Support vector regression-based short-term wind power prediction with false neighbours filtered

Wind power prediction has received much attention due to the development renewable energy sources using wind power. The paper presents a new approach which is a support vector regression (SVR) based local predictor (LP) with false neighbours filtered (FNF-SVRLP) to undertake short-term wind power perdition. The proposed predication method not only combines the powerful SVR with the reconstruction properties of time series, but also overcomes the drawback of the original local predictor by removing false neighbours. The proposed method (FNF-SVRLP) is evaluated with the real world wind power data, and the final performance is compared with the support vector regression based local predictor (SVRLP) and the autoregressive moving average (ARMA). The results demonstrate that the proposed method can achieve a better performance than the other methods.