Proceedings of the XIVth International Symposium on High Voltage Engineering, Tsinghua University, Beijing, China, August 25-29, 2005

D-14

Sag Calculation of Aged Overhead Lines

M. Muhr, S. Pack, S. Jaufer Institute of High Voltage Engineering and System Management Graz University of Technology, Graz 8010, Austria E-mail: surname@hspt.tu-graz.ac.at Abstract Due to the changing operation of the transmission network in Europe caused by the liberalisation of the energy market condition based diagnostics get more importance. Based on the large expansion of the network and the changed stress of the line components a considerable effort for condition evaluation is necessary. The Austrian transmission line system at the level 230kV and 400kV has an important role for the energy exchange in the centre of Europe and was built up in 50s to the70s of the last century, therefore parts of the system are older than 40 years. Based on this situation this paper focuses on the operation behaviour of ageing overhead lines with special regard to the sag calculation of span sections. In general, the conductor elongates by the permanent mechanical forces (everyday stress) during its live time. Additionally to this irreversible elongation, the sag of a transmission line will increase by the conductor temperature caused by the electrical load and specific environmental conditions or other rope forces (e.g. ice load). Key WordsOverhead line, Aging, Sag, Calculation INTRODUCTION Overhead lines are the backbone of every electrical power transmission system. Without overhead lines, it is not possible to use power plants with high output because the consumers are not concentrated at the location of the generation. Therefore, transmission line systems like overhead lines or power cable are in use to connect the power plants and the consumer. To increase the reliability many power plants and consumer are connected together to an integrated network. In former days the regional operators worked together, the network is used as back up, and electric energy was generated nearby the consumer. By the liberalisation of the European energy marked this agreement will be soften more and more. Now the power flow is given by the trading at the spot marked. The transferred electrical power over a transmission line fluctuates depending on the nationwide power flow. In addition, the cost pressure is increasing, and the grid operators have to save investments to make their transmission lines more economical. Therefore, they use their lines near the nominal limits and the maintenance strategy has changed from a time based to a condition base strategy. One of the important parameters is the ground clearance of the ropes. Depending on the system voltage and the approached object, a safety margin has to maintain over the whole expected useful life. In general, the conductor elongates by the permanent mechanical forces (everyday stress) during its live time. Additionally to this irreversible elongation, the sag of a transmission line will increase by the conductor temperature caused by the electrical load and specific environmental conditions or other conductor forces (e.g. ice load). ELONGATION MECHANISM The sag curve of a conductor, which is spanned between to points, can be specified by the funicular curve. The sag is depending on the tension in the conductor and distance between the fixed points. The wire tension is always present and variable by the wire temperature and the additional load. This everyday stress leads to an irreversible elongation in addition to the elastic and thermal elongation. The reason of this elongation is the behaviour of used wire materials. In general high conduction aluminium is used as conductor and a high strength material as mechanical support. For the common type ACSR overhead lines, galvanize steel wires used as support. The composite of this two material with different material properties leads to an average behaviour of wire. By a mechanical stress of an ACSR conductor, the mechanical tension is not equal in aluminium and steel. The lower elasticity module causes a lower tension in the aluminium than in the steel. Based on this circumstance the load distribution is also different depending on the construction and the strength of used materials as shown in Table 1. Table 1: Load distribution of different conductor types
Al/St Al St 1:1 22.5% 77.5% 3:1 45.5% 53.5% Area ratio 4:1 6:1 53.7% 65.3% 46.3% 36.5% 11:1 76.1% 23.9%

In addition, the different thermal expansion of the used material leads to a thermal influenced changing of the load and tension ratio. Aluminium has a higher thermal expansion coefficient than the used steel. With higher temperatures, the aluminium is released and the steel core will get a higher mechanic load. At a certain temperature, the steel core will carry the whole load and the aluminium has no mechanic load anymore. This temperature is also called knee-temperature and depends on the mechanical structure of the ACSR conductor. In Figure 2 some examples are shown.

Proceedings of the XIVth International Symposium on High Voltage Engineering, Tsinghua University, Beijing, China, August 25-29, 2005

D-14

100% Steel 80% load ratio 60%

Al/St 340/110mm Al/St 685/85 mm

With these two types of conductors, two different models of overhead lines were calculated: 1. 2. flat terrain, symmetric rod distribution flat terrain, asymmetric rod distribution

knee temperature 40% 20% 0% -40 -20 0 20 40 in C 60 80 100 120 Aluminium

Case 1 is typical for a free flat terrain without rivers, roads or building. Case 2 is a realistically situation in flat terrain with hindrances. CASE STUDY 230KV SYSTEM In this case study, an overhead line with symmetric and asymmetric rod distribution is calculated in a flat terrain. The construction is like a typical Austrian overhead line of the 230kV network. The used conductor is an ACSR 340/110mm.
Section: AB1 AB2
30 T1 T2 T3 AB1 AB2 1500
9.5cm 1500 AB2

Fig. 1: Thermal behaviour of the load ratio AGING EFFECTS The low melting point of aluminium with a level of 660C leads to a creeping at the normal operating temperatures. This creeping is also temperature depending and will increase at higher temperatures. Several investigations were carried out to analyse the creeping of overhead lines. [1] [2] [3] The creeping depends on the mechanical structure, the conductor tension, the maximum load and the temperature. Typically, overhead lines are not equipped with a monitoring system and so for most existing lines conductor temperatures or maximum loads are not available. For a new line, the measurement effort for the monitoring is very high and expensive. It is easier to measure the sag and the conductor temperature in certain intervals to achieve the elongation of the overhead line conductor and the ground clearance. [4] The elongation of the conductor caused by thermal or aging effects has different effects on the sags of a line section. OVERHEAD LINE MODEL In Austria, the 230kV network is typical equipped with one conductor per phase and the 400kV network with two or four conductors per phase. The length between two rods is typical between 200 and 350m and depends on the geographic situation. The used model of the overhead line is based on a typical Austrian overhead line for a system voltage of 230kV and for 400kV. For the modelling two types of conductor were used as shown in Table 2. Table 2: Conductor data and isolator data
Al area St area Weight modulus of elasticity E-modulus Al E-modulus St coeff. of linear expansion Isolator type Weight (tension/suspension) Total length (tension/suspension) ACSR 340/110 341.2mm 108.8mm 1815kg/km 8400daN/mm 6000daN/mm 20700daN/mm 1.67 10-05K-1 2 long rods 130/60kg 3.5/2.85m ACSR 680/85 678.6mm 85.95mm 2564kg/km 6800daN/mm 6000daN/mm 20700daN/mm 1.94 10-05K-1 3 long rods 240/180kg 6.7/5m

25 height in m 20 15 10 5 0

500 length in m

1000

Fig. 2: Initial state at 10C conductor temperature In Figure 2, the initial state for the sag calculations are shown. At this temperature, the horizontal forces are in each span equal and so the suspension insulators (T1...T4) hang perpendicular. The tension insulators (AB1 and AB2) will be linked depending on the conductor forces and their own weight. When the conductor temperature increases caused by the load current, the sag is increasing too, as shown in Figure 3.
Section: AB1 AB2
30 T1 T2 T3 AB1 25 height in m 20 15 10 5 0 0 500 length in m 1000 T4

9.5cm

Fig. 3: Sag calculation at 80C without elongation During the temperature rise the horizontal forces in the conductor decrease (-16.2%). In addition, clamping point at the end of the tension insulator is lowered at 9.5cm. The system has now a inhomogeneous horizontal force

T4

Proceedings of the XIVth International Symposium on High Voltage Engineering, Tsinghua University, Beijing, China, August 25-29, 2005

D-14

distribution and the suspension insulators are linked but the deflections are marginal and has low effect. Only the deflection of the tension insulators has a significant effect on the sag. During the live time the conductors elongates by the everyday stress. In the case study, an elongation of 0.336 and 0.435 was assumed which will accords an age of 50 and 120 years.
T1 T2 T3 24 AB1 22 20 height in m 18 T4 AB2

With rising conductor temperatures, the sag will increase unequal as shown in Figure 6. The tension insulators show a deflection about 10cm as in the symmetrical rod distribution. At the other hand, all suspension insulates have a significant deflection to the span with the greatest distance between the mounting. With this deflection the sag the shot spans is reduced and in the longer spans increase.
Section: AB1 AB2
40 30 height in m 20 15 10 5 0.33m 0.47m 0 500 length in m 0.61m 0.69m 1000 1500 0 AB1 T1 25 T4 35 T2 T3 AB2 0.78m T4 16cm 1500 AB2

16 14 12 10 8 0 500 length in m 1000 1500

Fig. 4: Sag calculation at 80C with elongation In this case, the sag increases by the elongation about 0.470.6m in all spans. Figure 5 shows the initial case with asymmetric rod distribution with the same overhead line conductor. This model has a heavy asymmetrical rod distribution (minimum 200m, maximum 400m).
Section: AB1 AB2
40 30 height in m AB1 T1 25 20 15 10 5 0 0 500 length in m 1000 1500 T4 35 T2 T3 AB2

Fig. 7: Sag calculation at 80C with elongation In Figure 7, the same elongation is assumed as in the previous examples and the sag increases by elongation about 0.260.78cm asymmetric depending on the span. CASE STUDY 400KV SYSTEM In this example, a 400kV overhead line with an ACSR 685/85mm conductor is modelled. The spans are the same as in the previous case. The insulators in this case are heavier and longer.
Section: AB1 AB2
40 T1 T2 AB1 30 height in m 25 20 15 10 5 0 0 500 length in m 1000 T3 35 16cm

Fig. 5: Initial state at 10C conductor temperature

Section: AB1 AB2
40 30 height in m 25 20 15 10 5 0 0 500 length in m 1000 1500 0.51m 0.36m 0.18m 0.47m AB1 16cm T2 9.5cm T1 18cm T3 AB2 T4 35

13cm

Fig. 6: Sag calculation at 80C without elongation with and without suspension insulator deflection

Fig. 8: Initial state at 10C conductor temperature and sag calculation at 80C without elongation With the longer and heavier tension insulators, the deflection is higher than in the 230kV example. The suspension insulators are neglect able deflected. The sag increase by temperature is lower than in the previous case study (1.57m than 1.78m). The reason for this circumstance is the lower elastic modulus witch compensates the higher coefficient of linear expansion of the ACSR 685/85mm conductor. Based on the lower steel fraction, the elongation is higher than at the ACSR 340/110mm conductor and so an elongation of 0.331, 0.435 and 0.567 was

Proceedings of the XIVth International Symposium on High Voltage Engineering, Tsinghua University, Beijing, China, August 25-29, 2005

D-14

assumed which will accords an age of 20, 50 and 120 years. The sag increasing by elongation is in a region from 0.360.62m depending on the age.
Section: AB1 AB2
50 45 40 35 30 25 20 15 10 5 0 AB2 T4

CONCLUSION In a case study, two different overhead lines were used for a sag calculation and results the following conclusion. This calculations show, that the deflection of the tension insulators affects the sag behaviour of the close-by spans. Is the rod distribution of a flat overhead line symmetric, the deflection of the suspension insulators is negligible and has no significant effect on the sag calculation. The deflection of the suspension insulators is not negligible when the rod distribution is asymmetric. The span with the largest distance between the rods will sag much more than the smaller spans. The elongation of the overhead conductor reduces this safety distance. In this example, the sag will increase in magnitude of 26 to 78cm. With a reduction of conductor steady-state temperature about 26 to 30C can this effect compensated. A lower permissible conductor temperature affects also the transmission capability with will dropped. Especially in the hot summer time, the capability limitation may cause negative effects on the power quality. SUMMARY In this work, a sag calculation based on typical Austrian overhead lines of a system voltage 230kV and 400kV was carried out in a case study. The main point of the calculations is that the deflection of the suspension insulators affects the sag and the phase to earth clearance significant by asymmetrical rod distribution. With symmetrical distribution, only defection of the tension insulator affects the result. The elongation of the overhead conductors caused by the everyday stress leads to an increasing of the sag. During the planning and construction a lower sag than necessary, gives a safety distance to the regulation clearance. With sag verification of aged overhead lines by measurement or calculation, the reliability of the overhead line can be improved and depending on the system voltage and the object type, the minimum phase to earth clearance can be assured. REFERENCES [1] WG 05 Electra N24, A Practical Method of Conductor Creep Determination, page 105ff, Paris October 1972 [2] WG 22.05 Electra N75, Permanent Elongation of Conductors, Predictor Equation and Evaluation Methods, page 105ff, Paris March 1981 [3] F. Kieling, P. Nefzger, U. Kaintzyk; Freileitungen, Springer Berlin, ISBN 3-540-42255-2, 2001 [4] H. Pohlmann, R. Thomas; Sag increases resulting from conductor creep on medium-voltage transmission lines, and the problem of measuring sag on live overhead lines, Electricity Distribution, 1993 CIRED, 12th International Conference on 17-21 May 1993 page(s):3.20/1 - 3.20/5 vol.3

height in m

AB1

T1

T2

500 length in m

T3 1000

1500

Fig. 9: Initial state at 10C conductor temperature Figure 9 shows the case with asymmetric rod distribution. With rising the temperature, the sag is increasing asymmetric and the suspension insulators deflect as shown in Figure 10.
Section: AB1 AB2
50 45 40 35 30 25 20 10cm 15 10 5 0 AB2 AB2 0.76m 1500 T4

T3

height in m

T2

14cm

10cm

T1

AB1

17cm

19cm 0.37m

10cm

0.41m 0.32m 0 500 length in m

0.14m 1000 1500

Fig. 10: Sag calculation at 80C without elongation, with and without suspension insulator deflection The effect of the deflection of the suspension insulators is marginal low than at the ACSR 340/110mm but big enough to affect the sag calculation.
Section: AB1 AB2
50 45 40 35 30 25 20 15 10 5 0 T4 AB1 0.36m 0.50m 0 500 length in m T1 T2 0.62m T3 0.70m 1000

Fig. 11: Sag calculation at 80C with elongation Is the elongation of the overhead line included, the sag increases from 36cm to 76cm depending on the span dimensions.

height in m