Fault Process Trend Prediction Based On Intelligengt Controll For Anode Backing Furnace

Statistical Learning Theory is based on the theory foundation and provides a uniform framework for learning subject of limited samples. Support Vector Machines (SVM) is a machine-learning algorithm based on statistical learning theory. This algorithm accomplishes the structural risk minimization principle.The fine performance of Support Vector Machines to limited samples attracts attention of investigators in fault diagnosis field. Fault diagnosis is a limited samples subject. The most predominance of SVM is proper for limited samples decision. The nature of the algorithm is acquiring connotative class information to great extent from limited samples. From the point of generalization, SVM is the more favorable for the practical engineering problem as fault diagnosis.The support vector machines method for predicting fault of the anode baking furnace is introduced.A correcting nonlinear errors of the thermocouple sensor based on Radial Basis Function Neural Network using particle swarm optimization is introduced. It solves the shortcoming of Thermocouple Sensor's application on large data. The result of experiment shows that the nonlinear calibration based on RBF-PSO has higher precision than the method based on BP and RBF. Then, PSO-RBF is used to test fire-path temperature in the anode baking. It is proved that the method is effective.The Wavelet Neural Network using particle swarm optimization for pressure transducer is introduced. The result of experiment shows that the nonlinear calibration based on WNN-PSO has higher precision.The study and the conclusion of this study have the important meanings to the predict fault of anode baking furnace based on SVM.