CIGRE Composite Insulators

21, rue d’Artois, F-75008 PARIS http : //www.cigre.org

B2-214

CIGRE 2012

ASSESSMENT OF THE CONDITION OF OVERHEAD LINE COMPOSITE INSULATORS

F. SCHMUCK Pfisterer Sefag Switzerland

J. SEIFERT Lapp Insulators Germany

I. GUTMAN STRI Sweden

A. PIGINI Consultant Italy

SUMMARY Composite insulators have become a highly developed alternative to conventional insulators in all transmission level voltage classes for AC and DC applications around the world. However, an important issue needs to be addressed: the difficulty in which to assess their conditions in service, especially before the application of live line working (LLW) techniques for the replacement or repair of insulators, insulator strings, dampers or the conductor. Despite broad experiences and advanced diagnostic tools are presently available, LLW on overhead lines with composite insulators has not yet been introduced in the present standards and international standardization committees work on this matter. This paper presents the current status of the work in CIGRE Working Group B2.21, which is participating in this topic as well. It can be shown that the diagnostic tools, first introduced more than fifteen years ago, have been further developed and if correctly used, provide sufficient information to evaluate the risk of the insulators subjected to LLW. Examples of how LLW is used in various countries are presented.

KEYWORDS Composite Insulators, Diagnostics, Life Estimation, Live Line Work, Corona [email protected]

1

STATUS QUO OF COMPOSITE INSULATOR USAGE

While the first application of composite insulators for the distribution network was seen in the nineteen seventies, the first industrial application of transmission class composite insulators began in the nineteen eighties. In the beginning, composite insulators were rather expensive and their application was limited to special areas, e.g. for areas requiring pollution performance above that of conventional insulators or areas affected by vandalism. However, with the definition of the “minimum requirements” for the composite insulator design in 1988 by CIGRE Working Group (WG) 22.10 [1] (the base for the first IEC 1109 edition in 1992 [2]), significant technological developments were triggered. At present, composite insulators are often more competitive in comparison to the conventional glass and porcelain insulators, because of factors like economies of scale in production. Many advantages of composite insulators have led to their wider use in transmission lines. New opportunities leading to an even wider use of composite insulators are as follows: 





With the introduction of UHV lines such as already existing 1000 kV AC or 800 kV DC and even discussed 1200 kV AC, considerable cost savings are possible when light and relatively short (in comparison to porcelain longrods) composite insulator strings are used. Insulators with lengths of 10 m and longer can be currently manufactured in one piece, as well as providing mechanical ratings up to 1000 kN and more. Due to the shift towards “green energy generation” in many countries, a voltage upgrade of existing transmission lines or voltage conversion of the existing transmission lines from AC to DC are the most practical ways to comply with the variations in energy flow. Depending on how much of additional power is required to be transported using the same towers and in some cases the same conductors, such refurbishment generally requires the use of composite insulators. This is also valid for AC/DC conversion, where composite insulators may provide the preferred solution due to their advantages in terms of pollution performance [3], [4]. Furthermore, the use of compact lines with aesthetically pleasing designs definitely requires the thin silhouette of composite insulators [5].

To measure the composite insulator penetration into the market and the insulator reliability, CIGRE WG B2.21 is currently working on a questionnaire, which will be sent to utilities worldwide. Their responses will form the basis for a further survey which will update the previous questionnaires from 1990 [6] and 2000 [7]. Today, based on preliminary and sparse information, only an estimation of the growth for AC can be made as shown in Fig. 1 [8], indicating an exponential trend.   s 25   n   o    i    l    l    i 20    M

Installed Composite Line Insulators ≥ 69 kV AC

15 10 5 0 1975

1980

1985

1990

1995

2000

2005

2010

2015 Year

Fig. 1: Estimated growth in composite insulator use in AC-transmission lines  With regard to DC transmission lines, according to another survey by CIGRE WG 22.03 in 1995 [9], the initial application of composite insulators started in the early nineteen eighties as well. More than 11`000 composite insulators are presently installed in Chinese ± 500 kV 1

and ± 800 kV transmission lines. Extrapolating the limited information available, the number of composite insulators installed in DC lines worldwide may be of the order of 100`000 [8]. The number is much lower than for AC, reflecting the current ratio between DC and AC transmission lines. However, as with AC, the increase in the use of composite line insulators for DC lines seems to be exponential too. As a matter of fact, the advantages of composite insulators in DC are even more evident than in AC, because of the superior performance in comparison to ceramic insulators under pollution, which is a dimensioning factor for DC overhead lines [10], [11]. As far as composite insulator reliability is concerned, an analysis by EPRI on the failures reported in the USA [12] is shown in Fig. 2. With an estimated number of 3 million insulators installed at that time, 315 failures were reported with a cumulative failure rate of 0.0105 %. The failures by brittle fracture and flashunder (i.e. by full or partial interface breakdown between the housing and the rod) can also be related to other primary root causes such as housing damage and rod exposure. As mentioned, the failure rate is a cumulative one and a large contribution to it is from the failures of the first generation of composite insulators with their well known weaknesses. As an example, the high number of brittle fracture failures is related to the use of rods with E-glass impregnated with polyester resin in the first insulator generations, which were widely installed in the USA network. With the progress in technology, electrochemically resistant glasses, such as ECR or E-CR, have become the standard for long rod insulator applications, thus limiting the brittle fracture risk as pointed out in the recently published IEC 62662 [13] and in [14]. The today’s generation of composite insulators is characterized by material components and manufacturing processes, which are proven by long-term outdoor experience.    5    1    3   y    l    l   a    t   o    t    f   o   n   o    i    t   u    b    i   r    t   s    i    D   e   r   u    l    i   a    F

50% 45%

EPRI Report on Failures in USA

40% 35% 30% 25% 20% 15% 10% 5% 0% Brittle Interface Discharge Mechanical Flashover Loss of End Fracture (Flashunder) Rod Failure Failure Fitting by Pullout

Fig. 2: Failures of composite insulators (EPRI survey) 

Another survey [15] describes the failure records of composite insulators in China, where impressive electrification programmes have been launched and where composite insulators have now become the prevailing insulator type. 58 failures are reported in the survey in relation to a number of installed composite insulators of 2.2 million by the end of 2006. Comparing the data of Fig. 3 with the statistics from the USA (Fig. 2), the range and causes of failures are quite similar. The Chinese cumulative failure rate is only 0.0026 %, which is lower than that reported from the USA. However, it should be noted that in the report from the USA, the first generation of composite insulators is included, which is recognized to have a higher failure rate because of the stage of product development and standard availability during the time of their installation. 2

100

   %   n    i   n   o    i    t   u    b    i   r    t   s    i    D

90

Number of Failures: 58 Reporting Period until 2006

80 70 60 50 40 30 20 10 0

  g   n    i   e   g    A

   t   c   e   u   r   u    d   l    i   o   r   a    P   F

  e   e   r    l   u    t    t    t    i   r   c   r    B   a    F

   l   a   e   c    i   n   g   a   a    h   m   c   a   e   D    M

  e   e   c   r   a   u    f    l   r    i   e   a    t   n   F    I

  g   n    i    l   a   e    S

  n   o    i    t    t   a   l    l    l   u   a   a   F    t   s   n    I

Failure Reason

Fig. 3: Survey from China – failures of composite insulators 

The reliability of the today’s generation of composite insulators, manufactured in accordance to the latest technology including a tight quality control and traceability is on par with that of ceramic insulators [8]. It is important however to stress that this level of reliability can only be reached, provided: 

 

2

The insulators characteristics (e.g. specific creepage distance) are correctly selected taking into account all the parameters which can affect their life (e.g. pollution condition of the site, mechanical stresses foreseen etc.). They are correctly handled and installed. Appropriate string design is used, including grading rings for grading of the E-field and corona and power arc control devices are correctly coordinated.

LIVE LINE DIAGNOSTIC METHODS AND PRINCIPLES

An important issue that limits an even wider application of composite insulators is the challenge to assess their conditions in service and especially before the application of LLW techniques for the replacement or repair of insulators, insulator strings or dampers or to repair the conductor. In particular, LLW on overhead lines equipped with composite insulators are not covered yet in the present standards and WG`s in IEC, CIGRE, IEEE are working on that issue. Some of the aspects under discussion are analyzed in the following and in particular: 



The progress in live line diagnostic methods and principles and methodology to evaluate the condition of insulators in service. The methodology to assess the conditions of the insulators specifically before LLW to ensure a safe operation.

Currently, both CIGRE WG B2.21 and IEEE 15.09.04.01 are working to update the state-ofthe-art of live line diagnostic techniques for composite insulators. However, it is important to state that the principles of diagnostic techniques remain those identified in [16]:  

 

Visual inspection and hydrophobicity assessment Infrared (IR) thermography Ultra-violet (UV) detection E-field measurements. 3

Visual inspection and IR/UV methods can be used both from the ground and from the air [1721]. Usually a helicopter is used for aerial inspections; however many more exotic devices from mini-helicopters to Zeppelins are on trial [21]. Despite the fact that the diagnostic technique principles are mostly the same as in [16], the diagnostic tools, like different camera systems, have been dramatically improved, making diagnostics much easier today. Furthermore, R&D activity has provided much better methods for interpretation of the measurements, which was a previous bottleneck, reducing their usage.

2.1

VISUAL INSPECTION

Visual inspection is presently the most commonly used inspection technique. It can be employed remotely from a long distance as well as by close-up visual inspections. Binoculars or telescopes are used to perform remote visual inspections. Better efficiency may be obtained when the inspections are made as close as possible to the insulator, e.g. operating from a tower, from a helicopter (including mini-helicopters), or from a bucket truck. A number of practical guides for visual inspection are available from CIGRE, EPRI and STRI. The guides typically include detailed descriptions of different types of possible defects with carefully selected colour photographic examples, enabling field personnel to quickly locate the photograph(s) and definition(s) of interest with respect to insulator deterioration and/or damages. CIGRE WG B2.21 collected and summarized the gained experience in the Technical Brochure TB 481, published in December 2011 [14]. In particular, it is important to define the criticality of a damage/defect, enabling required actions to be selected. The TB 481 can also assist in this regard. A simpler defect classification is necessary regarding LLW, as only conductive or semiconductive defects are recognized to be critical. This assumption is valid for the enforced rule that LLW is only permitted under dry weather conditions. While visual inspection allows most of the outside defects to be detected, internal defects, which might lead to flashunder, cannot be observed. Furthermore, visual inspection can generally provide rather qualitative information, which can be better quantified by the other diagnostic methods examined in the following sections.

2.2

INFRARED THERMOGRAPHY (IR)

A thermal emission is associated with local heating caused by a current flowing along a defective part of the insulator, characterized by relatively high conductivity in comparison to the intact insulating material. An example would be in the presence of tracking or other semiconductive paths as shown in Fig. 4.

Fig. 4: Examples of internal defects investigated [17] (left: tracking mark) and in [18] (right:  channel with low conductive moisture)  With IR, the temperature distribution along the insulator axis is measured by means of an infrared camera, searching for hot spots associated with possible local defects. Compact 4

cameras with high sensitivities and excellent performance were developed which currently permit a fast and reliable inspection of the insulators. An example of defect detection by IR is shown in Fig. 5. Guidelines for IR inspection from a helicopter are available [18]. The method is particularly sensitive to defects developing between the housing and the core, leading possibly to a flashunder. In this case the fault current is passing through the defective zone causing a significant temperature increase. The phenomenon is particularly evident when the tracking affects large parts of the insulator or when a sufficient conductivity is present all along the insulator. This could be because of humidity ingress in the interface or because of wetting of the part of the insulator not yet damaged [17], [18], [22]. On the contrary, temperature measurements by means of an infrared camera are not suitable to detect conductive or semi-conductive defects developing on only a small section of the insulator with the remaining part sound and characterized by high resistivity, especially if the measurements are made in a low humidity condition (with the insulator dry). In this case, corona may instead occur on the tip of the defect, with a very low current associated with it, thus leading to very limited temperature increase, hardly detectable in service [19], [20]. Corona measurements are thus an important tool as well, as discussed in the following chapter.

Fig. 5: Examples of clear IR detection of internal conductive defects (from left-to-right: in la-  boratory; at test station; in service) 

2.3

ULTRA-VIOLET DETECTION (UV)

The possibility of localizing initial corona activity constitutes an interesting technical challenge, especially in daylight conditions, thus different techniques are available for day and night measurements. For daylight corona cameras, the diagnostic indicator considered is the emission generated by the defects in the UV-C range (i.e. with wavelength in the range 240280 nm), a bandwidth in which the solar light is filtered by the atmosphere. Corona emission intensity is calculated using the number of pulses of light emission (named “blobs” - Fig. 6). A counter gives a number proportional to the quantity of “blobs” received by the sensor.

Clouds of blobs

Fig. 6: Example of “blob” counting by a daylight corona camera (245 kV tension insulators)  5

Very sensitive portable cameras are now available, and some of them are combined “multicameras”, which can also provide IR and visual observations with one instrument. The method is particularly efficient in detecting conductive/semi-conductive defects only developing on a section of the insulator, with the remaining part of the insulator in good condition and characterized by a very high resistivity, especially in dry conditions and low humidity. Under these conditions, the IR measurements indicated a very low temperature increase (in the order of 1-2 degrees) [19]. The sensitivity of the method was investigated by simulating conductive defects (using metal wires located on the insulator surface) at the live side, ground side and at a floating potential (in the middle of the insulator) [19]. The results are summarized in Fig. 7, where the minimum detectable defect length (in % of the insulator length) is presented for different insulator ratings. The UV camera was generally capable of detecting all conductive defects longer than about 20-30% of the insulator length. Similar results were obtained by substituting the metallic wire with a semi-conductive tape to reproduce typical tracking values [19]. 40 35

floating potential defects

30

ground potential defects live potential defects

   ] 25    %    [    L 20    /    d    L 15 10 5 0 100

150

200

250

300

350

400

Un [kV]

Fig. 7: Minimum conductive defect length detectable by UV for different defect positions and  for insulators of different ratings U n 

The question remained about the capability of the method to detect defects located under the insulator surface, possibly leading to insulator flashunder. While the UV method was not able to detect the defects when fully covered by the housing, removing a small part of the housing and thus uncovering the conductive end of the conductive defects (usual condition in case of severe defects [20]), the sensitivity of the method was seen again and very similar to that obtained with defects occurring on the surface. In general, IR thermography and UV measurements principally detect different physical properties (heat and enhancement of the electric field in the form of corona respectively), thus a combination of these two methods/cameras or use of a multi-camera would be the optimum solution for the remote inspection of composite insulators, especially when a certain failure tendency is known for the age or vintage of the insulators in question.

2.4

E-FIELD MEASUREMENTS (EF)

When a composite insulator is electrically defective, the electric field changes in the vicinity of the defective area. A portable, in-service and manually-operated diagnostic probe was developed from the version used for cap-and-pin insulators and has been tested in the laboratory and in the field to check its suitability. The principle of operation is based on the measurement and recording of the AC axial electric field along the insulator. The EF probe can easily be used by skilled LLW personnel. Defects that generate a distortion of the EF (i.e. conductive or semi-conductive defects) can be detected by comparing the EF pattern ob6

tained on the defective insulator with a reference fingerprint obtained on the sound insulator (see example in Fig. 8). To enhance the detection of defects, the data can be normalized by dividing the electric field value measured for each shed by the corresponding value of the plot taken as reference [19], [20].

Fig. 8: Examples of E-field probe application on a 420 kV insulator): left: defect at the fitting;  right: defect in the middle 

The maximum EF variation found between the corresponding values on defective and sound insulators (in p.u. of EF of the sound insulator) are shown in Fig. 9 as a function of the defect length and position. It can be shown that even relatively small conductive and semiconductive defects can be identified by this method. In the attempt to derive general rules, measurements were performed on insulators having different defects, generated during the manufacturing process (lack of primer, carbonization on the rod, break of the rod) or resulting from a long duration ageing test in different environmental conditions. Electrical field deviations higher than the intrinsic sensitivity of the methodology were confirmed for all types of defects investigated [19]. LIVE

GROUND

FLOATING

3.0    ]  .   u  .   p    [ 2.5   n   o    i    t   a    i   v 2.0   e    D    d    l 1.5   e    i    F   c    i   r 1.0    t   c   e    l    E0.5    X    A    M 0.0 0.0

0.1

0.2

0.3

0.4

0.5

Ld / L [p.u.]

Fig. 9: Maximum electric field deviation (p.u) as a function of the defect length and position  for conductive type defects 

Apart from being very sensitive, the advantage of this method relies on its capability of indicating defect size and location, as shown in Fig. 8. However, the method is quite demanding in terms of time/cost and expertise required and thus it is not economical to scan an entire overhead line. On the contrary, it may be very effective to ensure the safety of LLW on a specific insulator or insulator vintage. It was found empirically, that the EF-probe may not detect low severity level defects near the end fitting, due to electric field shielding by the corona ring. This by the way does not allow measurements closer than approximately 15 cm from the end fitting with corona rings. The atmospheric humidity during EF-probe measurements must be recorded. If an insulator is covered by a hygroscopic pollution layer, this situation must be considered when taking the reference measurement and analyzing the results. 7

3

EXPERIENCES WITH DIAGNOSTIC METHODS AND PRINCIPLES APPLIED TO INSULATOR STRINGS SUBJECTED TO LIVE LINE WORK

It is an important rule that LLW is always performed in fair weather, excluding conditions that could lead to pollution flashover, thus only the dielectric performance under transient voltages is of concern. If lightning is probable, LLW is usually forbidden or interrupted. So, only long front overvoltages are of concern, which are simulated in the laboratory by Switching Impulses (SI) [19], [20]. Many types of possible deteriorations of composite insulators in service do not affect the SI dielectric strength and are not of importance with respect to LLW feasibility. Examples of these defect types are pure external surface phenomena (chalking, color changes etc. [20]). Even shed damages (punctures, splitting etc.) may not be critical enough to cause a reduction of the dielectric strength under dry conditions required for LLW [19], [20]. Only conductive or partially conductive defects can affect the SI strength. Tracking type defects will always be potentially critical, since they are conductive or partially conductive even in dry conditions and are often hidden inside the insulator. However, the insulator can also become partly conductive following moisture ingress in the core and at the interface. This occurs, for example, when moisture penetrates through the punctures/splitting of the sheath towards the rod, thus increasing the conductivity of the rod or part of it [20]. Therefore, prior to LLW, the condition of the insulators of the span(s) subjected to maintenance are to be evaluated in order to detect any possible risk of flashover. Specific procedures are applied for this requirement, which is focused only on personnel safety. More specifically, the goal is to detect the quite large conductive or semiconducting type defects, since only large defects (e.g. 20-30% of the HV EHV insulator [19], [20]) create risk during the LLW. This specific requirement is very important for the assessment of the feasibility and suitability of the available diagnostic methods. LLW on composite insulators is carried out by many companies in the world, adopting different diagnostic tools, selected on the basis of their specific experience and taking into account the characteristics and the known weaknesses of the insulator generations installed in their system. In spite of the fact that R&D activities are going on to verify the efficiency of the different diagnostic methodologies, in many countries, like Australia and South America, LLW Example of the Italian Transmission Network up to 420 kV

Visual Inspection from Ground

Example of the Canadian Transmission Network up to 115 kV

IR-Inspection from Ground no

UV-Camera Corona Measurements

Temperature Increase measured?

No Action taken.

yes Tower Climbing

Tower Climbing

Close Visual Inspection from Tower

Noise, Heat or Visual Defect?

no Replace under Voltage.

yes EF-Measurement

Reschedule. Replace de-energized.

Fig. 10: Examples of diagnostic procedures before LLW  8

on composite insulators is still based mainly on careful visual inspection [24], [25], [26], [27], [28], [29]. Combined methodologies are proposed in other countries. As an example (Fig. 10) the Italian procedure [30] supplements visual inspection from ground level with both UVcamera and electric field measurements while in Canada, visual inspection is supplemented by thermal observation [22], [23]

4

CONCLUSIONS 











The many advantages of composite insulators have led to their wide use in transmission lines all over the world. The reliability of the present generation of composite insulators is considered to be similar to that of ceramic cap-and-pin insulators. An essential issue that limits an even wider application of composite insulators is the concern about the assessment of their conditions in service and especially before the application of live line working (LLW) techniques. The applicable diagnostic principles did not significantly change during the last 15 years. However, significant progress has occurred in the development of diagnostic tools and in the interpretation of measurements. On-line diagnostics, together with periodical tests on samples taken from service, enable reliable indications regarding the insulator conditions in service and regarding their life estimation to be obtained. Diagnostics is easier for LLW, since its aim is to identify only large conductive or semi-conductive defects which may be critical during the specific linemen activity. A well chosen combination of available diagnostic methodologies make it possible to identify the absence of critical defects of composite insulators and to carry out LLW safely on overhead lines equipped with composite insulators, in a similar way as it is for ceramic and glass insulators.

BIBLIOGRAPHY [1] [2] [3] [4] [5] [6] [7] [8] [9] [10]

CIGRE WG 22.10: “Technical basis for minimal requirement for composite insulators.” ELT_088_3, 1988 P-IEC 61109 Ed1: 1992: “Composite insulators for a.c. overhead lines with a nominal voltage greater than 1000 V - Definitions, test methods and acceptance criteria” K. Halsan, D. Loudon, I. Gutman, J. Lundquist: “Feasibility of upgrading 300 kV AC lines to DC for increased power transmission capability“, CIGRE-2008, B2-110 J. Lundquist, I. Gutman, K. Halsan: “Insulation Aspects for AC to DC Conversion of Transmission Lines“. World Congress & Exhibition on Insulators, Arresters & Bushings, Crete, 11-13 May 2009 http://www.tennet.org/english/tennet/publications/corporate_brochures/randstad38 0.aspx CIGRE SC 22 WG 03.01: Worldwide experience with HV composite insulators. ELECTRA 130, December 1990 CIGRE WG 22.03: Worldwide Service Experience with Composite Insulators. ELECTRA 191, August 2000. A. Pigini: “HV Composite Insulators Today: Market Size & Reliability.” INMR Issue 91 Quarter One 2011 Volume 19 Number 1 CIGRE WG 22.03: “Service Performance of Composite Insuators used in HVDC Lines”. ELECTRA 161, August 1995 A. Pigini, R. Cortina: “Insulation Challenges when Uprating EHV AC Lines or Converting EHVAC Lines to EHVDC Lines”. World Congress & Exhibition on Insulators, Arresters & Bushings, Seoul, Korea, 17-20 April 9

[11]

[12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30]

I. Gutman, K. Halsan, J. Seifert, W. Vosloo: “Pollution and ageing performance of DC line composite insulators: service experience vs. laboratory experience”. World Congress & Exhibition on Insulators, Arresters & Bushings, Seoul, Korea, 17-20 April 2011 “EPRI Survey of Utility Experiences with Composite/Polymer Components in Transmission Class (69 - 765 kV class) Substations”. EPRI, Palo Alto, CA, 2004, 1008817 IEC 62662 Ed1: 2008: “Guidance for production, testing and diagnostics of polymer insulators with respect to brittle fracture of core materials”. CIGRE WG B2.21: “Guide For The Assessment Of Composite Insulators in the Laboratory after their Removal From Service”. TB 481 published in December 2011 G. Zhicheng, W. Xiaoxing, J. Zhidong, Z. Ruobin: “Application and Research of Composite Insulator in China”. INMR Symposium 2011 in Seoul, South Korea CIGRE WG 22.03: “Review of “In service diagnostic testing” of composite insulators”, ELECTRA, No. 169, December 1996, p.p. 105-119 W.A. Chisholm: “Development of Live-Line Replacement Criteria for Partially Defective Non-ceramic Insulators,” ESMOL Panel Session on Live Working with NCIs (New Test Methods), April 9, 2003 I. Gutman: “Application of Helicopter-Based IR Technology for Detecting Internal Defects in Composite Insulators”, INMR, Issue 84, Quarter 2 - 2009, Volume 17, N. 2, 2009, p.p. 60-67 A. Pigini, G. Pirovano, G. Rocchetti: “Live line work on HV lines equipped with composite insulators: state of the art” INMR Seoul 2011 M. de Nigris, I. Gutman, A. Pigini: “Live-Line Maintenance of AC Overhead Lines Equipped with Non Ceramic Insulators (NCI)” TD conference and Exhibition, New Orleans 2010 paper TDC.2010.5484679 I. Gutman: “Helicopter & Ground-Based Methodologies to Detect Damaged Insulators”, World Congress & Exhibition on Insulators, Arresters & Bushings, Seoul, Korea, 17-20 April 2011 W.A. Chisholm: “Thermal Inspection Program Finds Failing Dead-End Polymeric Insulators”, INMR, Issue 83, Quarter 1 - 2009, Volume 17, N. 1, 2009, p.p. 34-39 M. Szeto, W.A. Chisholm: “Development of Live-Line Replacement Criteria for Partially Defective Polymer Insulators”, presented to the IEEE WG on Application and Inspection of In-Service Insulators, 23 July 2002 T. Gillespie: “Selection criteria&live line maintenance of composite insulators at an Australian Utility” INMR Q4-2005 “Researchers in Argentina explore new diagnostic methods to support live line replacement of composite insulators”. INMR Q3-2010 “South American Utilities plan live line replacement program to solve problems with composite insulators”. INMR Q1-201 Joel Rodolfo Elizarbe Córdova Cambio: “De Aisladores Poliméricos De Anclaje Y Suspensión En Líneas De Transmisión De 220 kV”. Del Dt Norte Cier Congress 2011 W. H. Santana Achury, I. A. Sanín Rincón, I. M. Loardo: “Metodología Tct Para El Cambio De Aisladores Poliméricos Con Condición Crítica”. Cier Congress 2011 H. Parisi J. Insogna: “Experiencia De Transba En La Deteccion De Aisladores Compuestos Deteriorados”. Cier Congress 2011 G. De Donà, C. Milanello, A. Posati, R. Gallo, C. Valagussa, U. Leva: “Dielectric behaviour of damaged composite insulating strings. Minimum approach distances calculation and individuation of the limit conditions for the safe live work”, ICOLIM 2008, Torun (Poland)

10