A Calculation Method for CO2 Emission in Utility Boilers Based on BP Neural Network and Carbon Balance

Abstract

For the accurate calculation of CO2 emission in utility boilers, the calculation model of CO2 emission suitable for most coal-fired boilers is built. In this model, BP neural networks were adopted to predict the carbon content of burning coal. This model took the proximate analysis data and fuel consumption flow / coal consumption as inputs, and output the CO2 emission. Under the BP network prediction model, the average difference of carbon content of burning coal was about 1.80%, while that of CO2 emission were 3.92% and 0.96%. Compared with emission factor method and capacity standard method, the method in this paper is more suitable to count the CO2 emission of utility boilers.