Abstract
In this paper, we report on a portable, compact and inexpensive fluorescence sensor for monitoring the ageing of retail lubricant oils ax a result of the operation In an internal combustion engine. The proposed sensor consists of a UV laser diode, an optical fibre probe for simultaneous excitation and collection of the fluorescence and a USB-powered palm-size spectrometer connected to a laptop. The sensor wax applied on a set of oil samples, taken from production engines by a leading car manufacturing company, at various mileages. The subtle variations in the spectra, caused by the ageing process of the oil, are resolved using two methods: Principle Component Analysis (PCA) and Artificial Neural Networks (ANN). Their performance is compared with regard to utilization in a portable dedicated system for continuous in-situ monitoring. Because of the "fingerprinting" approach which avoids detailed interpretation, the sensor has a wide range of application areas. © 2004 IEEE.
Original language | English |
---|---|
Title of host publication | Proceedings of IEEE Sensors|Proc. IEEE Sens. |
Editors | D. Rocha, P.M. Sarro, M.J. Vellekoop |
Publisher | IEEE |
Pages | 8-11 |
Number of pages | 3 |
Volume | 1 |
DOIs | |
Publication status | Published - 2004 |
Event | IEEE Sensors 2004 - Vienna Duration: 1 Jul 2004 → … |
Conference
Conference | IEEE Sensors 2004 |
---|---|
City | Vienna |
Period | 1/07/04 → … |
Keywords
- Fluorescence
- Neural Networks
- Oil
- Optical fibre
- PCA