This paper investigates the wind energy potential in Markazi Province, Iran, focusing on three cities: Tafresh, Khomein and Saveh. The primary objective of this study is to provide a comprehensive analysis of wind patterns using a combination of statistical approaches and artificial intelligence techniques. Wind data was collected from advanced meteorological stations in these cities over a two-year period (2018–2020), including detailed measurements of wind speed and direction at 10-minute intervals. This high-resolution dataset facilitated an in-depth examination of wind behavior and its suitability for energy production. Statistical analysis was conducted using the Weibull distribution and wind rose diagrams, which provided insights into the wind characteristics and their spatial variations. Additionally, Long Short-Term Memory (LSTM) networks were employed to predict wind speeds and temporal trends. These models effectively captured the complex relationships within the wind data and produced highly accurate forecasts. The comparison of actual and predicted wind rose diagrams demonstrated a strong alignment, validating the reliability of the LSTM-based predictions in reflecting real-world wind patterns. The results of this study demonstrate that combining traditional statistical methods with modern machine learning techniques provides a robust framework for analyzing wind energy potential. By leveraging these tools, the study offers valuable insights for sustainable energy planning and supports informed decision-making for renewable energy investments in Markazi Province.
