Abstract
Wind power production is variable by nature. These variations can be forecasted to some extent, by methods based on time series analysis or neural networks, for example. When forecasting more than 4 hours ahead, meteorological forecasts for wind speed are needed. The longer the time horizon, the more difficult it gets to forecast the production accurately, especially at hour-to-hour precision. Forecasting day ahead means having to correct on average 30-40 % of wind power production later. To handle the forecast errors, there can be trade up to the delivery hour, and the final imbalances to schedules are charged by regulating power prices. This paper outlines the forecast errors of wind power producers in the electricity market. The resulting benefits of shorter times between bids and delivery of production, as well as the benefits of pooling wind power production from larger areas are presented.
| Original language | English |
|---|---|
| Title of host publication | 2006 International Conference on Probabilistic Methods Applied to Power Systems, PMAPS |
| Publisher | IEEE Institute of Electrical and Electronic Engineers |
| ISBN (Print) | 978-91-7178-585-5 |
| DOIs | |
| Publication status | Published - 2006 |
| MoE publication type | A4 Article in a conference publication |
| Event | 9th International Conference on Probabilistic Methods Applied to Power Systems, PMAPS 2006 - Stockholm, Sweden Duration: 11 Jun 2006 → 15 Jun 2006 Conference number: 9 |
Conference
| Conference | 9th International Conference on Probabilistic Methods Applied to Power Systems, PMAPS 2006 |
|---|---|
| Abbreviated title | PMAPS 2006 |
| Country/Territory | Sweden |
| City | Stockholm |
| Period | 11/06/06 → 15/06/06 |
