Prediction Model for Concentration Measurement Error Based on SVR in the NDIR Sensor

DOI

10.2991/icca-16.2016.98How to use a DOI?

Keywords

NDIR sensor, The differential absorption detection technology, SVR, Concentration measurement error mode

Abstract

Affected by internal and external mechanical system, environment and other factors, which most likely cause lower measuring precision that can not reach the requirements of measurement and control system in present NDIR sensor. To deal with this problem, this paper puts forward the method of building model for concentration measurement error based on SVR: based on using the differential absorption detection technology to eliminate mechanical factors inside sensor, it adopt SVR learning algorithm, which can build concentration measurement error prediction model under dynamic environment in the given training set. Then we can make corresponding error compensation for gas concentration measurement by using the model. So the gas measurement concentration is largely close to the true value. The experimental result show that regression effect of the SVR learning algorithm is good for model of concentration measurement error, meet the demand of precision of measurement and control system.

Copyright

© 2016, 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/).

Download article (PDF)

TY  - CONF
AU  - Wenjuan Cheng
AU  - Shixiang Tang
AU  - Yongzhong Wu
PY  - 2016/01
DA  - 2016/01
TI  - Prediction Model for Concentration Measurement Error Based on SVR in the NDIR Sensor
BT  - Proceedings of the 2016 International Conference on Intelligent Control and Computer Application
PB  - Atlantis Press
SP  - 410
EP  - 413
SN  - 2352-538X
UR  - https://doi.org/10.2991/icca-16.2016.98
DO  - 10.2991/icca-16.2016.98
ID  - Cheng2016/01
ER  -