Enhancing wind power forecasting: A bootstrap resampling interpolated Markov model

Abstract

This paper presents an improved Markov forecasting for very short-term (1 hour) wind prediction in electrical power systems. The method utilizes the probability transition matrix, obtained for the Markov Model, to observe the trends of the data; this is used to forecast the power for an hour ahead in time. Next, improvement of the forecast is achieved using the method of weighted interpolation where weights are obtained using bootstrap resampling. Finally, the forecast is improved slightly using a hybrid of the two aforementioned approaches. Past wind farm power production data are used to develop the proposed model. Computer simulations using Northwestern weather recordings from the Bonneville Power Administration (BPA) website show good correlation between our predictions and the actual data.