An In2O3 Nanotubes based Gas Sensor Array combined with Machine Learning Algorithms for Trimethylamine Detection

Ren, Wenjie; Zhao, Changhui; Liu, Yingming; Wang, Fei

10.1109/NEMS51815.2021.9451424

2021

16th IEEE International Conference on Nano/Micro Engineered and Molecular Systems (IEEE-NEMS)

2474-3747

2474-3755

978-1-6654-3008-1

2021 IEEE 16TH INTERNATIONAL CONFERENCE ON NANO/MICRO ENGINEERED AND MOLECULAR SYSTEMS (NEMS)

1042-1046

APR 25-29, 2021

null,Xiamen,PEOPLES R CHINA

345 E 47TH ST, NEW YORK, NY 10017 USA

IEEE

This work focuses on developing an electronic nose system with machine learning algorithm for detection of trimethylamine (TMA). Pure and Ga-doped In2O3 nanotubes are synthesized by a simple electrospinning method, and four kinds of gas sensors (pristine, 1% Ga, 10% Ga, and 20% Ga-doped In2O3) are fabricated to form a sensor array. Results show that the sensor array can classify TMA effectively from interference gases (xylene, ethanol, hydrogen sulfide) by a support vector machine (SVM) algorithm. Several algorithms, including radial basis function neural network (RBFNN), back propagation neural network (BPNN) and principal component analysis combined with linear regression (PCA-LR), are used to predict the concentration level of each gas. For TMA gas, the trained algorithms can predict its concentration with average relative errors of 1.22% for RBFNN, 2.5% for BPNN and 13.34% for PCA-LR. Furthermore, the binary mixtures of TMA and ethanol are measured and used to train the above algorithms, and the lowest average relative error of 1.74% is achieved in the case of RBFNN algorithm.

Support vector machines Machine learning algorithms Radial basis function networks Nanotubes Interference Prediction algorithms Electronic noses

第一 ; 通讯

英语

CPCI-S ; EI

National Key Research and Development Program of China[2020YFB2008604]

Engineering ; Science & Technology - Other Topics ; Materials Science

Engineering, Electrical & Electronic ; Nanoscience & Nanotechnology ; Materials Science, Multidisciplinary

WOS:000693407600125

20213410818127

Backpropagation ; Binary mixtures ; Electronic nose ; Ethanol ; Gallium compounds ; Gas detectors ; Gases ; Hydrogen sulfide ; Indium compounds ; Nanosensors ; Nanotubes ; Radial basis function networks ; Sensory aids ; Sulfur compounds ; Support vector machines

Rehabilitation Engineering and Assistive Technology:461.5 ; Computer Software, Data Handling and Applications:723 ; Artificial Intelligence:723.4 ; Nanotechnology:761 ; Chemistry:801 ; Chemical Products Generally:804 ; Accidents and Accident Prevention:914.1 ; Solid State Physics:933

Web of Science

Southern Univ Sci & Technol, Sch Microelect, Shenzhen 518055, Peoples R China

Ren, Wenjie,Zhao, Changhui,Liu, Yingming,et al. An In2O3 Nanotubes based Gas Sensor Array combined with Machine Learning Algorithms for Trimethylamine Detection[C]. 345 E 47TH ST, NEW YORK, NY 10017 USA:IEEE,2021:1042-1046.