Determination of the composition of NO2 and NO mixture by thin film sensor and back-propagation network

In this work, vacuum deposited thin films of lead phthalocyanine (PbPc) were employed as gas sensor to detect mixtures of NO2 and NO. Data collected from sensor responses were used to train a two-stage back-propagation network (BPN) for the determination of the gas mixture composition. The success rate for the two-stage BPN in identifying categories of gas mixture approaches 100%. The maximum error for the two-stage BPN in predicting NO concentration in a gas mixture with fixed NO2 concentration is 14.7%.