Transient Response of a Double- Circuit Line to Direct and Indirect …en.rclp2018.com/upload/iblock/1b9/1b9df71235da5bf6d6053f... · 2018-04-26 · • In some areas, it is convenient

Transient Response of a Double-Circuit Line to Direct and Indirect Lightning Strikes

A. Borghettia, G. M. Ferrazb, F. Napolitanoa, C.A. Nuccia, A. Piantinic, F. Tossania

a. Dept. of Electrical, Electronic and Information Engineering, University of Bologna, Bologna, Italyb. High Voltage Equipment (HVEX), Itajubá, Brazil

c. Institute of Energy and Environment, University of Sao Paulo, Sao Paulo, Brazil

[email protected]

mailto:[email protected]

1. Introduction

2. Geometry of the Line, LIOV-EMTP Models and relevant assumptions

3. Time domain analysis

4. Statistical analysis

5. Conclusions

Transient Response of a Double-Circuit Line to Direct and Indirect Lightning Strikes by A. Borghetti, G. M. Ferraz, F. Napolitano, C.A. Nucci, A. Piantini, F. Tossani

VI Russian Conference on Lightning Protection, 17-19 April, 2018

• In some areas, it is convenient to install medium voltage (MV) distribution lines on the same pylons or poles holding a high voltage (HV) transmission line.

• Besides issues relevant to the electromagnetic coupling between the MV and HV phase conductors, the lightning protection of this double-circuit line configuration presents some peculiar characteristic, which are the object of this paper/presentation.



• Aim of this paper providing some insight of the lightning response of the multi-circuit configuration with particular focus on the occurring flashover phenomena.

• To accomplish that, use is made of the LIOV-EMTP-RV software, which has been suitably adapted to the specific characteristics of the case under investigation.

• Results of some statistical analysis will be presented too in this presentation.



1. Introduction




5. Conclusions





Geometry of the double-circuit overhead line



The HV line has an OGW grounded at every pole.

The MV line (15 kV insulation class) has a compact structure: insulated phase conductors (without shield) have reduced distances secured by periodical spacers suspended by an upper un-energized wire called Messenger.

Between two consecutive poles of the HV line, in the middle of each span, there is a shorter MV pole, having the function of sustaining the compact MV line. The span distance between the HV line poles is 70 m.

Fig. 1. Geometry and EMTP model of the considered concrete poles of the double-circuit overhead line: 69-kV transmission line.



Fig. 1. Geometry and EMTP model of the considered concrete poles of the double-circuit overhead line: 69-kV transmission line.

• Pole surge impedance Zp = 200 Ωaccording to the experimental data of [K. Nakada, H. Sugimoto, and S. Yokoyama, “Experimental facility for investigation of lightning performance of distribution lines,” IEEE Trans. PWDR, 2003].

• The pole of the 69-kV configuration has been split in two equal parts, one between the OGW connection and the M connection and the other between point M and ground.

• Dumping resistance Rp = 33 Ω and Inductance Lp = 5.33 μH according to [A. Ametani and T. Kawamura, “A method of a lightning surge analysis recommend-ed in Japan using EMTP,” IEEE Trans. PWDR, 2005.]



• Insulation breakdown occurrence taken into account by means of the Disruptive Effect (DE) criterion [M. Darveniza and A. E. Vlastos, “The generalized integration method for predicting impulse volt-time characteristics for non-standard wave shapes-a theoretical basis,” IEEE Trans. EI 1988].

v(t) voltage at the pole insulatorV0 min voltage to be exceeded before any breakdown process can startK dimensionless factort0 time at which |v(t)| becomes greater than V0A flashover occurs if integral D becomes greater than a constant value DE.

Table 1. Parameters assumed for the disruptive effect model

Voltage class (kV)

CFO (kV)

DE model parameters

V0 (kV)

K

DE (kVμs)

15 (with XLPE cable)

246

1

134

15 (bare conductor)

160

1

100

69

740

570

1.36

9180



• Assuming the critical flashover overvoltage (CFO) value for the HV lines is known, the parameters of the DE-model of the HV insulators are estimated by applying the following expressions [e.g. W. A. Chisholm, “New challenges in lightning impulse flashover modeling of air gaps and insulators,” IEEE El, 2010]

• For the MV line, DE model parameters have been inferred by using the V-tcurve provided by experimental data obtained for the 15-kV spacer and the bare conductors.

• For the covered conductors, the contribution of the XLPE cover (with thickness ranging from 2.3 to 3.3 mm) has been estimated by adding 100 kV to the voltage peak amplitudes of the above mentioned experimental curve for bare conductors [Wareing].

• The estimated V-t curve is in reasonable agreement with the experimental data for covered conductors presented in [R. E. de Souza, R. M. Gomes, G. S. Lima, F. H. Silveira, A. De Conti, and S. Visacro, “Preliminary analysis of the impulse breakdown characteristics of XLPE-covered cables used in compact distribution lines,” in Proc. 2016 33rd ICLP].

1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6

Time to breakdown ( s)

250

300

350

400

450

500

Vol

tage

(kV

)

Measured

Fitted

Modeled



Fig 2 shows the V-t experimental data along with the V-t curve obtained by means of the following function:

Fig. 2. Comparison between the V-tcurve fitted by using experimental data with XLPE covering and the one predicted by the DE model.

A, B and C are determined by a least-squares fitting procedure.

Assuming k = 1 and a 1.2/50μs voltage impulse parameters of the DE model (Table 1) are obtained by a nonlinear least-square minimization of the differences between the times to breakdown given by and the corresponding times given by



For the calculation of the induced voltages due to indirect strikes we have assumed the shielding effect of nearby buildings as negligible.

This effect has a significant influence on the lightning performance of overhead lines in urban areas as shown in [F. Tossani, A. Borghetti, F. Napolitano, A. Piantini, and C. A. Nucci, “Lightning Performance of Overhead Power Distribution Lines in Urban Areas,” IEEE Trans PWDR, 2017]

We have also neglected the effect of the lightning electromagnetic pulse in the calculation of the overvoltages due to direct strikes, as such a contribution has been shown to be minor in [F. Tossani, F. Napolitano, A. Borghetti, C. A. Nucci, G. P. Lopes, M. L. B. Martinez, and G. J. G. Dos Santos, “Estimation of the influence of direct strokes on the lightning performance of overhead distribution lines,” in 2015 IEEE Eindhoven PowerTech, PowerTech 2015, Eindhoven, Netherlands, 2015]



10-4

10-2

100

102

104

106

Current (A)

2

2.5

3

3.5

4

4.5

5

5.5

Volta

ge (V

)

104

Figure 3. Characteristic of the SAs in the MV line.

In the last decades significant improvements thanks to- more accurate models- constantly increasing computer performance



http://www.liov.ing.unibo.it

1. Introduction




5. Conclusions





M

D

RL

If not otherwise specified, the presence of SAs along the line is not considered.



Geometry

1400 m

50 m

Return stroke velocity 1.5×108 m/s Return stroke model Transmission Line (TL). Channel-base current 70 kA, equivalent Front time 3.8 μs

max di/dt = 26 kA/μs.



0 2 4 6 8 10 12 14

Time ( s)

0

50

100

150

200

250

300

350

Volta

ge (k

V)

M

L

D

R

Figure 4. Voltage induced in the MV conductors of the double-circuit line. Rg = 20 Ω.



Figure 5. Voltage induced in the MV conductors, in absence of HV line. Rg=20 Ω.

0 2 4 6 8 10 12 14

Time ( s)

0

50

100

150

200

250

300

350

Volta

ge (k

V)

M

L

D

R

0 2 4 6 8 10 12 14

Time ( s)

0

50

100

150

200

250

300

350

Vol

tage

(kV

)

M

L

D

R



Figure 4. Voltage induced in the MV conductors of the double-circuit line. Rg = 20 Ω.

Figure 5. Voltage induced in the MV conductors, in absence of HV line. Rg=20 Ω.

0 2 4 6 8 10 12 14

Time ( s)

0

50

100

150

200

250

300

350

Vol

tage

(kV

)

M

L

D

R

For both Fig. 4 and Fig. 5, the parameters for the DE model are those for the case in which the conductors are covered with XLPE.



Fig. 6 same as Fig. 5 but bare MV conductors. The voltage difference between conductors D and M is enough to cause a flashover. In case the HV line is present, for the same lightning return stroke, the phase to M voltage does not rise enough to cause a flashover, mainly due to the higher ground potential rise.

0 2 4 6 8 10 12 14

Time ( s)

0

50

100

150

200

250

300

350

Vol

tage

(kV

)

M

L

D

R

Figure 6. Voltage induced in the MV conductors, in

absence of HV line.Flashover in phase D. Rg=20 Ω.

Bare conductors.



0 200 400 600 800 1000 1200 1400

m

-50

0

50

100

150

200

250

300

Volta

ge (k

V)

5.5 us

L

D

R

M

Figure 7. Voltage peak amplitudes in the MV line, in absence of HV line, 5.5 μs after return stroke inception

0 200 400 600 800 1000 1200 1400

m

-50

0

50

100

150

200

250

300

Vol

tage

(kV

)

5.7667 us

L

D

R

M



The same situation is reported in Figure 8 but at t = 5.77 μs. The blue circle represents a flashover between D and M in the pole located in front of the return stroke location.

Figure 8. Voltage peak amplitudes in the MV line, in absence of HV line, 5.77 μs after return stroke inception. Blue circle: flashover in phase D.



In the simulations reported so far, in absence of further experimental data, we have assumed the same DE-model parameters for all the phases of the compact MV-line.

As shown in the previous figures, phase D is the one experiencing the larger potential difference with respect to M. However, due to the geometrical configuration of the spacer, it might be reasonable to expect that phase D is also the one with the highest withstand capability, since is the farthest from M.

It is therefore interesting to calculate the voltages in the MV-line conductors calculated by neglecting flashovers in phase D, as shown in Fig.9, next slide. The DE values of the lateral phases are the same reported in Table 1 but the values of V0 are set to 155 kV and 154 kV for phase R and L, respectively (in order to avoid simultaneous flashovers of both phases)



0 2 4 6 8 10 12 14

Time ( s)

0

50

100

150

200

250

300

350

Volta

ge (k

V)

M

L

D

R

Figure 9. Voltage induced in the MV conductors, in absence of HV line. Flashover in phase L. Rg=20 Ω. Bare conductors.



Geometry

DIRECT STROKESRg = 40 Ω Lightning channel impedance = 1 kΩ.

Channel-base current 31 kA, and 70 kA



Figure 10. Overvoltages in the 69-kV line due to a direct strike to OGW in the middle of the line: a) HV conductors; b) MV conductors

0 2 4 6 8 10 12 14

Time ( s)

-700

-600

-500

-400

-300

-200

-100

0

Vol

tage

(kV

)

OGW

Ph1

Ph2

Ph3

0 2 4 6 8 10 12 14

Time ( s)

-450

-400

-350

-300

-250

-200

-150

-100

-50

0

Vol

tage

(kV

)

M

L

D

R

We assume that the strike hits the messenger in the middle of the line. By comparing Fig. 9and Fig. 10 it can be seen that the MV line in the double circuit configuration is more protected also against direct strikes respect to the compact configuration alone.

31 kA 31 kA

0 200 400 600 800 1000 1200 1400

m

0

100

200

300

400

500

600

700

Volta

ge (k

V)

L

D

R

M

0 200 400 600 800 1000 1200 1400

m

0

100

200

300

400

500

600

700

Volta

ge (k

V)

L

D

R

M



70 kA70 kA

Figure 12. Maximum voltage peak amplitudes in the MV line due to a direct strike to OGW at the line center. Blue circles: flashovers phase D.

Figure 13. Maximum voltage peak amplitudes in the MV line due to a direct strike to OGW at

the line center. Blue circles: flashovers in D. SAs every 280 m.

The blue circles represent flashovers in phase D. The first flashover occurs at the pole struck by the lightning, which is also the one experiencing the larger ground potential rise. The voltage of the messenger, represented by the dashed thin line in Figure 11, decreases rapidly as the distance from the line center increases. At approximately 200 m from the line center the voltage potential difference between M and D is again large enough to cause flashovers in phase D of all the poles up to the line terminations

1. Introduction

2. Geometry of the Line and LIOV-EMTP Models and relevant assumptions



5. Conclusions





50 100 150 200 250 300

voltage (kV)

10 -2

10 -1

10 0

10 1

10 2

num

ber o

f eve

nts

havi

ng a

mpl

itude

larg

er th

an th

e ab

scis

sa/y

r/100

km

Total voltage - double-circuit line

Phase to ground - double-circuit line

Total voltage - MV line alone

Phase to ground - MV line alone

Comparison between the indirect lightning performance of the MV line in the double-circuit configuration (69-kV pole) and alone. Total voltage black Voltage respect to M red. Rg = 20 Ω, σg=1mS/m.



50 100 150 200 250 300

voltage (kV)

10 -2

10 -1

10 0

10 1

10 2

num

ber

of e

vent

s ha

ving

am

plitu

de

larg

er th

an th

e ab

scis

sa/y

r/10

0km

Total voltage - Unprotected

Phase to ground - Unprotected

Total voltage - SAs every 280m

Phase to ground - SAs every 280m

Indirect lightning performance of the MV line in the double-circuit configuration (69-kV) without SAs and with SAs (Rg = 20 Ω, σg=1mS/m).



0 100 200 300 400 500 600

voltage (kV)

10 -2

10 -1

10 0

10 1

10 2

num

ber o

f eve

nts

havi

ng a

mpl

itude

larg

er th

an th

e ab

scis

sa/y

r/100

km

Unprotected

SAs every 280m

SAs every 140m

Direct lightning performance (phase-to-ground voltages) of the MV insulators in the double circuit line configuration (69-kV pole) due to direct strikes with and without SAs. Rg = 20 Ω.

1. Introduction

2. Geometry of the Line and LIOV-EMTP Models and relevant assumptions



5. Conclusions





• The developed multi conductor line model based on the LIOV-EMTP-RV software, suitably adapted to the specific characteristics of the case under investigation, represents a powerful tool to study lightning-generated flashovers phenomena occurring at any of the line conductor-insulator combination;

• Although the number of direct strikes increases in the double-circuit configurations with respect to the MV line alone, the groundings of the overhead ground wire and of the underbuilt ground wire, when present, are overall quite effective in limiting the voltage stress of the MV line;

• The voltage induced in the overhead ground wire is larger than in the messenger, causing a significant ground potential rise that reduces the voltage across spacers and insulators.

• A preliminary deterministic investigation is necessary prior to any statistical study in order to be able to interpret the relevant obtained results.

Carlo Alberto Nucci – Full [email protected]

Alberto Borghetti – Associate [email protected]

Fabio Napolitano – Researcher (Assistant Professor)[email protected]

Fabio Tossani – Researcher (Assistant Professor)[email protected]

Dino Zanobetti – Emeritus Professor [email protected]

University of Bologna Research Group Composition at April ‘18

PhD students

Stefano Lilla [email protected]

Camillo [email protected]

Diego Rios [email protected]



mailto:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]

Слайд номер 1Слайд номер 2Слайд номер 3Слайд номер 4Слайд номер 5Слайд номер 6Слайд номер 7Слайд номер 8Слайд номер 9Слайд номер 10Слайд номер 11Слайд номер 12Слайд номер 13Слайд номер 14Слайд номер 15Слайд номер 16Слайд номер 17Слайд номер 18Слайд номер 19Слайд номер 20Слайд номер 21Слайд номер 22Слайд номер 23Слайд номер 24Слайд номер 25Слайд номер 26Слайд номер 27Слайд номер 28Слайд номер 29Слайд номер 30Слайд номер 31Слайд номер 32Слайд номер 33Слайд номер 34Слайд номер 35Слайд номер 36Слайд номер 37Слайд номер 38Слайд номер 39

Transient Response of a Double- Circuit Line to Direct and Indirect …en.rclp2018.com/upload/iblock/1b9/1b9df71235da5bf6d6053f... · 2018-04-26 · • In some areas, it is convenient

Documents