Estimation of dielectric properties of crude oils based on IR spectroscopy

Validerad; 2014; 20140917 (andbra)
Dielectric properties of crude oils play an important role in characterization and quality control. Measuring permittivity accurately over a wide range of frequencies is, however, a time-consuming task and existing measurement methods are not easily adapted for real-time diagnostics. IR spectroscopy, on the other hand, provides rapid measurements of fundamental molecular properties.In this paper we show that by using multivariate calibration tools such as PLS regression, it is possible to extract dielectric properties of crude oils directly from IR spectra, in addition to conventional interpretation of the spectra, hence reducing the need for direct electrical measurements. Results on 16 different oil samples show that the dielectric parameters obtained with the proposed method agree well with those obtained using direct permittivity measurements. The PLS regression method has also been extended with Monte-Carlo simulation capabilities to account for uncertainties in the data

Tomren, Andreas LingeDepartment of Chemistry, University of Bergen (författare)
Folgerø, KjetilMichelsen Centre for Industrial Measurement Science & Technology, P.O.Box 6031, NO- 5892 Bergen, NORWAY, Christian Michelsen Research AS, Bergen (författare)
Barth, TanjaDepartment of Chemistry, University of Bergen (författare)
Luleå tekniska universitetSignaler och system (creator_code:org_t)

Ingår i:Chemometrics and Intelligent Laboratory Systems: Elsevier BV139, s. 1-50169-74391873-3239

Av författaren/redakt...: Carlson, Johan E ...; Tomren, Andreas ...; Folgerø, Kjetil; Barth, Tanja

Om ämnet

TEKNIK OCH TEKNOLOGIER: TEKNIK OCH TEKNO ...; och Elektroteknik oc ...; och Signalbehandling

Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.