Economic and Technical Assessment of Wind Potential Using SARIMAX Time Series Models: Wind Speed Forecasting and Analysis

ABSTRACT The worldwide transition to renewable energy is gaining momentum, highlighting the importance of identifying and utilizing local wind resources as a viable approach to decrease dependence on fossil fuels and promote sustainable energy security. This investigation examined 15 years of wind speed data (2010–2024) from Khaf, Iran, utilizing machine‐learning methods alongside statistical modeling and economic evaluation to deliver an in‐depth assessment of the area's wind energy potential. The Weibull probability distribution was utilized to define the statistical characteristics of wind speeds and to determine the parameters necessary for calculating wind power density. The forecast for monthly mean wind speeds in 2025 was conducted using SARIMAX and Prophet models, resulting in a high level of predictive accuracy (R² = 0.85–0.98), which facilitates accurate estimation of wind energy capacity. This study combines machine‐learning forecasting, statistical methods, and economic analysis to offer a practical framework for evaluating the regional energy profile and enhancing turbine selection, thereby contributing to the sustainable development of wind farms in Khaf.