Quantitative discrimination of the binary gas mixtures using a combinational structure of the probabilistic and multilayer neural networks

In this study, the quantitative discrimination of seven different types of binary volatile organic gas mixtures were realized by using a proposed structure which was combination of probabilistic neural networks (PNNs) and multilayer neural networks (MLNNs). At the first phase of the discrimination, the binary gas mixtures were classified using PNNs. For comparison, the MLNN structures were also used at this phase. And at the second phase, the MLNNs were processed for the quantitative identification of individual gas concentrations in their gas mixtures. A data set consisted of the steady state sensor responses from quartz crystal microbalance (QCM) type sensors was used for the training of the PNNs and MLNNs. The components in the binary mixture were quantified applying the sensor responses from the QCM sensor array as inputs to the combined neural network structures. The performance of the combined network structure was discussed based on the experimental results.