Failure Rate Analysis of Power Circuit Breaker in High Voltage Substation

Similar documents
Integration of Internet of Thing Technology in Digital Energy Network with Dispersed Generation

Following a musical performance from a partially specified score.

Instructions for Contributors to the International Journal of Microwave and Wireless Technologies

INSTRUCTION MANUAL FOR THE INSTALLATION, USE AND MAINTENANCE OF THE REGULATOR GENIUS POWER COMBI

Hybrid Transcoding for QoS Adaptive Video-on-Demand Services

System of Automatic Chinese Webpage Summarization Based on The Random Walk Algorithm of Dynamic Programming

Simon Sheu Computer Science National Tsing Hua Universtity Taiwan, ROC

LOW-COMPLEXITY VIDEO ENCODER FOR SMART EYES BASED ON UNDERDETERMINED BLIND SIGNAL SEPARATION

Small Area Co-Modeling of Point Estimates and Their Variances for Domains in the Current Employment Statistics Survey

RIAM Local Centre Woodwind, Brass & Percussion Syllabus

Product Information. Manual change system HWS

AIAA Optimal Sampling Techniques for Zone- Based Probabilistic Fatigue Life Prediction

Product Information. Manual change system HWS

Statistics AGAIN? Descriptives

AMP-LATCH* Ultra Novo mm [.025 in.] Ribbon Cable 02 MAR 12 Rev C

A Scalable HDD Video Recording Solution Using A Real-time File System

Product Information. Universal swivel units SRU-plus

T541 Flat Panel Monitor User Guide ENGLISH

Optimized PMU placement by combining topological approach and system dynamics aspects

Simple VBR Harmonic Broadcasting (SVHB)

Technical Information

Color Monitor. L200p. English. User s Guide

tj tj D... '4,... ::=~--lj c;;j _ ASPA: Automatic speech-pause analyzer* t> ,. "",. : : :::: :1'NTmAC' I

QUICK START GUIDE v0.98

Correcting Image Placement Errors Using Registration Control (RegC ) Technology In The Photomask Periphery

THE IMPORTANCE OF ARM-SWING DURING FORWARD DIVE AND REVERSE DIVE ON SPRINGBOARD

Error Concealment Aware Rate Shaping for Wireless Video Transport 1

Quantization of Three-Bit Logic for LDPC Decoding

MODELING AND ANALYZING THE VOCAL TRACT UNDER NORMAL AND STRESSFUL TALKING CONDITIONS

A STUDY OF TRUMPET ENVELOPES

A Comparative Analysis of Disk Scheduling Policies

Product Information. Universal swivel units SRU-plus 25

Decision Support by Interval SMART/SWING Incorporating. Imprecision into SMART and SWING Methods

Reduce Distillation Column Cost by Hybrid Particle Swarm and Ant

SONG STRUCTURE IDENTIFICATION OF JAVANESE GAMELAN MUSIC BASED ON ANALYSIS OF PERIODICITY DISTRIBUTION

Modular Plug Connectors (Standard and Small Conductor)

The UCD community has made this article openly available. Please share how this access benefits you. Your story matters!

Analysis of Subscription Demand for Pay-TV

Improving Reliability and Energy Efficiency of Disk Systems via Utilization Control

3 Part differentiation, 20 parameters, 3 histograms Up to patient results (including histograms) can be stored

Detecting Errors in Blood-Gas Measurement by Analysiswith Two Instruments

TRADE-OFF ANALYSIS TOOL FOR INTERACTIVE NONLINEAR MULTIOBJECTIVE OPTIMIZATION Petri Eskelinen 1, Kaisa Miettinen 2

The Traffic Image Is Dehazed Based on the Multi Scale Retinex Algorithm and Implementation in FPGA Cui Zhe1, a, Chao Li2, b *, Jiaqi Meng3, c

User s manual. Digital control relay SVA

Accepted Manuscript. An improved artificial bee colony algorithm for flexible job-shop scheduling problem with fuzzy processing time

Why Take Notes? Use the Whiteboard Capture System

FPGA Implementation of Cellular Automata Based Stream Cipher: YUGAM-128

Conettix D6600/D6100IPv6 Communications Receiver/Gateway Quick Start

Novel Quantization Strategies for Linear Prediction with Guarantees

www. ElectricalPartManuals. com l Basler Electric VOLTAGE REGULATOR FEATURES: CLASS 300 EQUIPMENT AVC63 4 FEATURES AND APPLICATIONS

arxiv: v1 [cs.cl] 12 Sep 2018

Product Bulletin 40C 40C-10R 40C-20R 40C-114R. Product Description For Solvent, Eco-Solvent, UV and Latex Inkjet and Screen Printing 3-mil vinyl films

Lost on the Web: Does Web Distribution Stimulate or Depress Television Viewing?

Critical Path Reduction of Distributed Arithmetic Based FIR Filter

Craig Webre, Sheriff Personnel Division/Law Enforcement Complex 1300 Lynn Street Thibodaux, Louisiana 70301

Sealed Circular LC Connector System Plug

Modeling Form for On-line Following of Musical Performances

www. ElectricalPartManuals. com l Basler Electric P. 0. BOX 269 HIGHLAND, ILLINOIS 62249, U.S.A. PHONE FAX

SWS 160. Moment loading. Technical data. M x max Nm M y max Nm. M z max Nm

INTERCOM SMART VIDEO DOORBELL. Installation & Configuration Guide

DT-500 OPERATION MANUAL MODE D'EMPLOI MANUAL DE MANEJO MANUAL DE OPERA(_._,O. H.-,lri-D PROJECTOR PROJECTEUR PROYECTOR PROJETOR

Anchor Box Optimization for Object Detection

CONNECTIONS GUIDE. To Find Your Hook.up Turn To Page 1

S Micro--Strip Tool in. S Combination Strip Tool ( ) S Cable Holder Assembly (Used only

Study on the location of building evacuation indicators based on eye tracking

Discussion Paper Series

CONNECTIONS GUIDE. To Find Your Hook.up Turn To Page 1

A Quantization-Friendly Separable Convolution for MobileNets

Printer Specifications

User Manual ANALOG/DIGITAL, POSTIONER RECEIVER WITH EMBEDDED VIACCESS AND COMMON INTERFACE

current activity shows on the top right corner in green. The steps appear in yellow

Product Information. Miniature rotary unit ERD

AN INTERACTIVE APPROACH FOR MULTI-CRITERIA SORTING PROBLEMS

Scalable QoS-Aware Disk-Scheduling

zenith Installation and Operating Guide HodelNumber I Z42PQ20 [ PLASHATV

Cost-Aware Fronthaul Rate Allocation to Maximize Benefit of Multi-User Reception in C-RAN

Simple Solution for Designing the Piecewise Linear Scalar Companding Quantizer for Gaussian Source

Environmental Reviews. Cause-effect analysis for sustainable development policy

IN DESCRIBING the tape transport of

Step 3: Select a Class

User Manual. AV Router. High quality VGA RGBHV matrix that distributes signals directly. Controlled via computer.

Clock Synchronization in Satellite, Terrestrial and IP Set-top Box for Digital Television

ELEGT110111C. Servicing & Technology November Pick and place and holding fixtures. Whatever happened to if transformers

User guide. Receiver-In-The-Ear hearing aids, rechargeable Hearing aid charger. resound.com

JTAG / Boundary Scan. Multidimensional JTAG / Boundary Scan Instrumentation. Get the total Coverage!

User guide. Receiver-In-The-Ear hearing aids, rechargeable Hearing aid charger. resound.com

KW11-P program.m~ble real-time clock Illtlior user's manual LPA b (~ (Etch Rev F and up)

CASH TRANSFER PROGRAMS WITH INCOME MULTIPLIERS: PROCAMPO IN MEXICO

User guide. Receiver-In-Ear hearing aids. resound.com

SCTE Broadband Premises Technician (BPT)

Academic Standards and Calendar Committee Report # : Proposed Academic Calendars , and

JTAG / Boundary Scan. Multidimensional JTAG / Boundary Scan Instrumentation

in Partial For the Degree of

Multi-Line Acquisition With Minimum Variance Beamforming in Medical Ultrasound Imaging

Turn it on. Your guide to getting the best out of BT Vision

Fast Intra-Prediction Mode Decision in H.264/AVC Based on Macroblock Properties

Q. YOU SAY IN PARAGRAPH 3 OF THlf PAPER THAT YOU'VE

Management of Partially Safe Buffers

Loewe bild 7.65 OLED. Set-up options. Loewe bild 7 cover Incl. Back cover. Loewe bild 7 cover kit Incl. Back cover and Speaker cover

User interface. Abbreviations / Meanings

Transcription:

T. Suwanasr, M. T. Hlang and C. Suwanasr / GMSAR Internatonal Journal 8 (2014) 1-6 Falure Rate Analyss of Power Crcut Breaker n Hgh Voltage Substaton Thanapong Suwanasr, May Thandar Hlang and Cattareeya Suwanasr Abstract Ths paper proposes the falure rates of power crcut breakers at the system voltage level of 115kV for control and protectve system n power substaton. Frstly, the recorded falure data of the exstng power crcut breakers n the hgh voltage substatons are analyzed. Secondly, the data analyzes are performed such as the classfcaton of dfferent falure types by separatng three man groups such as lve parts and nsulaton, control parts, and operatng mechansm parts durng the falure event year perod from 1989 to 2011 ncludng the total number of faled 607 power crcut breakers. Fnally, falure rates and mean tme between falures (MTBF) for all components n each man part can be estmated by usng Webull dstrbuton technque s dscussed for mprovng the relablty of the hgh voltage substatons such as correct mantenance schedule or renovaton tasks of equpment. The proposed method can also used wth other hgh voltage equpment n the power system. Keywords Age, faled types, falure rate, falure statstc, power crcut breaker. 1. ITRODUCTIO Power crcut breaker s one of the most mportant protecton and control apparatus n the power system. It s functons are used to sense a fault current for the control relay to operate the trp openng mechansm and then nterrupt the electrc crcut for preventng the power supply nterrupton n the power system. The power crcut breakers wth hgh falure statstc should be analyzed ntensvely for preventng unpredctable falure n order to determne the relablty of crcut breakers components and system relablty. Varous numbers and technologes of power crcut breakers were nstalled n the power system. The deteroraton of power crcut breakers depends on equpment qualty, operaton such as load stress, mantenance, surroundng envronment such as temperature, mosture, polluton, and etc. In ths paper, power crcut breakers at the system voltage level of 115kV ncludng the total number of faled 607 breakers durng the perod from 1989 to 2011 have been analyzed. The objectve of ths paper s to descrbe a method for estmatng the key relablty parameters such as falure rate and mean tme between falure (MTBF) from recorded falure data of the Electrcty Generatng Authorty of Thaland (EGAT) by usng Webull dstrbuton technque for mprovng the Asst. Prof. Dr.-Ing. Thanapong Suwanasr (correspondng author) s wth the Srndhorn Internatonal Tha German Graduate School of Engneerng (TGGS), Kng Mongkut s Unversty of Technology orth Bangkok (KMUTB), 1518 Pracharat 1 Rd., Bangsue, Bangkok, 10800 Thaland. E-mal: thanapongs@kmutnb.ac.th. May Thandar Hlang s a Master student of Electrcal Power and Energy Engneerng at TGGS, KMUTB. E-mal: thandar.mtdh@gmal.com. Asst. Prof. Dr. Cattareeya Suwanasr s wth the Department of Electrcal and Computer Engneerng, Faculty of Engneerng, KMUTB; Tel. +66-255-2000 Ext. 8518; Fax. +66-2585-7350; E-mal: cattareeyas@kmutnb.ac.th. relablty evaluaton purpose. Then, the correct mantenance schedule or renovaton tasks of equpment wth the mnmum cost can be appled. 2. CLASSIFICATIO OF FAILURE TYPES FOR RELIABILITY AALYSIS As the normal openng and closng moton of the power crcut breakers depends on ther components workng capablty, control, and operatng mechansms are very mportant for the performance and relablty of power crcut breakers [1]. In ths paper, the components of crcut breaker are dvded nto three man parts such as lve parts and nsulaton, control parts, and operatng mechansm parts. Frstly, the lve parts and nsulaton are separated nto two sub-components that consst of (1)man nsulaton to earth and (2)makng and breakng unts. Secondly, control parts nclude (1)auxlary swtches and assocated drves, (2)contractors, relays, heaters, thermostats, fuses, (3)gas densty supervson, and (4)trpng and closng unts. Lastly, operatng mechansm parts take account of (1)actuator, (2)compressors, motors, pump, ppe unon, (3)control elements, (4)counter, (5)dampng devce, (6)energy storage, (7)mechancal transmsson, and (8)no return devce, respectvely. The numbers of faled breakers for each man part are 91unts, 153unts, and 363unts respectvely and totally 607 breakers are observed. Accordng to the recorded faled data, the most falure percentages are 52% at makng and breakng unts for lve parts and nsulaton, 30% at contractors, relays, heaters, thermostats, fuses for control parts, and 23% at counter for operatng mechansm parts. The second are the man nsulaton to earth 48%, gas densty supervson 27% and mechancal transmsson 14% for each man part respectvely as shown n the Fgure-1 to Fgure-3. 1

T. Suwanasr, M. T. Hlang and C. Suwanasr / GMSAR Internatonal Journal 8 (2014) 1-6 Table 1. All Components and Ther umber of Falures Fg.1. Falure Percentage for Lve Parts and Insulaton Fg.2. Falure Percentage for Control Parts Fg.3. Falure Percentage for Operatng Mechansm Parts Fg.4. Falure Rates for All Components of Crcut Breaker Table 1 represents the number of falures by 100CB years for all components n each man part [2]. Ther results are shown n Fgure-4. It can be seen that the most falures rate s 0.441 at compressor, motors, pumps, ppe unon of operatng mechansm part. The second s 0.388 at contractors, relays, heaters, thermostats, fuses of the controls parts, and the thrd falure rate s 0.335 at mechancal transmsson of operatng mechansm parts respectvely. Type of Components o. of falures CB years o of falures/ 100CB years Man nsulaton to earth 31 24472 0.127 Makng and breakng unts 60 24472 0.245 Auxlary swtches and assocate drves Contractors, relays, heaters, etc 12 24472 0.049 95 24472 0.388 Gas densty supervson 26 24472 0.106 Trppng and closng crcuts 20 24472 0.082 Actuator 65 24472 0.266 Compressors, motors, pumps, ppe unon 108 24472 0.441 Control elements 6 24472 0.025 Counter 4 24472 0.016 Dampng devce 55 24472 0.225 Energy storage 35 24472 0.143 Mechancal transmssons 82 24472 0.335 o return devce 8 24472 0.033 Total 607 3. METHOD FOR ESTIMATIG RELIABILITY PARAMETERS Webull dstrbuton technque s one of the most wdely used for accurate falure analyss, falure forecast, agng and relablty [3]. Frstly, the recorded falure data of the exstng power crcut breakers n the hgh voltage substatons are analyzed. Then an order of falure events s ranked. Fnally, medan rank F(t) or probablty of falure s calculated by Equaton (1). 0.3 Medan Rank = + 0.4 (1) where s the adjusted rank and s the number of falures observed. Webull parameters can be determned by usng straght lne formula as follows. y= mx + c (2) where, 1 y = ln ln (3) 1- F(t) m = = =1 =1 x y - =1 2 x =1 2 - x y x =1 (4) 2

T. Suwanasr, M. T. Hlang and C. Suwanasr / GMSAR Internatonal Journal 8 (2014) 1-6 x = ln (t) (5) y x =1 =1 c = -m (6) c - m η = e (7) The Webull probablty dstrbuton functon (PDF) represents the probablty of falure at specfc tme (t), as wrtten by Equaton (8). t f(t)= η η -1 - t e η (8) Fg.6. Straght Lne of F(t) for Control Parts The Webull cumulatve dstrbuton functon (CDF) represents the probablty of falure at specfc tme (t), as wrtten by Equaton (9) whle the relablty and falure rate by Equatons (10)-(11). t - η F(t) = 1- e (9) t - η R(t) = 1- F(t) = e (10) f(t) t λ(t) = = R(t) η η -1 (11) where s the shape parameter and η s the lfe or scale parameter. Mean tme between falures (MTBF) s calculated by usng Equaton (12). 1 MTBF = η Γ 1+ (12) where, 1 Γ 1 s the gamma functon evaluated at the + value of (1+ 1/). The Webull cumulatve dstrbuton functon F(t) for three man parts are as shown n Fgure-5 to Fgure-7[3]. The horzontal axs shows the values of (x) and the vertcal axs s the values of (y) by usng Equatons (5) and Equaton (3) [4]. Fg.7. Straght Lne of F(t) for Operatng Mechansm Parts 4. STATICAL FAILURE AALYSIS OF POWER CIRCUIT BREAKER Calculatng procedure of Webull analyss nvolves data acquston, data rankng, data plottng and outcome nterpretng. After rankng by usng Equaton (1), the number of falure events are known [3]. Table-1 shows the Webull relablty parameters, MTBF, and falure rate (λ) of power crcut breakers at the level of voltage 115kV are summarzed from hstorcal record from 1989 to 2011 whch are scatterng and paper-based n nature, are systematcally recorded n the central database. Benefts of Webull analyss are that t provdes ratonally accurate falure analyss [3]-[6]. Fg.8. Falure Rates for Lve Parts and Insulaton Fg.5. Straght Lne of F(t) for Lve Parts and Insulaton 3

T. Suwanasr, M. T. Hlang and C. Suwanasr / GMSAR Internatonal Journal 8 (2014) 1-6 falure rate s 0.0960 (freq/year) at lve parts and nsulaton. The second s 0.0869 (freq/year) at control parts and the last falure rate s 0.0777 (freq/year) at operatng mechansm parts. Table 2. Relablty Parameters, MTBF, and Falure Rate for Each Component Fg.9. Falure Rates for Control Parts Fg.10. Falure Rates for Operatng Mechansm Parts The most falure rate occurs at makng and breakng unt for lve parts and nsulaton as shown n Fgure-8. For the control parts, contractors, relays, heater, thermostats, fuses s the most falure occurred and the second falure rate s at gas densty supervson as shown n Fgure-9. Accordng to Fgure-10, the most falure rate s at counter and the second s at dampng devce for operatng mechansm parts. Table-2 represents the relablty parameters, MTBFs and falure rates for each component. It can be seen that the most falure rates are 0.1664(freq/year) at counter, 0.107(freq/year) at makng and breakng unts, and 0.1052(freq/year) at dampng devces.the second falure rates are 0.0957(freq/year) at contractors, relays, heaters, thermostats, fuses etc. and 0.0946(freq/year) at mechancal transmsson. The least falure rates are 0.0469(freq/year) at auxlary swtches and assocated drves respectvely [3]. Fgure-11 shows the falure rates and the MTBF for each part of the crcut breaker. The most falure rate s at lve parts and nsulaton, the second s at control parts and the least falure rate s at operatng mechansm parts. Table-3 shows the relablty parameters, MTBFs, and falure rates for each man parts. It can be seen that the lfe tme of breaker (MTBFs) are nearly the same amount of 17(year) at lveparts and nsulaton and control parts. For operatng mechansm parts s 21(year). The most Type of Components Man nsulaton to earth Makng and breakng unts Auxlary swtches and assocated drves Contractors, relays, heaters, thermostats, fuses Gas densty supervson Trppng and closng crcuts Webull Parameters η MTBF (year) Falure Rate (λ) (freq/yr) 1.5895 20.8894 18.8005 0.0697 2.3355 18.2773 16.4496 0.1070 1.4972 30.1995 27.1795 0.0469 1.8873 17.8765 16.0889 0.0957 2.4086 23.4442 21.0998 0.0880 1.8178 23.1004 20.7904 0.0699 Actuator 3.6509 29.7327 26.7594 0.0860 Compressors, motors, pumps, ppe unon 3.1011 28.5908 25.7317 0.0818 Control elements 2.9999 26.3143 23.6829 0.0871 Counter 2.3048 11.5177 10.3660 0.1664 Dampng devce 2.2862 18.9271 17.0344 0.1052 Energy storage 2.1336 21.3189 19.1870 0.0878 Mechancal transmsson 1.6907 16.1716 14.5544 0.0946 on-return devce 2.0399 25.9740 23.3766 0.0692 Fg.11. Falure Rates and MTBFs for Each Man Part 4

T. Suwanasr, M. T. Hlang and C. Suwanasr / GMSAR Internatonal Journal 8 (2014) 1-6 Table 3. Relablty Parameters, MTBF, and Falure Rate for Each Man Part Type of Components Lve parts and nsulaton Webull Parameters η MTBF (year) Falure Rate (λ) (freq/yr) 2.0676 19.0644 17.1580 0.0960 Control parts 1.9474 19.5946 17.6351 0.0869 Operatng mechansm parts 2.1140 23.7381 21.3643 0.0777 4. COMPARISO OF FAILURE RATES FOR ALL COMPOET Fgure-12 shows the Webull cumulatve dstrbuton functon F(t) for all components[3]. Smlarly, Fg.13 shows the falure rate and MTBF of all components n the breaker at assocated voltage level. In addton, Table-4 shows the values of Webull parameters, MTBF, and the falure rate of combned all components n the crcut breaker at 115kV voltage level. Each value are 2.0695, 21.8984, 19.7096, and 0.0812 respetvely. Fg.12. Straght Lne of F(t) for Combned All Components Fg.13. Falure Rate and MTBF for All Components Table 4. Relablty Parameters, MTBF, and Falure Rate for All Components Type of Components All components Webull Parameters 5. COCLUSIOS η 2.0695 21.8984 MTBF (year) 19.708 6 Falure Rate (λ) (freq/yr) 0.0812 In ths paper, the total number of faled 607 breakers are consdered durng the event year 1989 to 2011 whch are orgnated from separatng three man parts such as lve parts and nsulaton, control parts, and operatng mechansm parts at the system voltage 115kV. After that, all components falure data are combned and compared the results.the lfe tmes of breakers and the falure rates can be calculated from the recorded falure data of EGAT by usng Webull statstc dstrbuton technque. The value of shape parameter () can descrbe the falure mode of the devces, power crcut breakers n ths paper. Accordng to the results, the values of () for three man parts are greater than 1. It means that the falure rates are very low at the begnnng and sgnfcantly hgher wth ncreasng tme. Ths can be seen that the falure rates of the components depend on the agng. The lfe or scale parameter (η) represents the tme for whch the falure percentage s 63.2%. The MTBF represents the lfe tme of devce, n ths paper power crcut breaker. The MTBFs are nearly the same amount of 17 year at lve parts and nsulaton and control parts, 21 year at operatng mechansm parts and all components s 19 year respectvely. The falure rates are nearly the same amount of 0.09 for lve parts and nsulaton and control parts, 0.08 for operatng mechansm parts and all components. Each component has ts own falure rate, but the performance of crcut breaker as a whole part depends on the state of all components. For relablty pont of vew, the breaker components are connected n seres, so falure of a sngle component prevents proper fault nterrupton and causes breaker falure. If relablty of the components decreases, the system relablty can also be decreased. So, the correct mantenance schedule or renovaton tasks of equpment are requred for relablty evaluaton. Accordng to the results, the proposed method s approxmated lfe tme of the equpment for strategy plannng on relablty evaluaton. Fnally, the study results can be carred out and estmated falure rate, and the lfe tme (MTBF) of other hgh voltage equpment n the power system. ACKOWLEDGEMET The authors gratefully acknowledge the Transmsson System Mantenance Dvson at the Electrcty Generatng Authorty of Thaland (EGAT) for supportng the data. May Thandar Hlang would lke to gve specal grattude to Thaland Internatonal Development Cooperaton Agency (TICA) for fnancal support. 5

T. Suwanasr, M. T. Hlang and C. Suwanasr / GMSAR Internatonal Journal 8 (2014) 1-6 REFERECES [1] IEEE Std C37.10-1995 (R2002) IEEE Gude for Dagnostcs and Falure Investgaton of Power Crcut Breakers, 1995. [2] T.M Lndqust, L. Bertlng, R. Erksson Crcut breaker falure data and relablty modellng, publshed n The Insttute of Engneerng and Technology, Generaton, Transmsson & Dstrbuton, 2008, do:10.1049/et-gtd:20080127 [3] Robert B. Abernethy, The ew Webull Handbook ffth edton, Relablty & Statstcal Analyss for Predctng Lfe, Safety, Survvablty, Rsk, Cost and Warranty Clams, ovember 2006. [4] W. Tppachon, R. Boonruang, S. Boonpun, C. Khamtang,. Klarung, D. Rerkpredapong, J. Hokert, Falure Analyss of Protectve Devces n Power Dstrbuton Systems for Relablty Purpose. TECO 2006, IEEE conference publcatons, 2006 [5] Q.Bnh dam,a.p. Saks Meloppulos, Predcton of Crcut Breaker Tme-to-Falure consderng Generaton Capacty Growth, IEEE conference publcatons, 2008. [6] R.A. Jongen, P.H.F. Morshus, J.J. Smth, A.L.J. Janssen, E. Gulsk, Falure data of power transformers as nput for statstcal falure analyss, 15 Internatonal Symposum on Hgh Voltage Engneerng, Unversty of Ljubljana, Solovena, August 27-31 2007. 6