Variable Selections in Machine Learning Based Efficiency Estimation of the Combined Cycle Power Plant

Abstract

Combined cycle power plants (CCPP) are the top contenders in the electricity generation area. Not only are they highly efficient but they also use less fuel and produce fewer emissions than their counterparts. For their better utilization, in our earlier works, we carried out experiments to determine the efficiency of a CCPP with the help of machine learning models and therefore selected the top two models for the current paper. However, accurate predictions involve choosing the right parameters. In this paper, we took upon the task of investigating which parameters are extremely necessary for good efficiency prediction. This paper consists of the examination of thirteen variables, ranging from internal to environmental factors with the Adaptive Boost and Gradient Boosting model. By the end of the experiment, we found that inlet guide vanes, gas turbine, steam turbine, cooling tower, and megawatt produced were among the top priority variables.