relation: https://khub.utp.edu.my/scholars/6611/
title: Squared exponential covariance function for prediction of hydrocarbon in seabed logging application
creator: Mukhtar, S.M.
creator: Daud, H.
creator: Dass, S.C.
description: Seabed Logging technology (SBL) has progressively emerged as one of the demanding technologies in Exploration and Production (E&P) industry. Hydrocarbon prediction in deep water areas is crucial task for a driller in any oil and gas company as drilling cost is very expensive. Simulation data generated by Computer Software Technology (CST) is used to predict the presence of hydrocarbon where the models replicate real SBL environment. These models indicate that the hydrocarbon filled reservoirs are more resistive than surrounding water filled sediments. Then, as hydrocarbon depth is increased, it is more challenging to differentiate data with and without hydrocarbon. MATLAB is used for data extractions for curve fitting process using Gaussian process (GP). GP can be classified into regression and classification problems, where this work only focuses on Gaussian process regression (GPR) problem. Most popular choice to supervise GPR is squared exponential (SE), as it provides stability and probabilistic prediction in huge amounts of data. Hence, SE is used to predict the presence or absence of hydrocarbon in the reservoir from the data generated. © 2016 Author(s).
publisher: American Institute of Physics Inc.
date: 2016
type: Conference or Workshop Item
type: PeerReviewed
identifier:   Mukhtar, S.M. and Daud, H. and Dass, S.C.  (2016) Squared exponential covariance function for prediction of hydrocarbon in seabed logging application.  In: UNSPECIFIED.     
relation: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85006013732&doi=10.1063%2f1.4968061&partnerID=40&md5=2a3efda7e71c464c053a27aebc0eebf9
relation: 10.1063/1.4968061
identifier: 10.1063/1.4968061