Combining Support Vector Regression and Kernel Principal Component Analysis to Monitor NOx Emissions in Coal-Fired Utility Boiler

摘要

The formation of nitrogen oxides (NOx) associated with coal combustion systems is a significant pollutant source in the environment, and the monitoring of NOx emissions is an indispensable process for combustion optimization so as to control NOx emissions. In this paper, a hybrid model combining support vector regression (SVR) and kernel principal component (KPCA), named KPCA-SVR, was presented to map the complex and highly nonlinear relationship between the parameters of the boiler and the NOx emissions. The method was applied to a case boiler of 300MW steam capacity. The results showed that the hybrid model predicted NOx emissions much more accurate and certain than the widely-used BPNN model and simplex SVR model. This approach will be a good alternative and more suitable for its applicability in the actual power plants.