University of Twente Student Theses

Login

An assessment of wind power forecasting models and its financial implications for the traders

Amtsfeld, Jonas (2019) An assessment of wind power forecasting models and its financial implications for the traders.

[img] PDF
3MB
Abstract:With this paper, we develop a method to assess the two different wind power forecast models of DVEP and try to find the ideal bidding strategy at the day-ahead electricity market. We use the metrics of NBIAS, NMAE and NRMSE to assess the accuracy of the forecasts. This assessment is conducted under circumstances of different wind speeds, wind directions, temperatures and day hours. We found that overall the forecasts are not significantly different, however, when the different circumstances are considered, we could assign one forecast to be better. To find the ideal bidding volume, different strategies with both point and probabilistic forecasts are used. While the point forecasts are only based on the two wind power forecasts, the probabilistic forecasts use also resampled historical price and production data. We found that adding historical price data to the decision process of finding the bidding volume does not improve financial results. We recognize that the price uncertainty of both spot and imbalance prices is the main reason for this. Based on the analyzed month, July and August, there are only minor differences in monetary results for Forecast 1 and 2. When considering a probabilistic strategy, the method of resampling price and production data together with a VaR 0 lead to the best results. However, those results were under the results of the point forecasts.
Item Type:Essay (Master)
Clients:
DVEP, Hengelo, Nederland
Faculty:BMS: Behavioural, Management and Social Sciences
Subject:01 general works, 50 technical science in general, 83 economics
Programme:Industrial Engineering and Management MSc (60029)
Link to this item:https://purl.utwente.nl/essays/80233
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page