Research Article

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

Table 2

Classification results of the four bearing faults using the PNN with traditional HHT time-frequency features.

Sample Category vector of the normal-state samples Category vector of the inner-race fault samples Category vector of the
ball fault samples
Category vector of the outer-race fault samples

1100000.900.050.0500.100.890.0100.060.070.83
2100000.920.030.0500.120.870.0100.050.060.89
3100000.890.050.0600.120.790.0900.070.080.85
4100000.900.060.0400.130.850.1200.050.040.91
5100000.870.060.0700.100.780.1200.060.080.86
6100000.920.040.0800.130.760.1100.150.060.79
7100000.910.050.0400.070.870.0600.060.070.87
8100000.820.080.100.020.930.0500.150.120.73
9100000.870.080.0500.080.850.0700.070.080.75
10100000.930.050.0200.070.870.0600.050.070.88
11100000.690.140.1700.010.880.1100.060.080.86
12100000.740.120.1400.200.790.0100.070.080.85
13100000.800.120.0800.080.880.0400.030.060.91
14100000.860.090.0500.070.870.0600.070.040.89
15100000.750.230.0200.060.890.0500.080.090.83
16100000.780.100.1200.030.930.0200.090.100.81
17100000.860.120.1200.060.870.0700.070.040.89
18100000.870.060.0700.090.890.0200.030.080.89
19100000.830.130.1400.020.910.0700.070.010.83
20100000.920.050.0300.050.920.0300.070.110.82