This paper proposes a new method of mixed gas identification based on a convolutional neural network for time series classification. In view of the superiority of convolutional neural networks in the field of computer vision, we applied the concept to the classification of five mixed gas time series data collected by an array of eight MOX gas sensors. Existing convolutional neural networks are mostly used for processing visual data, and are rarely used in gas data classification and have great limitations. Therefore, the idea of mapping time series data into an analogous-image matrix data is proposed. Then, five kinds of convolutional neural networks—VGG-16, VGG-19, ResNet18, ResNet34 and ResNet50—were used to classify and compare five kinds of mixed gases. By adjusting the parameters of the convolutional neural networks, the final gas recognition rate is 96.67%. The experimental results show that the method can classify the gas data quickly and effectively, and effectively combine the gas time series data with classical convolutional neural networks, which provides a new idea for the identification of mixed gases.
Element doping is an effective method to improve the performance of ZnO varistors. Previous studies mainly focused on the variation of microstructures and Schottky barriers. In this study, the effects of Co dopant on electrical properties are investigated from the aspect of multiscale defect structures, including intrinsic point defects, the heterogeneous interface of depletion/intergranular layers, and interface states at grain boundaries. Combining with analysis of phase composition and energy dispersive spectroscopy, it is found that Co tends to dissolve into ZnO grains when slightly doped. It substitutes Zn 2+ with the same valence and affects little on densities of donors. Segregation of Co at grain boundaries would result in the formation of spinel phase Co(Co 4/3 Sb 2/3)O 4 and transformation of the intergranular phase from α-Bi 2 O 3 to δ-Bi 2 O 3. Meanwhile, densities of point defects are indirectly affected by oxygen ambient during sintering, resulting in abnormal variation of grain resistivity. And interface states are enhanced, leading to improved barriers at grain boundaries. Therefore, reduced leakage current, enhanced grain resistivity, and improved non-linear coefficient in Co-doped ZnO varistor blocks are understood from the underlying multiple defect structures. This presents a potential approach to explore short-term performance and long-term stability of ZnO varistors from the aspect of defect responses.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.