Skip to Main Content
Since metal oxide sensors grow to be more selective and more sensitive to low concentrations, we propose to study an electronic nose for an environmental air quality control application. In this aim, an array of six commercial gas sensors (TGS, Figaro) is tested to detect and quantify accurately in a reduced time, two pollutants gases H2S and NO2 alone or mixed in industrial areas. Known inconvenient of these resistive chemical sensors have been taken into account in this work, either by the characterisation procedure (humidity dependence, non linearity) or by the pattern recognition method (drift, non selectivity). So, the use of two or three representative parameters extracted from the time-dependent response (for respectively 5 or 15 minutes of gas exposure) and the discriminant factorial analysis algorithm, permit us to obtain a good estimation of the nature and the concentration of the studied gaseous atmospheres.