High performance of a gas identification system using sensor array and temperature modulation

A sensor array based on metal oxide gas sensors has been used to identify three single industrial and environmental gases (carbon monoxide, acetylene and hydrogen sulphide). A temperature modulation method is applied to a sensor array and compared to experiments with different constant temperatures. The former shows more powerful ability in gas discrimination than the latter. Both principal component analysis (PCA) and artificial neural networks (ANN) are performed to classify and identify the target gases. The classification results show the potentiality of the system. Moreover, unknown gases are tested and they are well identified.