This paper investigates the potential of mathematical algorithms, based on the principles of data mining, applied to data from a local power grid to identify customers with large fluctuations in their power curves. A literature study was performed to facilitate algorithm formation. The data provided by the local grid owners, Falkenberg Energi AB, was analysed in MATLAB and two novel algorithms was created. The results show that, by normalizing all the data, it is possible to find and select customers with large fluctuations in their power curves. Key performance indicators were then used to determine which algorithm performed better. One of the algorithms performed better in all tested indicators and was used to create a list with interesting customers to Falkenberg Energi AB. The conclusion of the study shows that the proposed algorithms can be applied on a local power grid to select customers, but more research is needed to validate these methods. The conclusion also indicates that a reduction of the power peaks, at the identified customers, mainly affect the local power grid and not the power supply from the overhead regional power grid.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:hh-42094 |
| Date | January 2020 |
| Creators | Larsson, Mikael, Persson, Simon |
| Publisher | Högskolan i Halmstad, Akademin för ekonomi, teknik och naturvetenskap, Högskolan i Halmstad, Akademin för ekonomi, teknik och naturvetenskap |
| 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 |
Page generated in 0.0021 seconds