A model to predict ammonia emission using a modified genetic artificial neural network: Analyzing cement mixed with fly ash from a coal-fired power plant

Abstract

Selective catalytic reduction (SCR) and selected non-catalytic reduction (SNCR) use ammonia as a catalyst for removing nitrogen oxides in power plants. However, unreacted ammonia in the flue gas denitrification system reacts chemically with SO3 and is adsorbed on the coal fly ash. Unfortunately, fly ash is commonly mixed with concrete because of its technological and economic benefits, without consideration of secondary environmental contamination. We measured the ammonia emissions from different types of fly ash mortar and found they have a strong correlation with the mixing ratio of fly ash, the mortar age, and the size. In order to find out this correlation and thus predict the ammonia concentration under different conditions, we constructed artificial neural network (ANN) models. The comparison results show that the ANN model initialized with the genetic ANN (GANN) algorithm has the smallest root-mean-square error (RMSE) between given and predicted outputs.