Gas Composition Recognition Based on Analyzing Acoustic Relaxation Absorption Spectra: Wavelet Decomposition and Support Vector Machine Classifier
- DOI
- 10.2991/iceea-18.2018.28How to use a DOI?
- Keywords
- gas compositions recognition; gas acoustic relaxation absorption spectrum; wavelet multi-resolution analysis; multi-class support vector machine
- Abstract
Gas acoustic spectrum represents properties of acoustic propagation, which can distinguish gas compositions. However in few existing methods of gas-composition-unknown recognition, the approaches of programmatically processing the acoustic spectrum curves have not yet been presented. We propose a method for gas-composition-unknown recognition by analyzing gas acoustic relaxation absorption spectrum (GARAS) based on wavelet multi-resolution analysis (MRA) and multi-class support vector machine (SVM). Features of GARAS are extracted by wavelet MRA, and then selected to obtain a few feature coefficients which are utilized to train and test multi-class SVM. Simulation results show that the proposed method completely classifies four examples of gas mixtures, and that it recognizes the mixtures with the same and similar concentration or temperature. This method realizes the numerically extracting and programmatically processing the information of GARAS, and implements gas-composition-unknown sensing based on acoustic spectrums.
- Copyright
- © 2018, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Yaqiong Jia AU - Bin Yu AU - Mingdi Du AU - Xiaoli Wang PY - 2018/03 DA - 2018/03 TI - Gas Composition Recognition Based on Analyzing Acoustic Relaxation Absorption Spectra: Wavelet Decomposition and Support Vector Machine Classifier BT - Proceedings of the 2018 2nd International Conference on Electrical Engineering and Automation (ICEEA 2018) PB - Atlantis Press SP - 126 EP - 130 SN - 2352-5401 UR - https://doi.org/10.2991/iceea-18.2018.28 DO - 10.2991/iceea-18.2018.28 ID - Jia2018/03 ER -