Development of a Predictive Maintenance Framework for Combined-Cycle Turbines Using Real-Time Sensor Data and Machine Learning

The reliability and efficiency of combined-cycle turbines are critical to ensuring optimal energy production and reducing operational costs. Unscheduled failures and maintenance activities can lead to significant financial losses and reduced system performance. This study presents a predictive maintenance framework that leverages real-time sensor data and machine learning to enhance the operational efficiency of combined-cycle turbines. By continuously monitoring key turbine parameters, the proposed framework enables early fault detection and failure prediction, minimizing downtime and maintenance costs. The integration of machine learning techniques allows for data-driven decision-making, improving the accuracy of failure forecasts and optimizing maintenance schedules. Unlike traditional reactive and preventive maintenance strategies, predictive maintenance enhances asset longevity and operational stability by addressing potential issues before they escalate into major failures. This framework contributes to the advancement of intelligent maintenance solutions in the energy sector, promoting sustainable and cost-effective turbine operations. The findings highlight the potential of predictive analytics in transforming turbine maintenance strategies, ensuring higher efficiency, reliability, and economic viability in power generation systems.