Influence of differences between cable system and overhead lines

One way to increase distribution network reliability is to replace traditional overhead lines with underground cables. To fully utilize these investments, network owners will have to adjust their reliability engineering methods to suit the new cable networks. In this paper different condition assessment methods as well as improved failure statistics for cable systems are considered. The paper is based on information from project reports and scientific papers. In addition a number of Swedish distribution network owners have been asked to share their opinion of reliability engineering work of today and the future. The methods used for distribution system condition assessment change as overhead lines are replaced by underground cable. Visual inspections can no longer be used and diagnostic methods are instead introduced. The diagnostic methods are costly to perform. To become frequently used they must prove to be efficient enough to justify the financial means required. New, improved, diagnostic methods are under development and since most cable failures are related to component properties the use of diagnostic methods is likely to increase. Statistics show that a majority of the cable system failures are ageing failures. The ageing failures do not to any large extent depend on environmental factors but on component properties. This implies that component lifetime standard deviation decreases as component data are related to the statistics, yet failure statistics is at present not related to component data. A majority of the network owners contributing to this paper agree that the nature of cable failures, except excavation failures, makes it interesting and useful to related cable data to failure statistics and to share the statistics with other network owners. Several Swedish distribution network owners are in the process of installing new program for network analysis. One challenge is to decide which component data that shall be related to the failure statistics. Operational age, maintenance history, manufacturer and year of manufacturer are four manageable factors which influence on failure statistics is already established but yet not thoroughly evaluated. In addition the method of cable excavation is suggested as a fifth interesting factor related to cable system reliability.