This paper thoroughly investigates needs and requirements for overhead distribution feeder inspection and develops models to investigate possible relations between short term inspections remarks and outages. The study was conducted in collaboration with Fortum Distribution AB that supplied extensive information about their overhead power feeders concerning both inspection and power outages. The investigated models where lognormal linear model, Poisson generalized linear model and negative binomial generalized linear model. All models were implemented utilizing offset terms to compensate for differences in feeder length and amount of overhead versus underground feeders. The Poisson generalized linear model was rejected at an early stage due to overdispersion and neither of the remaining models fit the data perfectly. There for conclusions were primarily concluded from similarities and differences amongst the models. The results either implicate that maintenance is scheduled sufficiently fast, with respect to short term inspection, to ensure a high system reliability or no relationships between short term inspection and feeder outages exist. Therefore the final conclusion was that maintenance and reinvestment decisions should not be based on short cycle inspection data.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:uu-219572 |
| Date | January 2014 |
| Creators | Brodersson, Anna Lilly |
| Publisher | Uppsala universitet, Tillämpad matematik och statistik |
| Source Sets | DiVA Archive at Upsalla University |
| Language | Swedish |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
| Relation | UPTEC STS, 1650-8319 ; 14002 |
Page generated in 0.0018 seconds