Artificial neural networks model for predicting oxygen content in flue gas of power plant

Abstract

The oxygen content in flue gas of power plant is one of the important variables for keeping the boiler combustion process stable and secure. Real-time monitoring and control for the oxygen content in flue gas of power plant is difficult at present. To address this problem, we propose a soft measurement method based on back-propagation neural network (BPNN) and genetic algorithm (GA) to predict the oxygen content in flue gas of power plant. In the algorithm, partial least squares (PLS) method is used to reduce dimensions of input variables. The model based on the data collected from the historical data of power plant is constructed by BP. The GA algorithm is utilized to optimize the parameters of BP for improving the accuracy of the model. The proposed method has been proved effective through iterative experiments.