Research On Judging Of Combustion Stability For Utility Boiler Based On Support Vector Machine

The boiler is an important burning equipment in a power plant. The combustion condition of the boiler has a direct effect to the safety, economy and environmental protections of the power plant. So the combustion stability of the boiler has become the focus studies in the few years. Firstly, the article summarized influence factors and the reflection factors of the combustion stability, then studied the relationship between furnace press and combustion stability. Secondly, this paper introduced statistical learning and support vector machine (SVM) theory. It is applied to the classification of coal combustion stability. Compared with traditional method, SVM classifier has better classification rate and advantages in selecting model, overcoming over-fitting and local minimum, etc. Based on the data from the boiler of certain 600MW boiler–turbine unit, the way to identify the flame combustion states using SVM from the factor of combustion stability is put forward. The further analysis suggests that SVM has good classification and generalization ability on judging of combustion stability for utility boiler.