Predictive fuzzy controller for bed temperature system of CFB boiler based on multi-sensor information fusion algorithm

Bed temperature of circulating fluidized bed (CFB) boiler is measured by multiple sensors equally fixed around the furnace wall. Big inertia of the combustion process and the large time delay of bed temperature measure are unfavorable for thermal process control. Based on the redundant distributed measuring property of bed temperature, a new sensor signal synthesis method was proposed. The new method is based on information fusion technique, which uses multi-sensor resource to decrease the measure uncertainty, eliminate the failure sensor/sensors, so the measure result is more credible. That is very important for optimal control of combustion system. Grey modeling theory and fuzzy control strategy are combined to control the combustion process. Numerical simulation results of bed temperature measure and grey fuzzy control for a CFB boiler show that the proposed new strategy is feasible and effective.