A multipoint optical evanescent wave U-bend sensor system based on Artificial Neural Network pattern recognition

An optical fibre (3 sensor) multipoint U-Bend evanescent wave absorption sensor system is reported which is capable of detecting contaminants in water and depositions by coating on its surface. The sensor is based on a continuous 1Km 62.5μm core diameter Polymer Clad Silicon (P.C.S.) fibre which has had its cladding removed in the sensing areas. The sensing fibre is addressed using an Optical Time Domain Reflectometer (OTDR), and is thus capable of resolving distance along its length allowing measurement at multiple points on a single fibre loop. Signals arising from optical fibre sensors can often be complex in nature and this is particularly so in the case of multipoint sensors. Due to cross-coupling effects of interfering parameters, it is difficult to interpret data from such systems using conventional detection techniques. Artificial Neural Network pattern recognition techniques are used for the signal analysis of the sensor, which allow classification of the samples under test, thus allowing the tree measurand to be recognised and separated from any cross-coupling effects that may be present. The system described is capable of recognising cross-sensitivity from interfering parameters such as time scale coating in hard water and the presence of other species e.g. alcohol in the water. Results are included that have been obtained from the sensors OTDR data. Also presented, are the resulting test outputs that have been obtained from a trained feed-forward neural network designed to interpret the sensor data. The system was 100% successful in classification of all test samples analysed.