Discrimination of individual gas/odor using responses of integrated thick film tin oxide sensor array and fuzzy-neuro concept

Abstract

A new neural network classifier for an Intelligent Gas Sensor (IGS) application is presented. The classifier is trained on fuzzyfied training set. Its superior classification and learning performance is demonstrated for discrimination of alcohols and alcoholic beverages using published data of thick film tin oxide sensor array fabricated and characterized at our laboratory. The new model proposed in this article not only gives a steep and monotone learning curve, but also exhibits lower sensitivity to learning parameter choices.