Detection of volatile organic compounds by using a single temperature-modulated SnO2gas sensor and artificial neural network

A dynamic measurement method is described for the rapid identification and determination of volatile organic compounds (VOCs) in ambient air. For the qualitative recognition of VOCs, only a single SnO2-based gas sensor operating in a rectangular temperature-modulation mode is required. The working temperature of the sensor was modulated between 250 and 300 °C and its dynamic responses to different concentrations of propane-2-ol, acetyl acetone and ethanol vapor were measured. The discrete wavelet transform (DWT) was used to extract important features from the sensor response. These features were then input to a (neural) pattern recognition algorithm. The species considered can be discriminated with a 100% success rate using a back propagation network and the concentrations of the organic vapor can also be accurately predicted.