V. Sommer et al., NEURAL NETWORKS AND ABDUCTIVE NETWORKS FOR CHEMICAL SENSOR SIGNALS - A CASE COMPARISON, Sensors and actuators. B, Chemical, 28(3), 1995, pp. 217-222
Both artificial neural networks (ANNs) and the abductory induction mec
hanism (AIM) have been proven to be suitable for the evaluation of sig
nals from chemical sensors with strong interactions of gas components
on the catalytically active surface. The algorithms allow the calculat
ion of calibration curves for a multisensor. AIM yields a good mean ap
proximation within a very short time; ANN covers a broader concentrati
on range with an adequate approximation. The signal evaluation of a se
t of two pellistors and another set of six MOSFETs is used for illustr
ation.