Classification of waxy crude oil odor-profile using gas sensor array

• Main Authors: Mawardzi, M.F.R.M., Japper-Jaafar, A., Najib, M.S., Daud, S.M., Ya, T.M.Y.S.T.
• Format: Conference or Workshop Item
• Published: Institute of Physics Publishing 2019
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85060969105&doi=10.1088%2f1757-899X%2f469%2f1%2f012073&partnerID=40&md5=d8b698ec6dc7ee616fe6847bc413f4ff; http://eprints.utp.edu.my/23570/

Nowadays, waxy crude oil becomes one of the major concerns in the oil and gas industry. The waxy crude oil affects the production and transportation of the crude oil from offshore to onshore. Differentiation by image visualization in determining the type of waxy crude oil, with or without wax sometimes appear rather similar between each other. Hence, a new method needs to be used to differentiate and classify the waxy crude oil type. An electronic nose (E-Nose) is one of the devices that could detect and measure the odor data of the waxy crude oil type using gas sensor array. This paper aims to classify four types of Malaysian waxy crude oil from different fields at room temperature. There are 16,000 odor data that has been collected by using the E-Nose. Then, the measured data were normalized and analyzed using boxplot analysis. The unique odor-profile for each type of waxy crude oil sample has been extracted and classified using intelligent classification technique. The four types of waxy crude oil have been classified 100 using k-NN intelligent classification technique with zero percentage of error in this paper. © Published under licence by IOP Publishing Ltd.