Journée des chercheurs en HE- projet Autodiag

1. Designing a tool to support fault diagnosis
and help with predictive maintenance of
HVAC electrical actuators
AUTODIAG
J. Vachaudez
23/11/2017
: :

2. Outline
Part 1 : Context
Part 2 : Data Extraction
Part 3 : Test Bench
Part 4 : Data Manipulation & Visualization
Part 5 : Imbalance Detection - A Machine Learning
Example
2 : :

3. Part 1
Context
3 : :

4. 1. Maintenance
1.1 Maintenance types
1.2 Predictive maintenance techniques
2. Measurement flow
3. Fault analysis
3.1 Fault types
3.2 Fault signature
4 : :

5. 1. Maintenance
1.1 Maintenance types
1.2 Predictive maintenance techniques
2. Measurement flow
3. Fault analysis
3.1 Fault types
3.2 Fault signature
5 : :

6. Maintenance types
Types of maintenance :
Reactive maintenance : Repair when it is broken
Preventive maintenance : Repair at regular intervals
Predictive maintenance : Follow the state of the equipement
All together . . .
Proactive maintenance : Operate in the best conditions
6 : :

7. Maintenance types
Types of maintenance :
Reactive maintenance : Repair when it is broken
Preventive maintenance : Repair at regular intervals
Predictive maintenance : Follow the state of the equipement
All together . . .
Proactive maintenance : Operate in the best conditions
6 : :

8. 1. Maintenance
1.1 Maintenance types
1.2 Predictive maintenance techniques
2. Measurement flow
3. Fault analysis
3.1 Fault types
3.2 Fault signature
7 : :

9. Predictive maintenance
techniques
Predictive maintenance techniques
• Vibration analysis
• Oil analysis
• Infrared thermography
• Ultrasonic analysis
• Other (electrical motor diagnosis, temperature measurement,
endoscopy, . . .)
8 : :

10. Predictive maintenance
techniques
Predictive maintenance techniques
• Vibration analysis
• Oil analysis
• Infrared thermography
• Ultrasonic analysis
• Other (electrical motor diagnosis, temperature measurement,
endoscopy, . . .)
8 : :

11. Unit conversion
x = X · sin(ωt) (1)
dx
dt
= v = ω · X · cos(ωt) (2)
dv
dt
= a = −ω2
· X · sin(ωt) (3)
• Displacement :
µm Pk − Pk
• Speed : mm/s RMS
• Acceleration : g s Pk
t
d
9 : :

12. Unit conversion
x = X · sin(ωt) (1)
dx
dt
= v = ω · X · cos(ωt) (2)
dv
dt
= a = −ω2
· X · sin(ωt) (3)
• Displacement :
µm Pk − Pk
• Speed : mm/s RMS
• Acceleration : g s Pk
t
d
9 : :

13. Unit conversion
x = X · sin(ωt) (1)
dx
dt
= v = ω · X · cos(ωt) (2)
dv
dt
= a = −ω2
· X · sin(ωt) (3)
• Displacement :
µm Pk − Pk
• Speed : mm/s RMS
• Acceleration : g s Pk
t
d
9 : :

14. Units – Order
Définition
The order number may be defined by the quotient of the analyzed
frequency and the reference shaft rotation frequency.
#ordre =
Fpeak[Hz]
Frot[Hz]
• The order number is dimensionless but often represents as a X
• It represents the number of times that the phenomenon occurs per
revolution
10 : :

15. Units – Order
Définition
The order number may be defined by the quotient of the analyzed
frequency and the reference shaft rotation frequency.
#ordre =
Fpeak[Hz]
Frot[Hz]
• The order number is dimensionless but often represents as a X
• It represents the number of times that the phenomenon occurs per
revolution
10 : :

16. 1. Maintenance
1.1 Maintenance types
1.2 Predictive maintenance techniques
2. Measurement flow
3. Fault analysis
3.1 Fault types
3.2 Fault signature
11 : :

17. Measurement flow
Step 1 : Transfer database from computer to collector
12 : :

18. Measurement flow
Step 2 : Measure equipments on customer site
12 : :

19. Measurement flow
Step 3 : Transfer data from collector to computer
12 : :

20. Measurement flow
Step 4 : Analyze data and send report to the customer
12 : :

21. 1. Maintenance
1.1 Maintenance types
1.2 Predictive maintenance techniques
2. Measurement flow
3. Fault analysis
3.1 Fault types
3.2 Fault signature
13 : :

22. 1. Maintenance
1.1 Maintenance types
1.2 Predictive maintenance techniques
2. Measurement flow
3. Fault analysis
3.1 Fault types
3.2 Fault signature
14 : :

23. Fault types
Detectable faults
• Imbalance
• Misalignement
• Looseness
• Bearing (BPFO, BPFI, FTF, BSF)
• Gear
• Belt
• Lubrication
• Cavitation
• . . .
15 : :

24. Fault types
Detectable faults
• Imbalance
• Misalignement
• Looseness
• Bearing (BPFO, BPFI, FTF, BSF)
• Gear
• Belt
• Lubrication
• Cavitation
• . . .
15 : :

25. 1. Maintenance
1.1 Maintenance types
1.2 Predictive maintenance techniques
2. Measurement flow
3. Fault analysis
3.1 Fault types
3.2 Fault signature
16 : :

26. Imbalance
Figure 1 – Static imbalance
f
d
1
Figure 2 – Spectral signature
17 : :

27. Bearing fault
Outter race fault - BPFO
Figure 3 – Outter race fault
Fe 2Fe
1
3Fe
f
d
Figure 4 – Spectral signature
18 : :

28. Belt fault
Figure 5 – Belt/pulleys
Fc 2Fc
f
d
Figure 6 – Spectral signature
19 : :

29. Part 2
Data Extraction
20 : :

30. 4. Introduction
5. Fault Frequencies Database Extraction
6. Measurements Database Extraction
7. Reports Database Extraction
8. Equipment Database
8.1 GSK Dashboard
8.2 Equipment database creation & modification
21 : :

31. Introduction
MHM .txt files MongoDB Autodiag
Extract mea-
surements
Store mea-
surements
Query stored
measurements
22 : :

32. Introduction
MongoDB Pre-Processing
Featurevector
Classifier
Decision
23 : :

33. 4. Introduction
5. Fault Frequencies Database Extraction
6. Measurements Database Extraction
7. Reports Database Extraction
8. Equipment Database
8.1 GSK Dashboard
8.2 Equipment database creation & modification
24 : :

34. Bearing Database Extraction
Figure 7 – Bearing text file
25 : :

35. 4. Introduction
5. Fault Frequencies Database Extraction
6. Measurements Database Extraction
7. Reports Database Extraction
8. Equipment Database
8.1 GSK Dashboard
8.2 Equipment database creation & modification
26 : :

36. Measurements Database Extraction
Need to create a tool, able to extract all measurements from the MHM
database, one at a time.
Results
• ≈ 10 days of extraction 24/7
• ≈ 15000 text files
• ≈ 8 GB of text files
27 : :

37. Measurements Database Extraction
Need to create a tool, able to extract all measurements from the MHM
database, one at a time.
Results
• ≈ 10 days of extraction 24/7
• ≈ 15000 text files
• ≈ 8 GB of text files
27 : :

38. 4. Introduction
5. Fault Frequencies Database Extraction
6. Measurements Database Extraction
7. Reports Database Extraction
8. Equipment Database
8.1 GSK Dashboard
8.2 Equipment database creation & modification
28 : :

39. Reports Database Extraction
Figure 8 – Report text file
29 : :

40. 4. Introduction
5. Fault Frequencies Database Extraction
6. Measurements Database Extraction
7. Reports Database Extraction
8. Equipment Database
8.1 GSK Dashboard
8.2 Equipment database creation & modification
30 : :

41. 4. Introduction
5. Fault Frequencies Database Extraction
6. Measurements Database Extraction
7. Reports Database Extraction
8. Equipment Database
8.1 GSK Dashboard
8.2 Equipment database creation & modification
31 : :

42. Equipment Database
GSK Dashboard
Access to a file from GSK containing :
• Informations about bearing ;
• Informations about belts ;
• Only for WN27.
32 : :

43. 4. Introduction
5. Fault Frequencies Database Extraction
6. Measurements Database Extraction
7. Reports Database Extraction
8. Equipment Database
8.1 GSK Dashboard
8.2 Equipment database creation & modification
33 : :

44. Equipment Database
Database creation
Needs
• Create an equipment database ;
• Bearing and belt informations for each equipments ;
• Other informations about equipment can be useful (criticity,
type,. . .)
• Need to easily edit the equipment database.
34 : :

45. Equipment Database
Database creation
Needs
• Create an equipment database ;
• Bearing and belt informations for each equipments ;
• Other informations about equipment can be useful (criticity,
type,. . .)
• Need to easily edit the equipment database.
Figure 9 – Equipments sheet header
34 : :

46. Equipment Database
Database creation
Needs
• Create an equipment database ;
• Bearing and belt informations for each equipments ;
• Other informations about equipment can be useful (criticity,
type,. . .)
• Need to easily edit the equipment database.
(a) Manufacturer (b) Types
Figure 9 – Manufacturer and Types drop list
34 : :

47. Part 3
Test Bench
35 : :

48. Test Bench - Instumentation
Figure 10 – Test Bench - Instumentation
36 : :

49. Part 4
Data Manipulation & Visualization
37 : :

50. 9. Data Visualization
10. Results
38 : :

51. Data Visualization
Development of a Bokeh application in order to visualize data.
Goal
• Select data based on :
Geographic location (customer, area, equipment)
Fault diagnose in the report
Type of measurement
• Compute indicators on each selected measurement
• Visualize data
39 : :

52. Data Visualization
1. Dataset selection.
Figure 11 – Dataset selection
40 : :

53. Data Visualization
2. Set required parameters to indicators.
(a) Filtering options (b) Global Value options (c) Entropy options
Figure 11 – Indicators options
40 : :

54. Data Visualization
3. Select indicators to display.
Figure 11 – Indicators selection
40 : :

55. 9. Data Visualization
10. Results
41 : :

56. Results
Relevant indicator - Example
Figure 12 – Global Value computed, between 0.95 and 1.05 orders, for all
equipments of area WN10
42 : :

57. Results
Irrelevant indicator - Example
Figure 12 – Kurtosis computed for all equipments of area WN10
42 : :

58. Results
(a) "NO FAULT" measurements (b) "IMBALANCE" measurements
Figure 13 – Comparison between "NO FAULT" measurements and
"IMBALANCE" measurements
43 : :

59. Part 5
Imbalance Detection
A Machine Learning Example
44 : :

60. Supervised learning
SVM
y
x
Figure 14 – Support Vector
Machines
Caracteristics
• Binary classifier
• Find the best separating
hyperplane that separates
the data
45 : :

61. Imbalance Detection
SVM results imbalance detection
SVM on WN10 : 45 < recognition rate < 92%
SVM on WN27 : 45 < recognition rate < 95%
SVM WN27 applied on WN10 : recognition rate ≈ 85%
46 : :

62. Imbalance Detection
SVM results imbalance detection
SVM on WN10 : 45 < recognition rate < 92%
SVM on WN27 : 45 < recognition rate < 95%
SVM WN27 applied on WN10 : recognition rate ≈ 85%
46 : :

63. Thank you for your attention.
47 : :