Fault Diagnosis of Lead-Zinc Smelting Furnace based on Multi-Class Support Vector Machines

Support vector machine (SVM) is powerful for the problem with small sampling, nonlinear and high dimension. A multi-class SVM classifier is applied to fault diagnosis of imperial smelting furnace in this paper. The input data is preprocessed through a special method based on the data reliability analysis technology, and six features are extracted as the input to multiple fault classifier for identify faults, which adapt an improved ´one to others´ algorithm. The real application results show that the classifier has an excellent performance on training speed and reliability.