An electronic nose is described, which consists of a gas sensor array combined with a pattern recognition routine. The sensor array used consists of ten metal-oxide-semiconductor field effect transistors with gates of catalytically active metals. It also contains four commercially available chemical sensors based on tin dioxide, so-called Taguchi sensors. In some studies, a carbon dioxide monitor based on infrared absorption is also used. Samples of ground beef and pork, stored in a refrigerator, have been studied. Gas samples from the meat were then led to the sensor array, and t h e resulting patterns of sensor signats were treated with pattem recognition software based on an artificial neural network as well as with an algorithm based on an abductory induction mechanism. When using all sensors for learning, the two nets could predict both type of meat and storage time quite well. Omitting the carbon dioxide monitor, both nets could predict type of meat, but storage time not so well. Finally, it is also shown how a net based on unsupervised training could be used to predict storage time for ground beef.
A very promising way of increasing the selectivity and sensitivity of gas sensors is to treat the signals from a number of different gas sensors with pattern recognition (PARC) methods. A gas sensor array with six metal-oxide-semiconductor field-effect-transistors (MOSFETs) operating at elevated temperatures was exposed to two types of multiple-component gas mixture, one containing 5-65 ppm of hydrogen, ammonia, ethanol and ethylene in air and the other containing hydrogen and acetone in air. The signals from the sensors were analysed with both conventional multivariate analysis, partial least-squares (PLS), and artificial neural network (ANN) models. The results show that both hydrogen and ammonia concentrations can be predicted with PLS models; the predictions were even better with ANN models. The predictions for ethanol and ethylene concentrations were, however, poor for both types of model. Hydrogen and acetone, from the two-component mixture, were best predicted from an ANN model.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.