Fault Diagnosis of Rotating Machinery Based on FDR Feature Selection Algorithm and SVM
To effectively avoid the loss of useful information, in this paper, feature information has been extracted from the fault signal of rotating machinery in different aspects such as amplitude-domain, time-domain and time-frequency domain. Then, for the multi-dimensional feature extraction was prone to the problem of “dimension disaster”, the principles of FDR was introduced in data mining to determine the classification ability of each individual feature, and the cross correlation coefficient was adopted to solve the problem that dealing with individual feature. Neglected the interrelationship between the features, a new feature selection algorithm was constructed. Finally, the eigenvectors were used for training and recognizing of SVM model. The experimental results showed the fault diagnosis system was valid and robust.
Liangchi Zhang, Chunliang Zhang and Tielin Shi
S. Li et al., "Fault Diagnosis of Rotating Machinery Based on FDR Feature Selection Algorithm and SVM", Advanced Materials Research, Vols. 139-141, pp. 2506-2512, 2010