On fault feature extraction and diagnosis of vertical mill

Abstract

In order to solve the problems of complicated fault diagnosis and poor fault diagnosis of vertical mill operation, this paper proposes a diagnostic method based on fisher and information entropy difference classification. By extracting the fault feature of the anomaly attribute–the maximum value of the attribute, and the possible faults can be determined according to the fault characteristics. Then the information entropy of each sample is calculated, and the entropy difference between normal and fault states is calculated. The normal and fault conditions can be classified by fisher classifier. This method can capture the instantaneous change of the fault and detect the moment when the fault occurs. And the effectiveness of the feature extraction method is verified by experiments.