Qualitative Identification of Single/Mixture Gases Based on Fe-ZnO Sensor Array and PSO-BP Neural Network

Fe-zinc oxide (ZnO) materials with self-assembled rod-flower structure were synthesized. X-ray diffraction (XRD), energy-dispersive spectroscopy (EDS), scanning electron microscope (SEM), and X-ray photoelectron spectroscopy (XPS) were used to characterize the morphology, elemental composition, and valence analysis of Fe-ZnO. It was verified that Fe-ZnO sensors have good performances for single/mixed test gases. Combining the sensor array with a back propagation neural network algorithm optimized by particle swarm (PSO-BPNN), qualitative identification of ten different gas concentration levels under three categories was achieved with a detection accuracy of 95%. High classification detection was achieved using the PSO-BPNN model even under the influence of different humidity levels (RH = 35%, 50%, and 80%). So, the combined Fe-ZnO sensor array with PSO-BPNN model can effectively detect toxic gases at different concentration levels and therefore has some potential practical values.