DOI: 10.5593/sgem2017H/15/S06.049


D.R. Jacota, C. Marinoiu, C. Popa
Thursday 23 November 2017 by Libadmin2017

References: 17th International Multidisciplinary Scientific GeoConference SGEM 2017, www.sgemviennagreen.org, SGEM2017 Vienna GREEN Conference Proceedings, ISBN 978-619-7408-26-3 / ISSN 1314-2704, 27 - 29 November, 2017, Vol. 17, Issue 15, 389-396 pp; DOI: 10.5593/sgem2017H/15/S06.049


Reservoir heterogeneity creates difficulties for every stage of production, the effects being more pronounced when little data is available. This is the case for many old and abandoned Romanian oil reservoirs that have very poor reservoir characterization. This paper discusses the efficiency of two prediction methods, ordinary kriging and artificial neural networks, for porosity as an routine core analysis parameter, with the hope of improving reservoir characterization by creating more accurate distribution maps. The present study is conducted in an unfortunate but very common situation regarding old Romanian oil reservoirs, meaning the presence of heterogeneities that have not been considered when the reservoir model was created. Among its many further applications, one of them relates to a new theory, the tertiary migration of hydrocarbons which is the basis of production resumption on abandoned oil reservoirs, in the end pointing out which of the candidate reservoirs is more suitable for production restarting. This paper covers only a small aspect of the whole study which is strongly sustained by the accidental production restarting of two abandoned oil reservoirs in Romania through a trivial and routine reentry in one of the closed wells.

Keywords: ordinary kriging, distribution maps, reservoir image enhancing, efficient tertiary migration, artificial neuronal network