ametani, A. Kojima,” Lightning induced voltages in an
Intelligent building”, 3rd International Symposium on
Consumer Electronics Proceedings, 14-16 Nov., Hong Kong,
pp. 133-138,1994.
[2] Ametai, N. Nagaoka, R. Koide, T. Nakanishi, “ Wave
propagation characteristics of iron conductors in an
intelligent building”, T. IEE Japan, Vol. 120-B(1), pp. 31-37,
2000.
[3] Y. Tsuchiya, M. Nayel, A. Ametani, S. Sekioka, Y.
Miyamoto, T. Kosumi,“ An Experimental investigation of
transient voltages in an intelligent building due to lightning

with special reference to grounding”, IEEJ Trans. PE, Vol.
123, No.11, pp. 1273-1279, 2003.
[4] A. Ametani, I. Fuse, “ Approximate method for calculating
impedance of multiconductor with arbitrary cross-section” Y.
IEE Japan, vol. 111-B(8), pp. 896-902, 1991.
VII. BIOGRAPHIES
M. Nayel was born in Assiut , Egypt, on April 15 1973. He received
the B.Sc. and M.Sc. degrees from Assiut University, Assiut, Egypt,
in 1996 and 1999. He get his Dr. Eng. from Doshisha University,
Japan. He was employed by Assiut University from 1996 to 1999 as
a demonstrator and from 1999 to 2004 as an assistant lecture and as
lecture from 2004. He is now a postdoctoral for two years in China
Southern Power Grid Company and Tsinghua University, China.
J. Zhao was born in Guangxi, China, on Jan. 10, 1961. He
received the M. Sc. Degree from Guangxi University in 1989.
He is now the director of TRC of China Southern Power Grid
Company.
Jinliang He (M’02–SM’02) was born in Changsha, China, in 1966.
He received the B.Sc. degree in electrical engineering from
Wuhan University of Hydraulic and Electrical Engineering,
Wuhan, China, in 1988, the M.Sc. degree in electrical engineering
from Chongqing University, Chongqing, China, in 1991, and the
Ph.D. degree in electrical engineering from Tsinghua University,
Beijing, China, in 1994, respectively. Currently, he is Vice Chief
of High Voltage Research Institute at Tsinghua University,
Beijing, China. He became a Lecturer in the Department of
Electrical Engineering at Tsinghua University in 1994, and an
Associate Professor in the same department in 1996. From 1994 to
1997, he was the head of the High Voltage Laboratory at Tsinghua
University. He was also a Visiting Scientist in Korea
Electrotechnology Research Institute, involved in research on metal
oxide varistors and high voltage polymeric metal oxide surge
arresters from 1997 to 1998. In 2001, he was promoted to a
Professor at Tsinghua University. Dr. He is a senior member of the
China Electrotechnology Society, and a member of the
International Compumag Society, the China representative of IEC
TC 81, the Vice Chief of China Lightning Protection
Standardization Technology Committee, and member of the
Electromagnetic Interference Protection Committee and the
Transmission Line Committee of China Power Electric Society, a
member of the China Surge Arrester Standardization Technology
Committee, a member of the Overvoltage and Insulation
Coordination Standardization Technology Committee and Surge
Arrester Standardization Technology Committee in Electric Power
Industry. Dr. He is the chief editor of the Journal of Lightning
Protection and Standardization (in Chinese).
N. Nagaoka was born in Nagaoya, Japan, on October 21, 1957. He
received B.Sc., M.Sc. and Dr. Eng. degrees all from Doshisha
University in 1980, 1982 and 1993, respectively. He joined the
Faculty of Engineering, Doshisha University in 1985, and has been
a professor since 1990. Dr Nagaoka is a member of both IEE and
IEEE.
Z. Cai was born in Guangzhou, China in July 10, 1967. He received
the B.Sc. And MBA degree from HUST and Jinan University,
China. He is now employed by China Southern Power Grid
Company.
Q. Wang was born in Shandong, China in Oct. 13, 1979. He
received the B.Sc. and M.Sc. degree from Xi’an Jiaotong
University, China in 2002 and 2005. He is now employed by China
Southern Power Grid Company.