A Comparative Study for CO and NOx Emission Forecasting Using Random Forest and Decision Tree Methods

Abstract

In this study, CO and NOx (NO + NO2) gas emission forecast of a gas turbine located in Turkey's north western region is done with machine learning techniques. These techniques are decision tree (DT) and random forest techniques (RF).

At first, the best performed architecture in each technique is determined for CO and NOx emission forecasting. Then, the technique which gives the best solution is determined among these techniques. Results show that, Random Forest gives the best result among these techniques.

Keywords: Machine learning techniques, prediction, decision tree, random forest