Principal component analysis and artificial neural network based approach to analysing optical fibre sensors signals

• Published in: Sensors and actuators. A. Physical. Vol. 136; no. 1; pp. 28 - 38
• Main Authors: Lewis, E., Sheridan, C., O’Farrell, M., King, D., Flanagan, C., Lyons, W.B., Fitzpatrick, C.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01-05-2007
• Subjects: Artificial neural networks processing for sensor data; Fluorescent coated optical fibres; Food quality sensor; Optical fibre colour sensor; Optical fibre temperature sensor; PCA; Water quality measurement; Water quality measurement; Fluorescent coated optical fibres; Optical fibre temperature sensor; Artificial neural networks processing for sensor data; Food quality sensor; Optical fibre colour sensor; PCA
• ISSN: 0924-4247; 1873-3069
• DOI: 10.1016/j.sna.2007.02.012

This paper investigates the use of artificial neural networks (ANNs) coupled with principal components analysis (PCA) to interpret complex optical spectrum and time resolved signals from optical fibre sensors. Specific reference is made to two application areas addressed by optical fibre sensors which are examples of systems deployed to measure food colour (reflection spectra) as it cooks in a full scale industrial oven and time resolved and Fourier transformed signals received from an optical time domain reflectometer (OTDR) for water monitoring. The method of analysis is different in each case but the same principles apply to each measurement.