The rapid development of wind energy in Sweden created a volatile environment for the electricity market. Variance in the daily prices and the reductions of the average prices over the years due to the merit order effect of intermittent wind energy resulted in increased unpredictability in financial returns, which led to many wind projects being cancelled. In this thesis, in order to shed more light on the impact of wind energy development on spot prices, an artificial neural network (ANN) electricity price forecasting model is designed in order to predict Sweden’s four electricity regions Nord Pool Elspot day-ahead electricity spot market prices. The model's final result displayed a mean absolute error of 3.3398 €/MWh. In order for the model to be able to generalize better, a ridge regression regularizer is added to the ANN. Alternative wind scenarios for Sweden are introduced and their spot prices are predicted by the ANN model. The results show that each 10% increase in wind energy production leads to a 0.9% spot price reduction in the Nord Pool Swedish energy market prices.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:uu-339817 |
| Date | January 2018 |
| Creators | Kasimoglu, Ata |
| Publisher | Uppsala universitet, Institutionen för geovetenskaper |
| 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 0.0701 seconds