Neural control for a solid waste incinerator

TLDR: In this work, a neural control scheme to regulate carbon monoxide and nitrogen oxides emissions for a solid waste incinerator is proposed and a recurrent high order neural network (RHONN) is proposed, which is trained with an extended Kaiman filter (EKF) algorithm.

Abstract: In this work, a neural control scheme to regulate carbon monoxide (CO) and nitrogen oxides (NO x ) emissions for a solid waste incinerator is proposed. Carbon monoxide emissions are avoided by oxygen regulation in the incinerator; nevertheless nitrogen oxides emissions are difficult to control because the sludge composition varies continuously. The air flow is selected to be the control input because it have a great influence in CO and NO x formation. The air flow can guarantee a complete combustion and therefore, a good incineration quality because it avoids pollutant formation. In order to obtain the sludge combustion model, it is proposed to use a recurrent high order neural network (RHONN), which is trained with an extended Kaiman filter (EKF) algorithm. The proposed neural controler performance is illustrated via simulations.