Identification of Key Parameters Influencing Load Progression During Startup in Combined Cycle Power Plants (CCPP): Implications for Smart Grids

This paper investigates integrating advanced analytics techniques into smart grids to enhance energy system efficiency and reliability. It focuses on identifying key operational parameters influencing load progression during startup processes in combined cycle power plants (CCPPs). Using a dataset categorized into different startup classes (Hot, Warm 1, Warm 2, Cold), it employs feature selection techniques such as PCC, LASSO regularization, and RF-RFE to identify influential parameters. Predictive modeling evaluates each technique's efficiency in capturing system dynamics, revealing variations in predictive performance across startup classes. The study highlights crucial variables contributing significantly to load progression and discusses implications for smart grid applications like predictive maintenance and load prediction. It advances understanding of CCPP operation within smart grids, providing insights for enhancing energy system intelligence and resilience.