Abstract
This paper presents the development of the prediction algorithm of lubricant degradation based on Beer Lambert's transmittance theory by using Multiple Linear Regressions (MLR). Recently, an increasing amount of wasted lubricant has been due to the unnecessary changing of lubricant even though the lubricant still remains its lubrication behavior. Therefore, a condition based technique is introduced to monitor the degradation parameters in lubricating oil by using optical approach. This work focuses on Total Acid Number (TAN) that has been identified as the main parameter in determining the lifetime of lubricant and it occurred at band location from 1,050-1,250cm-1 and 1,700-1,730cm-1. The best input parameter has been identified for sensor development and signal processing. Then, the prediction model is used to validate the measured and the predicted value of degradation. The high correlation between the predicted and measured data shows the prediction algorithm can be used for prediction purposes efficiently.
| Original language | English |
|---|---|
| Title of host publication | International Conference on Electronic Devices, Systems, and Applications |
| Pages | 194-197 |
| Number of pages | 4 |
| DOIs | |
| Publication status | Published - 2012 |
| Event | IEEE International Conference on Electronics Design, Systems and Applications, ICEDSA 2012 - Kuala Lumpur Duration: 5 Nov 2012 → 6 Nov 2012 |
Other
| Other | IEEE International Conference on Electronics Design, Systems and Applications, ICEDSA 2012 |
|---|---|
| City | Kuala Lumpur |
| Period | 5/11/12 → 6/11/12 |
Fingerprint
Keywords
- Degradation
- Lubricant
- Multiple Linear regression (MLR)
ASJC Scopus subject areas
- Computer Science Applications
- Hardware and Architecture
- Software
- Electrical and Electronic Engineering
Cite this
Implementation of multiple linear regressions in lubricant degradation prediction algorithm. / Idros, M. F M; Manut, Azrif; Yahya, R.; Md Ali, Sawal Hamid.
International Conference on Electronic Devices, Systems, and Applications. 2012. p. 194-197 6507795.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Implementation of multiple linear regressions in lubricant degradation prediction algorithm
AU - Idros, M. F M
AU - Manut, Azrif
AU - Yahya, R.
AU - Md Ali, Sawal Hamid
PY - 2012
Y1 - 2012
N2 - This paper presents the development of the prediction algorithm of lubricant degradation based on Beer Lambert's transmittance theory by using Multiple Linear Regressions (MLR). Recently, an increasing amount of wasted lubricant has been due to the unnecessary changing of lubricant even though the lubricant still remains its lubrication behavior. Therefore, a condition based technique is introduced to monitor the degradation parameters in lubricating oil by using optical approach. This work focuses on Total Acid Number (TAN) that has been identified as the main parameter in determining the lifetime of lubricant and it occurred at band location from 1,050-1,250cm-1 and 1,700-1,730cm-1. The best input parameter has been identified for sensor development and signal processing. Then, the prediction model is used to validate the measured and the predicted value of degradation. The high correlation between the predicted and measured data shows the prediction algorithm can be used for prediction purposes efficiently.
AB - This paper presents the development of the prediction algorithm of lubricant degradation based on Beer Lambert's transmittance theory by using Multiple Linear Regressions (MLR). Recently, an increasing amount of wasted lubricant has been due to the unnecessary changing of lubricant even though the lubricant still remains its lubrication behavior. Therefore, a condition based technique is introduced to monitor the degradation parameters in lubricating oil by using optical approach. This work focuses on Total Acid Number (TAN) that has been identified as the main parameter in determining the lifetime of lubricant and it occurred at band location from 1,050-1,250cm-1 and 1,700-1,730cm-1. The best input parameter has been identified for sensor development and signal processing. Then, the prediction model is used to validate the measured and the predicted value of degradation. The high correlation between the predicted and measured data shows the prediction algorithm can be used for prediction purposes efficiently.
KW - Degradation
KW - Lubricant
KW - Multiple Linear regression (MLR)
UR - http://www.scopus.com/inward/record.url?scp=84877829009&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84877829009&partnerID=8YFLogxK
U2 - 10.1109/ICEDSA.2012.6507795
DO - 10.1109/ICEDSA.2012.6507795
M3 - Conference contribution
AN - SCOPUS:84877829009
SN - 9781467321631
SP - 194
EP - 197
BT - International Conference on Electronic Devices, Systems, and Applications
ER -