Optimization study of low-NO combustion in boiler based on pulverized coal preheating solution

The goal of this study is to develop a method to optimize the control of pulverized coal boiler combustion in order to reduce NO x emissions and improve the operating efficiency of the boiler. Firstly, the deep belief neural network (DBN) is used to extract the boiler combustion rules from the large data of the DCS system of the coal-fired power plant. Secondly, the deep belief neural network (DBN) and long short-term memory network (LSTM) were used to establish the combustion model of pulverized coal boiler based on the ignition temperature, outlet temperature, powder feeding frequency and other parameters, as well as the coupling relationship between NO x and CO emissions. Finally, considering the actual situation of coal-fired power plants, we used the beetle swarm optimization (BSO) algorithm on the basis of the three optimization objectives, and took the pyrolysis temperature and powder feeding frequency as the main input parameters of combustion optimization to optimize the combustion model. The results show that when the weights are set to α =0.70 and β =0.30, the boiler can be in the best operating condition and the NO x and CO emissions are reduced. In summary, the optimization method proposed in this study can not only reduce NO x emissions, but also improve combustion efficiency, which has certain reference value for combustion optimization of coal-fired power plants.