Ultra-short-term wind power prediction based on double decomposition and LSSVM

To reduce the influence of random fluctuation on wind power prediction, a new ultra-short-term prediction model, based wavelet decomposition (WD), variational mode (VMD), and least-squares support vector machine (LSSVM), is proposed in this paper. The method double LSSVM, where sequence decomposed by WD into low- high-frequency components, which are further VMD to obtain many modal components with tendency periodicity. Multiple LSSVM models then established historical data weather as inputs predicted values multiple components. final achieved fusion outputs these models. experimental results show that MAPE (mean absolute percentage error) combined model 4.66%, best compared nine benchmark This demonstrates high performance WD-VMD-LSSVM for short-term power.