Available under a Creative Commons Attribution Non-Commercial Share Alike 4.0 International Licence
This paper explores wind energy forecasting consistency by considering the error benchmarks associated with the generation output of a small wind farm in comparison to the national forecasting as provided by Eirgrid, the Irish TSO. This percentage error analysis will contrast the predicted (Eirgrid) capacity and actual wind energy output observations (Wind farm) and postulations that consider alternative prediction metrics are discussed. The findings suggest that in monthly like for like comparisons over a twelve month period, total MWh percentage errors of -0.36% and 5.7% are observed respectively for the actual generation and the forecasted prediction, when a monthly averaged window is considered. However if one considers higher frequency observations, such as those available through Eirgrid (15 minute periods), a mean absolute error of 28.5% is evident for the national wind generation capacity over the course of the same year. In the context of the proposed Irish wholesale and integrated single electricity market (I-SEM), an error of this magnitude could have severe financial implications for the wind energy sector; particularly if wind is to become the primary component of the future Irish energy mix.
McDonald, M., Woolington, T. & Sunderland, K. (2015). Wind Energy and Ireland: Could forecasting errors lead to a flawed market? In: UPEC (ed.) 50th International Universities Power Engineering Conference (UPEC 2015). Staffordshire University, Stoke on Trent IEEE.