A System For Forecasting NO_x Emission And The Unburned Carbon Of The Fly Ash Based On Genetic Algorithm And BP Neural Network

To meet the requirement of high efficiency and low emission, the Pulverized Coal-fired Boiler combustion optimization method combining genetic algorithm and BP neural network was researched. Taking twelve kinds of influencing factors including load of boiler, volatile matter, oxygen content in flue gas, flue gas temperature, net calorific value, carbon and so on as conjunction points of input layer, NOx emission and the unburned carbon content as conjunction points of output layer. NOX emission and the unburned carbon content of the Pulverized Coal-fired Boiler is forecasted using genetic algorithm and BP neural network. Analysising of the results predicted showed that the present method in convergence speed and precision are in line with requirements of on-line monitoring.