As part of that work, Louka et.al. introduced modelling that is being applied in Greece[16].
For instance, the University of Athens used SKIRON modelling to forecast wind speed for upto 5 days ahead. At the same time, the Regional Atmospheric Modelling System (RAMS) was developed at Colorado State University and Mission Research Inc. ASTeR Division. RAMS can forecast wind upto 48 h later. In addition, an adaptive fuzzy neural network (F-NN) also applied wind power prediction for upto120 h ahead. However, these methods cannot handle systematic errors that are caused by local adaptation problems. The authors proposed Kalman filtering to improve the performance of the aforementioned methods. Kalman filtering is one of the statistically optimal sequential estimation procedures for dynamic systems. There sults from the research show that systematic errors can be eliminated using Kalman filtering.