Research Article

Time-Frequency Fault Feature Extraction for Rolling Bearing Based on the Tensor Manifold Method

Table 3

Classification results of the inner-race fault using the PNN with the proposed tensor manifold features.

Sample Category vector of the normal-state samples Category vector of the inner-race mild-damage
fault samples
Category vector of the inner-race moderate-damage
fault samples
Category vector of the inner-race severe-damage
fault samples

1100000.970.010.0200.010.980.0100.010.020.97
2100000.960.020.0200.020.970.0100.020.020.96
3100000.980.010.0100.010.970.0200.030.020.95
4100000.950.020.0300.030.950.0200.020.020.96
5100000.960.020.0200.010.980.0100.010.030.96
6100000.970.010.0200.010.960.0300.050.020.93
7100000.940.050.0200.020.950.0300.030.020.95
8100000.950.030.0200.020.970.0100.010.010.98
9100000.940.020.0400.040.940.0200.030.030.94
10100000.930.020.0500.040.930.0300.030.010.96