Automatic detection of outlier data received in multi-parametric capillary sensors of diesel fuels

Multi-parametric capillary sensor with local sample heating has been shown as effective tool for diesel fuel classification at a laboratory level of technology, where trained operator performs experiments. The sensor consists of disposable capillary optrode, head and measurement control unit. Technology level increasing of the sensor requires automatization of samples handling or automatic rejection of uncertain data base of outlier data analysis. Such data collection may come from imprecision of capillary optrode diameters, inaccuracy of optrode filling with sample, inaccuracy of corking the sample as well as inaccuracy of optrode positioning in head. Mentioned inaccuracies of preparation of the measurement may lead to outlier data, which disables proper sample classification. Therefore, in this paper automatic detection of outlier data received in multi-parametric capillary sensors of diesel fuels is proposed and examined with data collected by untrained and trained operator. Performed experiments show that direct statistical tools using at raw data leads to improper results of outlier data pointing. The proper outlier data pointing taking place for raw data converted to array pattern data on the base of physical phenomena described by experimental data.

Keywords: capillary sensor, multi-parametric sensor, automatic data processing, diesel fuel, outlier data, uncertain data, fuel quality sensor, diesel fuel quality

Authors: M. Borecki(a), P. Prus(b), M.L, Korwin-Pawlowski(c), P. Doroz(a), J. Szmidt(a)

(a) Warsaw University of Technology, 75 Koszykowa Str., 00-662 Warsaw, Poland
(b) BOI, sp. z o. o., 31/101 Lwowska Str., Oleśnica, Poland
(c) Université du Québec en Outaouais, 101 rue Saint-Jean-Bosco, Gatineau, J8X 3X7, Canada