Nonlinear Modeling of Wind Power Output Using Box-Cox Transformation and Its Application to Very Short-term Prediction

We propose a modeling method of the wind power output for very short-term prediction. The model has cascade structure composed of two parts: One is a linear dynamical part that is driven by a white Gaussian noise and described by an auto regressive model. The other is a nonlinear static part that is driven by the output of the linear part. This nonlinear part is designed such that the distribution of its output matches with that of the wind power output. The constructed model is utilized to one-step ahead prediction of the wind power output. Furthermore, we evaluate the relation between the prediction accuracy and the sampling period for modeling.