Soft sensing technology based on PCA and SVM for coal powder amount in medium speed mill

For the accurate measurement of the amount of coal powder, this paper proposed the soft sensor method. This method combined principal component analysis (PCA) and support vector machine (SVM), then established the model of the amount of coal powder. The method uses principal component analysis to compress the modeling data, which can reduce the modeling difficulty of support vector machine. The actual operation data in the power plant verifies that the soft sensor method could effectively track the change trend of the amount of coal powder. Its calculation is simple, and has a better promotional and application value.