Machine Learning Approaches for Short-Range Wind Power Estimation: A Perspective

^* Department of Medical Laboratory Technics, Al-Zahrawi University College, Karbala, Iraq
^† Department of Electronics & Communication Engineering, IES College Of Technology, IES University, Madhya Pradesh 462044 India, Bhopal .
^‡ Assistant Professor, Department of S&H, Prince Shri Venkateshwara Padmavathy Engineering College, Chennai - 127
^§ The Islamic university, Najaf, Iraq
^** Department of Management, Uttaranchal Institute of Management, Uttaranchal University, Dehradun-248007, India
⁶ Assistant Professor, Department of BCA, K.S.Rangasamy College of Arts and Science (Autonomous), Tiruchengode. Mail Id: vimalrmail@gmail.com

^* Corresponding Author :ahmeds909091@gmail.com
^† research@iesbpl.ac.in
^‡ p.sharmila_phy@psvpec.in
^§ abathermahmood560@gmail.com
^** Ankitajoshi2010@yahoo.com

Abstract

The evolution of wind energy production, especially in near and offshore farms, has seen significant advancements due to the integration of novel technologies and the reduction in economic costs. This paper reviews the work in the domain of wind power estimation, emphasizing the innovative approaches leveraging satellite data and artificial intelligence (AI) methodologies. A notable method integrates Sentinel satellite imagery analysis in a two-phased approach, combined with machine learning techniques, to forecast wind speed. This method utilizes sentinel-1 and sentinel-2 satellite images for wind speed and bathymetry analysis, respectively. Furthermore, a hybrid forecasting model, comprising the generalized regression neural network (GRNN) and the whale optimization algorithm (WOA), has been introduced. Another pivotal advancement comes from the National Center for Atmospheric Research (NCAR), which has revamped its wind power forecasting system. This enhancement focuses on short-term forecasting, uncertainty quantification in wind speed prediction, and the prediction of extreme events like icing. The integration of numerical weather prediction with machine-learning methods, such as the fuzzy logic artificial intelligence system, has further elevated the accuracy and efficiency of these forecasting models. Collectively, these advancements offer a comprehensive perspective on the future of shortrange wind power estimation.