Prediction of gaseous emissions from industrial stacks using an artificial intelligence method

Abstract

AbstractA novel technique based on artificial intelligence methods able to predict pollutant emission concentrations from industrial stacks is presented. This procedure combines regression and classification problems into a unified technique, named minimax decision procedure. The core of this procedure is based on the minimax probability machine regression model. Using experimental databases, the trend of pollutant emissions and the level of pollution for one industrial thermal power station stack were presented. Based on this unified technique, numerical experiments provided the estimates of concentrations of CO, NOx, NO, and SO2 confirming the predictive power of this procedure.