Noninvasive Alcohol Testing Using Near-Infrared Spectroscopy and Partial Least Square Method

Author

Chen, Guoqiang ; Luo, Site ; Li, Zengyong

Author_Institution

Sch. of Mech. Eng., Shandong Univ., Jinan, China

fYear

2012

fDate

21-23 May 2012

Firstpage

1

Lastpage

3

Abstract

Near-infrared spectroscopy (NIRS) and Partial Least Square (PLS) are used for rapid and accurate testing of alcohol in this study. PLS calibration method is applied to establish a model to study the relationship between NIRS and alcohol. In this model, the related coefficient is 0.9828 which proves that predicting model has high linearity and can be used for further analysis. The value of RMSEC is 2.54(V/V, %) and the value of MREP is 4.83%. By comparing the predicted value in model with the measured value, the results show that the predicted model has a high precision. Our results suggest that NIRS can be applied to the alcohol testing as a quick, accurate and real-time method.

Keywords

calibration; chemical variables measurement; infrared spectroscopy; least squares approximations; mean square error methods; spectrochemical analysis; PLS calibration method; mean relative error of prediction; near infrared spectroscopy; noninvasive alcohol testing; partial least square method; root mean square error of calibration; Absorption; Analytical models; Calibration; Mathematical model; Predictive models; Spectroscopy; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Photonics and Optoelectronics (SOPO), 2012 Symposium on

Conference_Location

Shanghai

ISSN

2156-8464

Print_ISBN

978-1-4577-0909-8

Type

conf

DOI

10.1109/SOPO.2012.6270460

Filename

6270460