Fig 1.
The proposed fault diagnosis scheme based on image processing.
Fig 2.
Approximated second order derivatives with box filters.
Fig 3.
Filters Dxy for two successive scale levels (9×9 and 15×15).
Fig 4.
Haar wavelet filters.
Fig 5.
Determining the main direction of the feature point.
Fig 6.
Basic structure of PNN.
Fig 7.
Self-priming centrifugal pump data acquisition system.
Table 1.
Experiment items of centrifugal pumps fault injection.
Fig 8.
Bi-spectrum counter maps under bearing roller wearing fault condition.
Fig 9.
Bi-spectrum counter maps under bearing inner race wearing fault condition.
Fig 10.
Bi-spectrum counter maps under bearing outer race wearing fault condition.
Fig 11.
Bi-spectrum counter maps under normal condition.
Fig 12.
Bi-spectrum counter maps under impeller wearing fault condition.
Fig 13.
The first three features extracted using t-SNE (a) and without using t-SNE (b).
Table 2.
The data composition of the self-priming centrifugal pump for cross validation.
Fig 14.
Results of 4 groups of cross validation.
Table 3.
The error rate of 4 groups of cross-validation.
Fig 15.
Axial piston hydraulic pump system.
Fig 16.
Bi-spectrum counter map of normal.
Fig 17.
Bi-spectrum counter map of valve plate wearing.
Fig 18.
Bi-spectrum counter map of piston shoes and swashplate wearing.
Fig 19.
The first three features extracted using t-SNE (a) and without using t-SNE (b).
Table 4.
The data composition of axial piston hydraulic pump for cross validation.
Fig 20.
Result of 4 set cross validation.
Table 5.
The error rate of 4 groups of cross-validation.