Hybrid machine learning models forwind speed prediction in renewable energy applications

Abstract

Wind power is a clean, eco-friendly renewable energy source that can beadopted as an alternative for other exhaustible energy sources such as coal.The generated wind power is highly non-linear in nature as it is directly in relation tothe wind speed, which is a highly nonlinear entity. So, for wind power generationand management, the wind speed forecasting is an inevitable process to be carriedout, in the proposed thesis various machine learning and neural network-based windspeed forecasting models are developed and their prediction performances areanalyzed to present a better method for efficient and accurate forecasting. newline