Detection Method Using FBG Sensing Signal to Diagnose Rolling Bearing Fault

Chen Yong; An Wangyue; Liu Huanlin; Chen Yawu

doi:10.3788/CJL202047.1104004

Chinese Journal of Lasers, Volume. 47, Issue 11, 1104004(2020)

Detection Method Using FBG Sensing Signal to Diagnose Rolling Bearing Fault

Chen Yong^1、*, An Wangyue¹, Liu Huanlin², and Chen Yawu¹

¹Key Laboratory of Industrial Internet of Things & Network Control, Ministry of Education, Chongqing University of Posts and Telecommunications, Chongqing 400065, China

²Key Laboratory of Optical Fiber Communication Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(14)

Fig. 1. Improved CNN structure

Download full size

Fig. 2. Flowchart of decoupling classification algorithm

Download full size

Fig. 3. Fault diagnose process

Download full size

Fig. 4. FBG bearing failure detection experimental platform. (a) Bearing fault detection platform; (b) FBG installation location

Download full size

Fig. 5. Different faulty bearings. (a) Outer fault; (b) inner fault; (c) roller fault; (d) outer & inner fault; (e) outer & roller fault; (f) inner & roller fault

Download full size

Fig. 6. Flowchart of rolling bearing fault diagnosis

Download full size

Fig. 7. Time-domain signal diagrams of different faulty bearings. (a) Normal; (b) outer fault; (c) inner fault; (d) roller fault; (e) outer & inner fault; (f) outer & roller fault; (g) inner & roller fault

Download full size

Fig. 8. IMF component in outer fault

Download full size

Fig. 9. Schematic of IMF combined into a matrix

Download full size

Fig. 10. Multi-class confusion matrix of different methods. (a) Traditional convolutional neural network; (b) proposed method

Download full size

Table 1. Parameters of 6203 bearing
View table
Table 1. Parameters of 6203 bearing
Parameter Inside diameter /mm Outside diameter /mm Width /mm Weight /kg Number of balls
Value 17 40 12 0.065 8

Table 2. Numerical table of evaluation indicators under outer fault
View table
Table 2. Numerical table of evaluation indicators under outer fault
Indicator IMF1 IMF2 IMF3 IMF4 IMF5 IMF6 Mean value
C_Pearson 0.0186 0.8362 0.8148 -0.0019 -0.0235 0.0028 0.2745
K 33.1020 1.3698 1.3160 4.3158 4.1448 1.8571 1.6594
C_index 0.6157 1.1454 1.0723 -0.0082 -0.0974 0.0052 0.4555

Table 3. Numerical table of comprehensive evaluation indicator of each components under other faults

View table

Table 3. Numerical table of comprehensive evaluation indicator of each components under other faults

Fault type	IMF1	IMF2	IMF3	IMF4	IMF5	IMF6	Mean value
Inner fault	0.5224	1.0962	1.1458	-0.0079	0.0529	-0.0036	0.4676
Roller fault	0.6391	1.2017	0.8473	-0.0198	0.0753	0.0045	0.4580
Outer & inner fault	0.5736	1.0145	0.9434	0.0028	-0.0974	0.0056	0.4071
Outer & roller fault	0.6419	1.1256	1.0593	0.1064	-0.0632	0.0130	0.4805
Inner & roller fault	0.7150	1.0127	0.8723	0.0974	-0.0152	0.0043	0.4478

Table 4. Size of each layer of improved convolutional neural network
View table
Table 4. Size of each layer of improved convolutional neural network
Layer name Size
Input 4096×3×1
Conv1 256×3×1
Pool1 128×3×1
Conv2 128×3×1
Pool2 64×3×1
Epoch 20
Batch size 64

Tools

Get Citation

Copy Citation Text

Chen Yong, An Wangyue, Liu Huanlin, Chen Yawu. Detection Method Using FBG Sensing Signal to Diagnose Rolling Bearing Fault[J]. Chinese Journal of Lasers, 2020, 47(11): 1104004

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites