Three Techniques to Mitigate Lightning

Mar 1, 2008 12:00 PM
By Ignaz Hübl, Michael Marketz and Robert Schmaranz, KELAG Netz GmbH, and Stephan Pack and Mich

Austria's 110-kV Distribution System is a Key Component of the Distribution Network. A substantial proportion of this overhead-line system is located in the high alpine area. In this rocky terrain, the local grounding conditions are often poor, with values ranging from 200 Ω to 7000 Ω. Additionally, this area often experiences thunderstorms resulting in a high number of lightning strikes to ground. Therefore, the local weather conditions, particularly the lightning strikes, adversely affect the reliability of the overhead-line circuits.

As a result of the poor circuit reliability of the 110-kV system, Austrian supply company KELAG began to record and analyze circuit outages in 1995. It was apparent that the transient voltage dips were resulting in a negative impact on the supplies to several specialist industrial companies. The monitoring of multiphase faults revealed the fault incidence due to lightning on one particular 110-kV overhead distribution line was abnormally high, prompting the need for further investigation.

LINE PARAMETERS

The circuit identified for investigation was a 110-kV overhead distribution line erected across the Kreuzeckgruppe mountain range that links the Oberdrauburg substation to the Auβerfragant substation (Fig. 1). This double-circuit line has an over-running ground wire. The line is 36 km (22 mile) long, has 108 steel towers and crosses a mountain range that rises to 2300 m (7546 ft) above sea level. Figure 2 shows KELAG's 110-kV distribution network.

The 110-kV circuits from Tower No. 2 to Tower No. 56 are operated in parallel for the 15-km (9.3-mile) section of the mountain range that rises from 640 m (2100 ft) to 2320 m (7612 ft) above the timber line in a high alpine area with rocky soil. The remaining sections of these lines operate independently, and the span of Tower No. 59 to Tower No. 60 has no ground conductor because of an overhead 220-kV line crossing.

The tower heights vary from 30 m (98 ft) to 55 m (180 ft) and have concrete foundations and a grounding system. Span lengths vary from 120 m (394 ft) to 420 m (1378 ft). The insulators are dimensioned accordingly for the 110-kV system and arcing horns are installed on all isolator strings.

In the area where the 110-kV distribution line is located, the number of atmospheric discharges to the line conductors and into the surrounding environment is higher than the average value in Austria, according to data recorded by Austrian Lightning Detection & Information System (ALDIS; Vienna, Austria). The average location accuracy for all detected lightning strokes in Austria recorded by ALDIS is less than 1 km (0.6 miles).

The lightning density for the region where the overhead line is erected ranges from three to more than six lightning strikes per square kilometer per annum. Although these values vary annually, they are always higher than the average values for the region. Figure 3 shows the lightning density in the area of the overhead line for 2006.

Nearly one-third of the 110-kV overhead-line route is in a high alpine region that has rocky soil and bad tower grounding conditions, with values of the footing resistances in the range of 200 Ω to 1200 Ω (Fig. 4). The number of line outages for KELAG's 110-kV distribution network is, on average, lower than two outages per annum. However, one 110-kV overhead-line circuit has 49 outages in the five-year time frame, some 5 to 10 times higher than the average system circuit outage values (Fig. 5).

AN ANALYTICAL EVALUATION

An analytical process was undertaken to evaluate the relevant parameters to develop a practical solution.

Firstly, the footing resistance was evaluated. This specific high alpine distribution line has high footing resistances. These were determined in 2001, when the footing resistances at five towers were measured with and without the ground wire connected on the tower top. Values between 100 Ω and 1200 Ω were measured.

Secondly, the shielding angle was analyzed. The effectiveness of the shielding area of the ground wire was studied for the defined line section. The 110-kV overhead line crosses a mountain range, so the line is located on a hillside. The maximum lightning currents, which can probably hit the phase wires directly, were determined using the geometrical-electrical model. By determining the currents this way, the hillside situation was comparable with the lightning flash density experienced in this region. For the most exposed towers, the maximum currents for a direct strike to the phase conductors were evaluated and are shown in the table.

Thirdly, the line arrester location was evaluated. To identify the locations for surge arresters along the overhead line, the alternative transient program (ATP) was used. Two models were used: one to represent the steady-state conditions for overhead line, grounding conditions and protection devices, and the second for the transient source, which represents the lightning discharge.

For this high alpine region, the grounding of each tower was modeled individually. A multiplicity of computations were performed to evaluate the performance of this line section, and the applications of surge arresters were varied based on the number of protected phases and the number of protected towers.

MEASURES TO IMPROVE PERFORMANCE

To improve the line performance and to decrease the line outage rate, KELAG applied the following practical measures to the 110-kV line in accordance with findings from theoretical studies by Graz University of Technology (Graz, Austria):

System rearrangement
The double-circuit, three-phase 110-kV overhead line was converted into a single circuit with one conductor in reserve. The two-top phase conductors were connected to each steel tower and, as a result of these connections, became two additional ground wires increasing the shielding angle for the single circuit line.
Surge arrester application
In accordance with the transient calculations for the application of the surge arresters along the overhead line, 18 surge arresters were installed along the overhead line and six arresters were installed in the Oberdrauburg and Auβerfragant substations. All three phases of the substations were equipped with surge arresters. In the selected line section (9 towers) only two of the three phases were equipped with surge arresters (Fig. 6).
Improvement in the grounding situation
To improve the grounding situation, the old ground wire was replaced with a ground wire that had an integrated fiber-optic conductor. In view of the high contact resistance between the ground wire and the tower, an additional shunt wire between the ground wire and the tower was installed. The ground wire between Towers No. 59 and No. 60, the span with a 220-kV system crossing, also was installed.
Arrester discharge records
To register and record the discharge behaviour of the installed surge arresters, a Rogowski inductor and an arrester discharge logger (ADL) were installed at every arrester location.

FIELD EXPERIENCE

As a result of the scientific cooperation between KELAG and Graz University of Technology, a correlation between the registered outages and registered ADL records was produced, supported by ALDIS's lightning activity data:

Lightning activity
ALDIS observed the lightning activity in the area close to the overhead-line route from 2004 to 2006. The observation corridor was 3 km (1.9 miles) wide, or 1.5 km (0.9 miles) wide on each side of the 30-km (19-mile) line section. ALDIS detected a total of 799 lightning strikes and a total of 1463 flashes. The average value of the amplitude of the detected strikes was approximately 15 kA. As shown in Fig. 7, most of the lightning strikes were of negative polarity, with 90% to 95% of all lightning strikes having negative polarity in the first two years of record. This confirms established data on polarity distribution. In 2006, an uncommonly high 31% of strikes were of positive polarity discharges.
Outages
Based on the utility's internal event recording system, the number of two- and three-phase faults and phase-to-ground faults for the observed line are shown in Fig. 8. Eight faults were registered that involved two or three phases. Importantly, none of these multiple-phase faults were within the rearranged line section. Furthermore, a total number of eight transient phase-to-ground faults could be correlated with lightning activity in the line corridor. Therefore, the prevailing lightning activity led to ground faults within the rearranged line section. Outside the rearranged line section, the number of multiple-phase faults was comparable with the failure rate during the 1995 to 2000 period.

PRACTICAL MEASURES

Due to an abnormally high outage rate on a double-circuit 110-kV overhead line, a scientific evaluation project was established to determine ways to increase the performance of this circuit. A number of important tasks, like the geographic situation of the line, the tower footing resistances, the shielding angle analyses, the lightning activity and the numerical calculations for insulation coordination, were undertaken.

As a result, the line was converted to single-circuit operation with the original ground wire and the two upper-phase conductors that had been grounded, a design change that significantly improved the grounding conditions and surge arresters that had been installed on the line section most vulnerable to lightning activity. Three years of operational experience has shown that the theoretical studies performed and the practical measures taken have improved the system reliability of this circuit by significantly reducing lightning-caused outages.

Ignaz Hübl graduated from Graz University of Technology in 1986 and spent four years with Siemens (Berlin, Germany) developing and testing high-voltage switchgear. Currently, he is head of the Power System Operation department at KELAG Netz GmbH. ignaz.huebl@kelagnetz.at

Michael Marketz studied electrical power engineering at Graz University of Technology before joining KELAG in 1999 in the department of Power System Operation. In 2003, he received his Ph.D., and he currently is head of the department of Power Plant Operation and Maintenance. michael.marketz@kelag.at

Robert Schmaranz graduated from Graz University of Technology in 2001 and, following three years at the Institute for Electrical Power Systems, he received his Ph.D. in 2004. Schmaranz is now working in KELAG's department of Power System Operation, where his specialist interests include power system protection and operation. robert.schmaranz@kelagnetz.at

Stephan Pack studied electrical power engineering at Graz University of Technology, where he earned his Ph.D. Since 1985 he has worked with the Institute of High Voltage Engineering and System Management and has been involved in the university's accredited Test Laboratory. Pack is the Austrian delegate at CIGRÉ for System Technical Performance and is chairman of the Austrian Technical Committee for Lightning. pack@tugraz.at

Michael Muhr studied electrical engineering at Graz University of Technology, where he earned his Ph.D., before joining the Institute of High Voltage Engineering and System Management in 1971. He was appointed head of the Institute and managing director of the Test Institution of High Voltage Engineering, a position he has held since 1990. In addition to publishing more than 150 Papers, Muhr serves as the Austrian delegate on CIGRÉ Study Committee D1 “Materials and Emerging Technologies” and on two IEC committees. muhr@tugraz.at