Mixed vapour identification using partition column-QCMs and Artificial Neural Network

Abstract

This Paper presents the identification of mixed vapour using electronic nose system composed of Quartz Crystal Microbalance (QCM) sensor array and a partition column of gas chromatography. The polymer coated QCMs produced a specific frequency shift. The data set was processed by an Artificial Neural Network using Backpropagation algorithm as a pattern recognition. The result showed that this equipment was able to identify five types of vapours namely benzene, acetone, isopropyl alcohol, non-polar and polar mixture (i.e. benzene and acetone), and also polar and polar mixture (i.e. isopropyl alcohol and acetone) with the identification rate of 96%.