Abstract:Pattern recognition is the core component of the electronic nose (E-nose). Traditional machine learning algorithms highly rely on the feature data selected manually for model training and testing. A complete experiment must be performed before the data can be further processed. To realize the automatic extraction of response features and simplify the model's training and application process, a multi-task convolutional neural network (MTL-CNN) with a dual-block knowledge-sharing structure is designed to train a model for the E-nose system. This model can simultaneously perform three different classification tasks, for the purposes of target discrimination, concentration prediction, and state judgment. Only a few consecutive seconds of response data are needed to be input into the trained model to obtain various information about the E-nose. With the utilization of an unmanned gas-sensing test system, large-scale measurements of the E-nose can be carried out automatically. A baseline tracking algorithm (BTA) is designed based on the relative changes of short-term data, reducing the impact of long-term shifts. Over thousands of gas response processes and more than 10 million sensing data have participated in the training of the deep learning model. The 5-fold cross-validation method shows that the fully trained model has an outstanding generalization performance. After the baseline is tracked automatically, the accuracy of three tasks towards 12 kinds of volatile organic compounds (VOCs) is about 95% (type recognition: 95.2%, concentration prediction: 92.1%, status judgment: 97.3%) using only 4 seconds of sensing data during the response status of the E-nose. Our work shows the distinct advantages of combining "big data" and "deep learning" in the gas-sensing field and further proves that the application of MTL-CNN can significantly improve the training and application efficiency of the E-nose.

Type discrimination and concentration prediction towards ethanol using a machine learning–enhanced gas sensor array with different morphology-tuning characteristics

Classification and Concentration Prediction of VOCs With High Accuracy Based on an Electronic Nose Using an ELM-ELM Integrated Algorithm

Enhanced performance of the tangerines-like CuO-based gas sensor using ZnO nanowire arrays

Pattern Recognition with Temperature Regulation: A Single YSZ-Based Mixed Potential Sensor Classifies Multiple Mixtures of Isoprene, N-Propanol, and Acetone.

Discriminative analysis of volatile organic compounds using machine-learning assisted Au loaded ZnO and TiO 2 -based thin film sensors

Multivariable Sensor Based On Mxene And Machine Learning For Selective Detections Of Vocs

Machine-learning-aided Identification of Ethanol in Humid Air Using Zinc Complex Capped CsPbBr3 Resistive Sensors

Several ML Algorithms and Their Feature Vector Design for Gas Discrimination and Concentration Measurement with an Ultrasonically Catalyzed MOX Sensor

Discrimination of VOCs with e-nose consisting of a single MEMS sensor using lightweight deep learning model based on SqueezeNet

Machine learning-based discrimination of indoor pollutants using an oxide gas sensor array: High endurance against ambient humidity and temperature

Target discrimination, concentration prediction, and status judgment of electronic nose system based on large-scale measurement and multi-task deep learning

High energy facets and oxygen vacancies guided hierarchical tin dioxide microcubes assembled by cross-stacked nanoslices for ethanol gas-sensing

Selective Sensing of Volatile Organic Compounds Using an Electrostatically Formed Nanowire Sensor Based on Automatic Machine Learning

Integrated Mixed Potential Gas Sensor with Efficient Structure for Discriminative Volatile Organic Compounds Detection.

A New Device Structure Yttria-Stabilized Zirconia-Based Mixed Potential Gas Sensor for Volatile Organic Compounds Gas Classification

Microwave Preparation and Remarkable Ethanol Sensing Properties of ZnO Particles with Controlled Morphologies in Water-Ethylene Glycol Binary Solvent System

One-step Electrospun SnO2/MOx Heterostructured Nanomaterials for Highly Selective Gas Sensor Array Integration

An Enhanced Gas Sensor Data Classification Method Using Principal Component Analysis and Synthetic Minority Over-Sampling Technique Algorithms

Machine Learning-Driven Gas Concentration Forecasting: A Case Study with Ws2 Nanoflower Gas Sensor

Utilizing Machine Learning for Rapid Discrimination and Quantification of Volatile Organic Compounds in an Electronic Nose Sensor Array

Identification and Pattern Recognition Analysis of Chinese Liquors by Doped Nano Zno Gas Sensor Array