FA49F4E0d01 - On the Statistical Testing of Solid...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
On the Statistical Testing of Solid Insulation Systems M.M. Morcos, K.D. Srivastava Author Affiliation: Kansas Statc Univcrsity, M;inhattan, KS; Uni- versity of British Columbia, Vancouver, Cen;ida Introduction: The design ol insulalioii systems for power systcni equipment has been primarily deterministic in nature. Ncw matcrials were usually evaluated with a Set of valUcS charecteriring a number of propcrtics such as breakdown voltage, tensilt. strength, antl mehing tcmpcraturc. Safety fhctors based on experience with similar materials wcrc introduccd, and tlic iniximum permissible electric slress, lemper- ature, and mechanical slrengtlis Sor a reasonablc servicc lifc of tlic ma- tcrid wcre cstablished for design purposes. The determinislic approach might be supplcmeiited by comparison tests against an existing design with a provcn service recold. Designs bccmne slantliirtli~etl once male- rial properties are agreed upon by many panics. The deterministic design approach has becn gr;dually rcplacctl by probabilistic dcsign procedorcs. Thc iniiiii rcaon fiir this Il.ansitimi is a hroad recognitioii lhat inlaterial properties are variable even for idcnti- cal samples of lhe same material. In addition, the stresses to which the equipmenl will be exposed Eire not constant; they vary with ccrtain probabilities of occurreiicc. Furtliermorc, safcty factors wecl in a deter- ministic design arc often difficult 111 justify and are subiecl to be re- duced to savc money. Sincc a statistical dcsign pmceilure takes iill such factors into account, cquipmcnt with ii pmjojcctcd life under specific up crating conditions can he menuf;ictured more economicelly [I I. The choicc of tcst prnccdurc is dctcrmincd by the tesl ob,jective. Constant-voltage tests providc rcliablc comprchcnsive d;its for tlie dis- tribution Sunclion of the breakdown lime, but determillation of a lifc characteristic is time consuming. Rising-voltage tcst. sivc, but gciicralizatiim of thc tcst rcs~lt is difficult to implcmenl. I1 should be pointed out that solid insulilting inaterials are not self-restoring; a separale tcsl piecc is conscquently nccdcd for each ill- dividual [est. Since solid lest pieces are rclalivcly cxpcosive, reseiirch is fvrccd to go with a sinall numbcr of test pieces. Thc iipplicatiun of stiilislictl mcthorls LO thc smdy of thc pcrlbr- malice of cables with extriided insulation lids bcen thoroughly veriiied. Generally, all lhc authors agree on the validily or the two-dimensional Weiboll distribution, which is in fairly good agrccmcnt with cxpcri- tiientiil results. The possibility of carrying out the I~rgcsI nuniber of tests on miniature cables has bccii cunsidcrcd so as 10 reduce tir a mini- mum the numbcr of tcsts on full-sizc cahlcs and to cnahle savings on both timc and costs to bc madc. Ah, full-size cables were employed in the tests. Breakdown iii Solids: Solid insuliitions
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 06/08/2011 for the course ELECTRICAL 124 taught by Professor Ghjk during the Spring '11 term at Institute of Technology.

Page1 / 3

FA49F4E0d01 - On the Statistical Testing of Solid...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online