Modeling and Optimization of Efficiency and NOx Emission at a Coal-Fired Utility Boiler

Abstract

In order to improve boiler efficiency and to reduce the NOx emission of a coal-fired utility boiler using combustion optimization, a hybrid model, by combining support vector regression (SVR) with simplified boiler efficiency model, was proposed to express the relation between operational parameters of the utility boiler and both NOx emission and boiler efficiency. SVR' parameters were determined by the grid search method and 5-fold cross validation method. The predicted NOx emission and boiler efficiency from the hybrid model, compared with that of the BPNN-based hybrid model, shows better agreement with the measured. Then, based on the hybrid model, the modified center particle swarm optimization (CenterPSO) was employed to optimize the two objectives, the one is minimization of NOx emission and maximization of boiler efficiency and the other one is maximization of boiler efficiency under NOx emission constraint. The optimized results indicate that the proposed method can effectively control NOx emission and improve boiler efficiency.