Information Technology Reference
In-Depth Information
at 6 o
'
clock and outer raceway fault located at 12 of
'
clock of drive end bearing affect
the contributions at the same scales (Figs.
9
a
b), but the contribution amplitudes are
different so they can be used to diagnose the faults.
In order to diagnose each motor fault LDA is used. It searches a linear trans-
formation that maximizes class separability in a reduced dimensional space. LDA is
proposed in Fisher (
1936
) for solving binary class problems. It is further extended
to multi-class cases in Rao (
1948
). In general, LDA aims to
-
nd a subspace that
minimizes the within-class scatter and maximizes the between-class scatter simul-
taneously. PCA contributions are used as features input to the LDA algorithm.
Tables
3
,
4
and
5
show the classi
cation accuracy. The results are the average
classi
cation accuracy of each motor conditions (i.e. healthy motor, rolling element
fault, inner raceway fault, outer raceway fault located at 3, 6 and 12 of
'
clock) and of
4 motor loads: from 0 to 3 hp (motor speeds of 1,797
-
1,720 RPM). Table
3
shows
the classi
cation accuracy at different wavelet decomposition level L and acquisi-
tion time of faults occurred at drive end bearing with fault diameter of 0.007 in. The
classi
cation accuracy is over 99 % for each level L and acquisition time, so a low
wavelet decomposition level and acquisition time can be chosen to diagnose
effectively this fault.
Table
4
shows the classi
cation accuracy at different wavelet decomposition
level L and acquisition time of faults occurred at drive end bearing with fault
diameter of 0.021 in. The classification accuracy is over 99 % at each level L and
acquisition time higher 0.3 s, so a low wavelet decomposition level and acquisition
Table 3 Average
classi
cation accuracy of
drive end bearing fault
with fault diameter of
0.007 in.
Acquisition time s
0.1
0.2
0.3
0.5
0.7
%
L
1
99.23
99.96
99.98
100
100
2
99.40
99.85
99.99
100
100
3
99.55
99.95
100
99.99
100
4
99.67
99.98
100
100
100
5
99.47
99.95
100
100
100
6
99.29
99.85
99.99
100
100
Table 4 Average
classi
cation accuracy of
drive end bearing fault
with fault diameter of
0.021 in.
Acquisition time s
0.1
0.2
0.3
0.5
0.7
%
L
1
92.66
97.73
99.19
99.91
99.96
2
94.76
98.36
99.51
99.75
99.88
3
95.10
98.87
99.73
99.98
99.99
4
94.64
98.39
99.21
99.81
99.97
5
95.00
98.54
99.50
99.79
99.96
6
94.38
98.25
99.35
99.83
99.88
Search WWH ::
Custom Search