Estimation of chosen combustion parameters of an individual pulverized coal burner based on optical signals

There are several tens of burners operating in a power boiler and their control based on averaged and delayed measurements (e.g. gas analyzers located inside the chimney) is often not enough effective. The article describes attempts to obtain information about levels of emission of nitrogen oxides and carbon monoxide from the signals of the fiber-optic flame monitoring system developed in the Department of Electronics of the Technical University of Lublin. Artificial neural networks are used for estimation of emission. The article contains description of an object and the measurement system as well as results of research of nitrogen oxides and carbon monoxide modeling.