The intelligent control and optimisation of power station boilers firing pulverised coal and coal/biomass blends (Smartburn) : final report

The Smartburn project was designed to increase the amount of biomass used in utility power stations. The project enabled this by focussing on the problem of combustion instability which can occur during the co-firing of coal and biomass and in its severest form can lead to flame extinction. As a result of poor combustion high emissions of CO or NOx may emanate from the burner. In order to co-fire higher concentrations of biomass a system was required that could provide warning of these conditions. The objective of the Smart burn project was the development of an innovative 'intelligent' monitoring, control and optimisation system to control coal/biomass fired burners. The system developed is based on the remote monitoring of individual flames using sensors that detect the spectral emissions of the combustion flame in the ultra-violet (UV), infra-red (IR) and visible (VIS) bands and the application of thermo-acoustic sensors was also investigated. Advanced signal processing was applied based on joint time-frequency methods and an Artificial Neural Network (ANN) was used to identify characteristics that could be related to operational parameters such as local air/fuel ratio, NOx emissions, and flame stability. The developed system was successfully tested at pilot scale (500 kWt) and on a full scale power plant (burner 20 MWt) where the ability of the system to monitor and optimise the combustion for a variety of unseen coal/biomass blends was demonstrated. It was also demonstrated how the system could be applied to balance air/fuel supply to multi-burner applications and detect potential flame out conditions.