Optimization of Coal-fired Boiler on LS-SVM Model and PSO Algorithms

With the deterioration of dust storm in recent years, environmental protection has become a topic for everyone. The most important problem of environmental protection comes from coal combustion which can be improved by combustion optimization. And combustion optimization has been proved to be an effective way to reduce the Nox emissions and improve boiler combustion efficiency by setting the operating parameters. The aim of this work is to achieve optimization of the coal-fired boiler by least square-support vector machine (LS-SVM) model and PSO algorithm. In this paper, LS-SVM was applied to build Nox emissions model, carbon content of fly ash model and flue gas temperature model. Thereafter, based on the above models, we select PSO algorithm and GA to solve the problem. The results of the experiment demonstrate that PSO algorithm is superior to GA and it is effective on improving boiler's efficiency and reducing Nox emissions.