The integration of Internet of Things (IoT) technology for monitoring wind turbines has emerged as a pivotal advancement within the renewable energy sector. Wind energy has enormous potential to meet the rising global energy demand with high efficiency and reliability, while also reducing environmental deterioration. Industry 4.0, characterized by the convergence of IoT, AI, and data analytics, has the ability to revolutionize energy production. The authors explore the application of IoT-driven machine learning models for predictive maintenance of wind turbines, leveraging data from SCADA systems. The study employs a RandomForestRegressor to forecast wind turbine performance and a GradientBoostingRegressor to identify optimal energy generation hours. Data quality and relevance are ensured through data preprocessing, feature engineering, and exploratory data analysis. Results demonstrate a high predictive accuracy, with R-squared values of 0.973 and 0.946 for active power output and energy prediction models, respectively. Predictive maintenance models achieved an accuracy of 0.955, underscoring their effectiveness in minimizing downtime and reducing maintenance costs. The analysis identified wind speed as a critical factor in energy generation, with the optimal hour for energy production pinpointed at 20:00. This research highlights the transformative impact of IoT and machine learning in optimizing wind turbine operations, providing actionable insights for enhancing renewable energy efficiency and sustainability.