FUZZY NEURAL COMPUTING OF COFFEE AND TAINTED-WATER DATA FROM AN ELECTRONIC NOSE

In this paper we compare the ability of a fuzzy neural network and a c ommon back-propagation network to classify odour samples that were obt ained by an electronic nose employing semiconducting oxide conductomet ric gas sensors. Two different sample sets have been analysed: first, the aroma of three blends of commercial coffee, and secondly, the head space of six different tainted-water samples. The two experimental dat a sets provide an excellent opportunity to test the ability of a fuzzy neural network due to the high level of sensor variability often expe rienced with this type of sensor. Results are presented on the applica tion of three-layer fuzzy neural networks to electronic nose data. The y demonstrate a considerable improvement in performance compared to a common back-propagation network.