This thesis explored different approaches of clustering residential data. The goal was to develop a model with applications in load forecasting contexts, specifically in situations where only a limited amount of residential data is available. Four different types of approaches were explored, one of which utilised not only data pertaining to the load profile but also data related to the residency. Effects of seasonal and weekly variations were studied to identify how the load profiles were affected by these parameters. In the end the developed clusters were evaluated using silhouette scores as well as using load forecasting models developed outside of the thesis.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:kau-90240 |
| Date | January 2022 |
| Creators | Jones, Philip |
| Publisher | Karlstads universitet |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.2079 seconds