Bayesian Method Used In Optimal Operation Of Large Power Station Boilers

Energy conservation has become more and more important with the increase of energy consumption and environmental protection needs.Improving the efficiency of coal-fired units, reducing the emission of pollutants is the important task of structural adjustment. NOx is the major pollutants produced by coal-fired boilers. High of content of fly ash is the main reason of low efficiency of coal-fired units. The coal and the operation parameters have a great influence on the emission of the boiler, carbon content of the fly ash. And it is hard to express by conventional way. As the reason of that, Bayesian approach was introduced to build the model. The factors which affect the target were introduced as the input, NOx emissions and carbon content were introduced as the output. The plant samples were used as training samples.RKHS BR、RKHS BL and RKHS were used to build the model of the NOx emission of the300MW boiler. Field data was used to train and validate the model. Meanwhile, colony algorithm was introduced to establish the optimal model.In addition, the119cases which were chose from the670cases were selected to build the model.The model was compared with the SVM method. The result showed that RKHS method was better than SVM method.