An Investigation Into The Methodology To Develop An Insulator

Document Sample
An Investigation Into The Methodology To Develop An Insulator Powered By Docstoc
					> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <

1

An Investigation Into The Methodology To Develop An Insulator Pollution Severity Application Map For South Africa
D Pietersen* JP Holtzhausen* and W.L. Vosloo**

Abstract— The development of an Insulator Pollution Severity Application Map (IPSAM) for South Africa is being investigated. This paper discusses the results that were obtained from Directional Dust Deposit Gauges (DDG’s) and Equivalent Salt Deposit Density gauges (ESDD’s) across South Africa. An investigation into the effect of distance-from-seacoast on pollution severity levels is discussed. Correlations between DDG and ESDD and between DDG and corrosion test results are also included. Keywords—Distance-from-sea, pollution insulator pollution, corrosion. classification,

utilized in conjunction with the results from the 400 kV test site to investigate the effect of distance-from-sea on pollution severity levels. II. POLLUTION M EASUREMENTS

Various methods of obtaining a measure of insulator pollution severity are investigated. Two direct methods of pollution measurements were utilised: directional dust deposit gauges (DDG) and equivalent salt deposit density (ESDD). Metal corrosion depends on the presence of moisture and ionic salts, the same factors causing insulator pollution flashover. It was therefore decided also to install standard corrosion measurement probes at each measurement point. Direct insulator pollution measurements These methods collect the pollution material directly and the collected samples are evaluated monthly intervals. DDG (Directional dust deposit gauges) The DDG consists of four vertical collection tubes with slots milled in the sides to collect air-borne pollutants. The four slots face the four cardinal points of the compass: North, East, South, and West as shown in figure 1. The DDG pollution index is defined as the mean value of the conductivities of all four directional gauges, expressed in µS/cm and arithmetically normalised to a 30-day month. The equations to calculate normalised conductivity and average (mean) conductivity is given below: σ = C (V/500) (30/N) N where σN C V N = = = = normalised conductivity (µS/cm) volume conductivity (µS/cm) volume of distilled water (ml) amount of days since previous removal of the containers Average σ= (σ + σ + σ + σ )/4 N S E W (2) (1)

I.

INTRODUCTION

P

ower lines pass through the following major pollution environments: marine, industrial, desert, and agricultural. These pollution deposits accumulate on insulator surfaces and form a conductive electrolyte when the insulator surface is wetted by rain or fog. This allows the flow and increase of leakage currents over the insulator surface, which in turn results in the decrease of the electrical withstand of the insulator and finally leads to a flashover. Although the ultimate aim of this project is to produce a map for South Africa, the present study includes data from the Western Cape and South Cape, which are used as the basis to develop a methodology to compile a map. Pollution measurements are taken at various intersections along a 400 kV power line up to 35 km away from the coast. The aim is to assign pollution severity classification levels along the line. Pollution measurements from other pollution monitoring stations along the coast, up to 35 km’s away, are
Manuscript received April 23, 2004. This work was supported in part by ESKOM and the University of stellenbosch. *D. Pietersen is a consulting engineer and is busy with an MSc Engineering Sciences degree at the University of Stellenbosch. (phone: +27 982 7714, email: holtzhau@sun.ac.za). *J.P. Holtzhausen is a senior lecturer with the Univ ersity of Stellenosch and holds a PhD. degree. (email: holtzhau@sun.ac.za). **W.L. Vosloo is Chief Consultant on insulators for Eskom TSI and he holds PhD. degree. (email: wallacevosloo@sun.ac.za).

> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < ESDD (Equivalent salt deposit density) A unit string, consisting of seven glass disks is used for the ESDD measurements, as shown in figure 2. These discs are not energised and collect air-borne pollutants under the influence of the environmental conditions, the shape and material of the insulator. The discs are removed and the pollution on the surface is washed, using demineralised water, as indicated in figure 2. The ESDD pollution index is obtained from the measurements of the volume conductivity, temperature, and volume of the wash water. The conductivity probe measures the volume conductivity, σt, at temperature t. The volume conductivity is related to 20 °C using equation 3:

2

Metal coupons The metal coupon test was based on the principles of the ASTM G50 [10] and ISO 9226 [12] standards. The corrosion rate, rcorr, in mirometers per year (µm/a) is given as [12]: rcorr = ∆ / (A⋅d⋅ t) m (16)

where ∆ = is the mass loss (mg) m A = is the surface area (m2) t = is the exposure time in years d = is the density (g/cm3) where d Fe = 7.86 (g/cm3), d Zn = 7.14 (g/cm3), d Cu = 8.96 (g/cm3), d Al = 2.70 (g/cm3).

σ20 = σt [1 – 0.02277(t –20) e -0.01956(t-20) ]

(3)

Where σ is the measured volume conductivity (µS/cm), t is the t solution temperature (°C) and σ is the volume conductivity 20 corrected to 20 °C. The salinity, Sa (kg/m3), of the suspension at 20°C is obtained with equation 4 and the equivalent salt deposit density (ESDD) in mg/cm2 is then determined with equation 5: Sa = (5.7 * σ20)1.03 ESDD = Sa V / A (4) Figure 1: Directional dust deposit gauges (5) Dummy disk 7 Analysed every two years Analysed every year Analysed every six months Analysed every three months Analysed every month Dummy disk 1 CLIMAT For the CLIMAT test; preparation, testing and analysis were done as prescribed by ASTM G116 and ISO 9226 [9], [12]. In principle, this test method is a galvanic corrosion (or dissimilar metal) effect. The CLIMAT ( CLassify Industrial and Marine ATmosphere) test is also referred to as the wire-on-bolt test. The ISO equation for the corrosion rate, rcorr in mirometers per year (µm/a) is given as [12]: rcorr = 0.25 x ∆ m⋅D / (m⋅ t) where ∆ = m D = m = t = is the mass loss (mg) is the wire diameter (mm) is the original mass (grams) is the exposure time (in years) (6)

Where V is the volume of distilled water (cm3) used and A is the area of the washed insulator (cm2). Table 1 is a summary of the DDG and ESDD pollution indices as given in [8]. Corrosion measurements Two methods of corrosion testing were incorporated in the study, CLIMAT and metal coupons.

Figure 2: ESDD unit string CLIMAT Mild steel

Figure 3: CLIMAT and Metal coupons

> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < Figure 3 is a photograph of an actual completed metal coupon assembly installed at a test site 15 km from the coast while Table 2 provides the corrosion indices and classification for the metal coupon test [13]. Table 1: DDG and ESDD pollution indices DDG ESDD Severity monthly monthly monthly Class average maximum maximum (µS/cm) (mg/cm2) 0 to 75 0 to 175 < 0.06 Light 76 to 200 176 to 500 0.06 – 0.12 Medium 201 to 350 501 to 850 > 0.12 - 0.24 Heavy > 350 > 850 > 0.24 Very Heavy Table 2: Corrosion indices Metal coupons Corrosivity Mild steel Zinc (category) (µm/a) < 1.3 < 0.1 Very low (C1) 1.4 to 25 0.2 to 0.7 Low (C2) 25 to 50 0.8 to 2.1 Medium (C3) 51 to 80 2.2 to 4.2 High (C4) 81 to 200 4.2 to 8.4 Very High (C5)

3

and figure 11 show corrosion rates against the distance-fromcoast along the Koeberg-Muldersvlei 400 kV line for CLIMAT (wire-on-bolt) and metal coupon corrosion rates respectively. Figure 8 (average values) and 9 (maximum values) are plots of the ESDD results.

Figure 4: Map of South Africa indicating pollution measurement areas

III.

LOCATION OF MONITORING STATIONS

Figure 4 is a topographical map showing the location of monitoring stations within South Africa; the areas are the Cape Peninsula, West coast and South Cape Figure 5 is a topographical map of the monitoring area from the Koeberg Nuclear Power station to Muldersvlei substation, spanning a dis tance of 30 km from the coast. This site was set up to investigate the effect of aerosol migration from the coast and the impact on DDG and corrosion levels along the Koeberg-Muldersvlei 400kV line. Directional dust deposit gauges and corrosion specimens were installed at the KIPTS (Koeberg Insulator Pollution Test Station) monitoring station, and at every 1 km interval for up to 10 km’s, at 15 km, 20 km and 30 km from the coast. The position of each monitoring station was determined using GPS (global positioning system) mapping technology. In the analysis of the distance-to-coast effect, the results from the Koeberg-Muldersvlei line (figure 5) were combined with data from other pollution monitoring stations in the West and South Cape (figure 4).

IV.

RESULTS AND DISCUSSION

The DDG pollution measurement results from all the monitoring stations up to 35 km’s from the seacoast, including data from the Koeberg-Muldersvlei 400 kV test area, are shown in figure 6 and 7 for average and maximum values respectively. Figure 10

Koeberg NPS Muldersvlei substation DDG positions Substation (monitoring)

Figure 5: Topographical map of the monitoring area along the Koeberg-Muldersvlei 400 kV line

> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
900

4

12

800

10
700

Al-Fe Al-Cu Power (Al-Fe) Power(Al-Cu)

Average Nornmalised Conductivity(uS/cm)

600

8
CLIMAT Index (%)

500

6

400

300

4

200

2
100

0 0 0.8 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

0
0.04 0.8 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Distance from coast (km)

Distance from the coast (km)

Figure 6: DDG measurement results (average values)
60 3200 3000 2800 2600 2400 2200 Average normalised Conductivity (uS/cm) 2000 1800 1600 1400 1200 1000 800 600 400 200 0 0 0.8 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Distance from coast (km) 0 10 Corrosion Rate (um/a) 40 50

Figure 10: CLIMAT corrosion testing results
Mild Fe Power (Mild Fe)

30

20

0.04 0.8 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Distance from the coast (km)

Figure 11: Mild steel corrosion rates
0.12

Figure 7: DDG measurement results (maximum values)
0.10
0.12

0.08
0.1

ESDD (mg/cm)

M
0.06

0.08

ESDD (mg/cm2)

L
0.04

0.06

0.02
0.04

L
0.00
0.02

M
200

H

VH
400 600 800 1000 1200 1400
DDG conductivity (uS/cm)

0

0
0 0.8 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Distance from coast (km)

Figure 12: Correlation between ESDD and DDG measurement results (average values)
0.30

Figure 8: ESDD results (average values)
0.28 0.26 0.24 0.22 0.2 0.18

VH
0.25

0.20 ESDD (mg/cm2)

0.15

ESDD (mg/cm2)

0.16 0.14 0.12 0.1 0.08 0.06 0.04

H M L

0.10

0.05

0.00
0.02 0 0 0.8 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

0

500

1000

1500

2000

2500

3000

3500

DDG conductivity (uS/cm) Distance from coast (km)

Figure 9: ESDD results (maximum values)

Figure 13: Correlation between ESDD and DDG measurement results (maximum values)

> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
7

5

60

6

50
5

Corrosion rate - mild Fe (um/a))

40

Climat Index (%)

4

30

3

20

2

10
1

0
0 0 50 100 150 200 250 300 350 400 450 500

0

200

400

600

800

1000

DDg conductivity - Average values (uS/cm)

DDG conductivity (uS/cm)

Figure 14: Correlation between DDG and CLIMAT (AL-FE), average values
60 7

Figure 16: Correlation between DDG and Mild steel (average values)

6 corrosion rate, Mild Fe (um/a)

50

5 CLIMAT Index (%) )

40

4

30

3

20

2 10 1

0 0 200 400 600 DDG conductivity (uS/cm) 800 1000 1200

0 0 500 1000 1500 2000 2500 3000 3500
DDG conductivity, maximum values (uS/cm)

Figure 15: Correlation between DDG and CLIMAT (AL-FE), maximum values Table 4: Correlation parameters Description DDG vs ESDD avg max avg DDG vs mild Fe max avg DDG vs CLIMAT max Mild Fe vs CLIMAT (Al-Fe) CLIMAT vs CLIMAT (Al-Fe vs Al-Cu)

Figure 17: Correlation between DDG and Mild steel, maximum values

R2 0.617322 -0.608362 0.7558815 0.748623 -0.111266 -0.132 -0.446458 0.589449

Standard error 0.0147 0.08419 7.07787 7.182327 1.9617 1.979958 6.0517 3.150925

Slope, b 0.0001099 0.000118 0.060688 0.01489 0.013788 0.006093 2.8534 1.518899

P -value 9.016 x 10 -5 4.64 x 10-5 2.59 x 10-6 3.05 x 10-6 0.000431 0.000479 0.000581 0.000493

King et al reported salt deposition close to the coast (within 103 m) is highly influenced by salt production at surf beaches and by local terrain, whilst salt at 103 m to 106 m inland is more influenced by transport of the finer aerosol produced by ocean white caps [3]. Comparing pollution levels (figures 6 to 9) close to the coast for up to 1 km, it is clear that the order of magnitude is twice (or even 3 times) that of pollution levels further inland, which agrees with findings from various distance-from-seacoast corrosion studies [3-7].

Impact distance The pollutants at KIPTS were found to be predominantly sodium chloride (NaCl) [2], with the sea as the source of the salt-laden aerosol and affecting areas further inland. Table 3 is a summary of the impact distance for each pollution level based on the distance-from-seacoast analysis. It should be noted that pollution levels could penetrate further than indicated; however, the distances are reflective of the actual measurement results.

> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < Table 3: Impact distance-from-seacoast Maximum distance of impact (km) Description Very Heavy Medium Light Heavy max 3 18 30 >30 DDG Avg 5 18 30 >30 max 0.08 23 30 > 30 ESDD avg 0.08 > 0.08 Mild steel 0.08 > 30 Correlations Figures 12 to 17 are plots from regression studies for various monitoring parameters with the regression for all forced through the origin. Table 4 is a summary of results from regression analyses. A hypothesis test (t -test) was carried out to determine the usefulness of the relationship. A P-value greater than 0.01 (P>0.01) indicate that there is no useful linear relationship between the modeled parameters. To the author’s knowledge no other study have been carried out comparing direct insulator pollution measurement results with corrosion testing results. It should be noted that the data point appearing as an outlier is not an outlier in the sense of the word. It is ascribed to the location of monitoring stations leaving a gap in the data points between the coastal station (KIPTS) at 40 m from the shore and the next station located 800 m away, producing results in the order of 2 to 3 times in magnitude at KIPTS compared to the next station 800 m away. It would be a good practice to have more monitoring stations over the first f hundred meters ew from the coast in order to capture data points that are more spread out over the range of the measurement values since the decay is more dominant in this region (see figures 6 to 11) and it becomes more linear after about 800 m from the coast. Climatic conditions Relative humidity (RH) is an indication of the moisture level in the atmosphere. When the RH percentage is high (>75 %) there is a good chance that the insulator surface could be wetted, which could dissolve into a conductive electrolytic layer, which further result in increased leakage currents [2]. Vosloo conducted a study at the KIPTS (Koeberg Insulator Pollution Test Station) monitoring site and found that there is a 50% probability that RH will be higher than 94% in winter and 85% in summer. The probability that the RH levels will exceed 75% is approximately 88% in winter and 78% in summer. Corrosion rates are also significantly increased at RH levels above 75%, which also explains why the distance-fromseacoast pollution profiles are similar for the insulator and corrosion pollution measurement results (see figures 6 to 11). V. CONCLUSION

6

exceptional pollution severity over the first 500 to 800 meters from the coast; thereafter a rapid decay is noticeable moving further inland. The DDG and ESDD pollution results are very similar in profile compared to that of the corrosion tests (CLIMAT and metal coupons). The first few hundred meters are exceptional and a rapid decay is noted after about 1 km (103 m) from the coast. A very strong relationship exists between DDG pollution levels, distance from the coast and mild steel corrosivity with 2 a n R value of 0.75, while a moderate linear relationship between DDG and ESDD pollution values with an R2 value of 0.62 have been produced. Corrosivity and insulator pollution levels at the coast (KIPTS) are 2 times those at sites 800 m from the coast, -3 reflecting the exceptional pollution concentration over the first few hundred meters (up to 103 m) from the seacoast, which is mainly attributed to the high NaCl concentration levels and high relative humidity in the region closer to the coast. It can be concluded that a corrosion map can be used to predict insulator pollution severity levels in the case where direct insulator pollution measurements are not available.

A CKNOWLEDGMENT The author would like to thank Eskom for the support. REFERENCES
[1] D. Pietersen, W.L. Vosloo and J.P. Holtzhausen: “An investigation into the measurement techniques to compile an Insulator Pollution Severity ApplicationMap for South Africa”, 6th IEEE AFRICON Conference, 2002 IEEE Transactions,, South Africa, October 2002, pp. 585 - 591. W.L. Vosloo : “A comparison of the performance of high -voltage insulator materials in a severely polluted coastal environment”, Ph. D Dissertation, pp. 52-70, March 2002. G. A. King, W. D. Ganter and I. S. Cole : “Studies at Sites Progressively Inland from the Coast to aid Development of a Geographic Information System Map of Australian Corrosivity” Journal of Corrosion and Prevention, 1999, pp. 1-12. P.W. Haberecht, C.D. Kane and S. J. Meyer: “Environmental Corrosivity in New Zealand: Results after 10 Years of Exposure”, Conference Paper No 70, 14th International Corrosio n Congress, September 26 – October 1, Cape Town (South Africa), 1999, pp. 2-12 S. Feliu, M. Morcillo and B. Chico : “Effect of distance from sea on Atmospheric corrosion rate” Journal on Corosion, Vol.55 No.9, pp 883-891, 1999 D. Frumuseli and C. Radu : “Impact of a Polluting environment on Overhead Power Delivery Systems” Journal of Material Selection and Design, August 1999, pp. 62-66. P.W. Haberecht : “Atmospheric corrosion in New Zealand: corrosivity modelling and measurements” Corrosion and Prevention Journal, Auckland, New Zealand, 2000, pp. 1-14.

[2]

[3]

[4]

[5]

[6]

[7]

The results from the distance-from-coast study show an

> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
[8] W.L. Vosloo, R.E. Macey, C. deToureil, “The Practical Guide to Outdoor High Voltage Insulators”, Book to be published by Eskom, 2004. [9] ASTM Standard, G116 – 93, “Conducting Wire-on-Bolt Test for Atmospheric Galvanic Corrosion”, Annual Book of ASTM Standards, 1999. [10] ASTM Standard, G50 – 76, “Conducting Atmospheric Corrosion Tests on Metals”, Annual Book of ASTM Standards, 1999. [11] ASTM Standard, G1 – 90, “Standard Practice for Preparing, Cleaning, and Evaluating Corrosion Test Specimens”, Annual Book of ASTM Standards, 1999. [12] International Organization for Standardisation, ISO 9226, “Determination of Corrosion Rate of standard specimens for the evaluation of Corrosivity - Corrosion of Metals and Alloys – Corrosivity of Atmospheres”, 1992 . [13] International Organization for Standardisation, ISO 9223, “Classification: Corrosion of Metals and Alloys – Corrosivity of Atmospheres”, 1992.

7


				
DOCUMENT INFO
Shared By:
Stats:
views:50
posted:12/16/2009
language:English
pages:7
Description: An Investigation Into The Methodology To Develop An Insulator