Chemical sensor arrays and abductive networks

It is shown how a type of adaptive learning network can be used to predict the concentration of individual components in a gas mixture from the signals obtained from a field-effect-transistor sensor array. The network is formed by the use of software based on the so-called abductory induction mechan...

Full description

Saved in:
Bibliographic Details
Published in:Sensors and actuators. B, Chemical Chemical, 1992, Vol.9 (2), p.127-131
Main Authors: Sundgren, H., Lundström, I., Vollmer, H.
Format: Article
Language:English
Subjects:
Analytical chemistry
Chemistry
Exact sciences and technology
General, instrumentation
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:It is shown how a type of adaptive learning network can be used to predict the concentration of individual components in a gas mixture from the signals obtained from a field-effect-transistor sensor array. The network is formed by the use of software based on the so-called abductory induction mechanism (AIM). It is demonstrated how AIM defines very simple networks which can predict the hydrogen and ammonia concentrations in mixtures of hydrogen, ammonia, ethylene and ethanol in air quite well. Furthermore, the learning time is found to be very short. The use of self-organizing methods for the evaluation of data from chemical sensor arrays is discussed, with special emphasis on abductory induction.
ISSN:0925-4005
1873-3077
DOI:10.1016/0925-4005(92)80205-C