Back to the programme printer.gif Print

Delegates are invited to meet and discuss with the poster presenters in this topic directly after the session 'Forecasting: Maximizing grid deliverability and leading your business processes to profitability' taking place on Thursday, 13 March 2014 at 11:15-12:45. The meet-the-authors will take place in the poster area.

(1) METEOROLOGICAL AND HYDROLOGICAL SERVICE, ZAGREB, Croatia (2) Croatian Transimission System Operator, Zagreb, Croatia

Printer friendly version: printer.gif Print


Wind forecasting for energy management in Croatia


The increase of the wind energy share in Croatia becomes a major challenge, since larger uptake of wind energy increases threats to safe power system operation. The efficient way to effectively regulate the power system is to utilize high-performance forecasts of electricity production. The Croatian TSO has access to NMS wind forecast since 2009. and use it for prediction of wind energy production. The aim of this work is to estimate energy prediction errors by comparing forecasted and actual aggregated wind farms output for region of Croatia. Obtained energy prediction errors indicate how valuable existing wind forecast is for energy management.


Approach to the verification of wind energy production in Croatia is based on forecasted and aggregate energy production at three wind farm locations in one year period. Usual statistical verification indicators such as MAE and RMSE have been used for estimate the wind speed and energy production forecast errors. Emphasis has been given to weather situations with strong bura wind.

Main body of abstract

The Croatian Transmission System Operator (HOPS) is the key player in supporting the accelerated uptake of wind energy and the main actor in defining the national limits of maximum wind energy penetration into the power network. By utilizing forecasts of electricity production from wind power plants the HOPS balance reserves and schedule energy production. The model used for energy production forecast is the Wind Power Planning Tool (WPPT) (Gieble and Nielsen, 2008). The prediction covers a horizon corresponding to the prediction horizon of the meteorological model - 72 hours.
The Croatian Meteorological and Hydrological Service (CMHS) uses ALADIN limited area model for the operational weather forecast. The operational ALADIN model forecast is run on a Lambert-projection domain with 8 km horizontal resolution. The model fields are subsequently going through a dynamical adaptation procedure that produces 2 km resolution forecast of 80 m wind speed and gusts.
Statistical verification carried out with the use of mean absolute error (MAE) and root-mean-square error (RMSE), suggests that the model performance is successful specially for 6-18 h forecast range.
Forecasted and actual aggregated wind plants output for region of Croatia during the one year period show that the forecast overestimate energy production for 2-4% in summer months and up to 10% during the weather situations characterized by the strong and gusty bura wind.


The entire Croatian wind energy sector lacks efficient wind energy management, and in particular a dedicated state-of-the-art wind and wind power forecasting system designed for the specific and challenging wind climate of Croatia. The state-of-the-art performance of the predictive technologies is essential, because excessive prediction errors and uncertainties raise the costs of the wind energy utilization and hinder the efforts to integrate more wind energy into the Croatian power system.

Learning objectives
Generic wind prediction, although gives very valuable input for wind energy production forecast, is of limited use in the wind energy sector, resulting in energy prediction errors reaching twice the state-of-the-art in the TSOs in the EU. Consequently, a strong need for improved site-specific predictions of wind conditions and quantitative assessment of the prediction system uncertainties is evident.