A New Method of Mixed Gas Identification Based on a Convolutional Neural Network for Time Series Classification.

Lu Han,Kaitai Xiao,Chongchong Yu,Xia Zhao

doi:10.3390/s19091960

Abstract

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.

Highlights

The electronic nose is an electronic system that uses the response pattern of a gas sensor array to identify gases
Mixtures gas is stopped for 240 s to 300 s, which is the "recovery time" of the sensor, and the sensor returns to Convolutional neural networks are a special type of artificial neural network, which are the baseline value
This section briefly introduces the structure of the convolutional neural networks (CNN) model, including the input layer, output layer and hidden layer, as well as the typical architecture VGG, Resnet of the CNN

Summary

Introduction

The electronic nose is an electronic system that uses the response pattern of a gas sensor array to identify gases. The most widely used are metal oxide gas sensors called MOX gas sensors [1]. MOX gas sensors have the advantages of small volume, fast response speed, low cost and long service life. They are widely used in the field of detection of gases such as industrial exhaust gases, flammable and explosive gases, and the analysis of smells in terms of the intensity of the smell or hedonic quality, etc. The sensor array can obtain multi-path response signals, which provides a feasible means for detecting and analyzing the composition of mixed gases. The types and number of sensors used is determined by the nature of the gas being measured [8,9]

Methods

Results

Conclusion