Short-term wind power forecasting based on dynamic system of equations

The aim of this study is to identify a class of models that appropriately describe the wind power production in wind farms and to provide reliable forecasts in a 72-hour horizon. The activity of the turbines in a wind farm determines the total power produced at a point in time. Their functioning depends on climatic conditions and on electrical and/or mechanical breaks. The latter can be either controlled (in case of maintenance) or uncontrolled (in case of dysfunctions). Therefore, both the uncertainty related to the climate (wind speed, direction, air density etc.) and to the functioning of turbines affect the total power generation in the wind farm. We propose a new approach for modeling these sources of variability, focusing on the average power production rather than on the overall wind farm power production, like the previous literature. The advantage of this strategy is to reduce the high variability of the total power, separating the climatic component from the mechanical functioning of the turbines. We propose several models, all in the form of a dynamic system of equations, of increasing complexity due to the number of variables that have to be estimated in the system.