Forecasting of Wind Power Generation with the Use of Artificial Neural Networks and Support Vector Regression Models

摘要: Abstract The stochastic character of wind power generation suggests limitations on the increased shares energy in electricity systems and challenges market integration power, mainly due to fact that nowadays, new parks are set cope with more dynamic pricing mechanisms. In this environment, where advanced bidding strategies need be adopted from actors, introduction novel elements support address inherent impact variability is thought a prerequisite. To end, current study expands work previous studies by examining different methods prediction regards forecasting. More specifically, both Artificial Neural Networks (ANNs) Support Vector Regression (SVR) models trained tested basis horizons, using as case real speed measurements park operating Greek territory. Models an in-house forecasting tool, results obtained reflecting better fit SVR method overall, especially for time horizons longer than 6 hours ahead. At same time, effort made order optimize through combination approaches via clustering areas. This approach improvement predictions obtained, despite already performs sufficiently even 24