Predicting Full Load, Partial Load Efficiency of a Combined Cycle Power Plant Using Machine Learning Methods

The generation of electricity takes place essentially via the power plants. The thermal efficiency of these plants depends on internal factors as much as it depends on the environmental ones. Hence, it becomes of utmost importance that these plants be monitored for maintaining their efficiency. Due to fluctuation in demands, the load at which the plant is operated also varies. In our work, we have taken a closer look at these loads which are generally divided into two parts: full load and part load. We have explored different machine learning algorithms ranging from regression methods to ensemble methods for predicting the thermal efficiency of the plant for various loads. Thirteen variables have been selected as inputs and the results illustrate that K-Nearest Neighbor performs the best in regression methods while Ada Boosting on Decision Tree performs best in ensemble methods.