A novel strategy for wind power forecast through neural networks : Applications to the uruguayan electricity system
Resumen:
In systems with a high penetration of wind power generation, the precision of the forecasts is a critical input for the electricity dispatch planning. In this paper, we present the methodology that has been used to implement a complete update of the wind power forecast model in Uruguay. The new model increases the precision of the forecasts both in low and high power scenarios. It allows to perform a more efficient short-term electricity dispatch, improving the resource valuation, the inter-systems energy exchanges and the prevision of the wholesale electricity market spot price. According to the simulations performed, the new model increase the precision of wind power forecasts between 7% and 32%. The model is on its production phase and their results can be accessed through pronos.adme.com.uy/svg and latorrex.adme.com.uy/vates.
| 2023 | |
|
Training Adaptation models Uncertainty Wind speed Wind power generation Predictive models Wind farms Renewable energy systems Forecasting Wind energy Neural networks Wind turbine power curve |
|
| Inglés | |
| Universidad de la República | |
| COLIBRI | |
| https://hdl.handle.net/20.500.12008/42459 | |
| Acceso abierto | |
| Licencia Creative Commons Atribución - No Comercial - Sin Derivadas (CC - By-NC-ND 4.0) |
| Sumario: | In systems with a high penetration of wind power generation, the precision of the forecasts is a critical input for the electricity dispatch planning. In this paper, we present the methodology that has been used to implement a complete update of the wind power forecast model in Uruguay. The new model increases the precision of the forecasts both in low and high power scenarios. It allows to perform a more efficient short-term electricity dispatch, improving the resource valuation, the inter-systems energy exchanges and the prevision of the wholesale electricity market spot price. According to the simulations performed, the new model increase the precision of wind power forecasts between 7% and 32%. The model is on its production phase and their results can be accessed through pronos.adme.com.uy/svg and latorrex.adme.com.uy/vates. |
|---|
