Offshore wind speed short-term forecasting based on a hybrid method: Swarm decomposition and meta-extreme learning machine

As the share of global offshore wind energy in electricity generation portfolio is rapidly increasing, grid integration large-scale farms becoming interest. Due to intermittency wind, stability power systems challenging. Therefore, accurate and fast short-term speed forecasting tools play important role maintaining reliability safe operation system. This paper proposes a novel hybrid model based on swarm decomposition (SWD) meta-extreme learning machine (Meta-ELM). approach combines advantages SWD which has proven efficiency for non-stationary signals, with Meta-ELM provides faster calculation lower computational burden. In order enhance accuracy stability, signal decomposed by implementing swarm-prey hunting algorithm SWD. To validate model, comparison against four conventional state-of-the-art models performed. The implemented are tested two real datasets. results demonstrate that proposed outperforms counterparts all performance metrics considered. can also improve as well-known robust method.