Document details

Stratigraphic interpretation of Well-Log data of the Athabasca Oil Sands of Alberta Canada through Pattern recognition and Artificial Intelligence

Author(s): Igbokwe, Onyedikachi Anthony

Date: 2011

Persistent ID:

Origin: Repositório Institucional da UNL

Subject(s): Automatic Stratigraphic Interpretation; Oil Sand; Eletrofacies; Lithofacies; Artificial Intelligence


Automatic Stratigraphic Interpretation of Oil Sand wells from well logs datasets typically involve recognizing the patterns of the well logs. This is done through classification of the well log response into relatively homogenous subgroups based on eletrofacies and lithofacies. The electrofacies based classification involves identifying clusters in the well log response that reflect ‘similar’ minerals and lithofacies within the logged interval. The identification of lithofacies relies on core data analysis which can be expensive and time consuming as against the electrofacies which are straight forward and inexpensive. To date, challenges of interpreting as well as correlating well log data has been on the increase especially when it involves numerous wellbore that manual analysis is almost impossible. This thesis investigates the possibilities for an automatic stratigraphic interpretation of an Oil Sand through statistical pattern recognition and rule-based (Artificial Intelligence) method. The idea involves seeking high density clusters in the multivariate space log data, in order to define classes of similar log responses. A hierarchical clustering algorithm was implemented in each of the wellbores and these clusters and classifies the wells in four classes that represent the lithologic information of the wells. These classes known as electrofacies are calibrated using a developed decision rules which identify four lithology -Sand, Sand-shale, Shale-sand and Shale in the gamma ray log data. These form the basis of correlation to generate a subsurface model.

Dissertation submitted in partial fulfillment of the requirements for the Degree of Master of Science in Geospatial Technologies.

Document Type Master thesis
Language English
Advisor(s) Pebesma, Edzer; Costa, Ana Cristina; Mahiques, Jorge Mateu
Contributor(s) Igbokwe, Onyedikachi Anthony
facebook logo  linkedin logo  twitter logo 
mendeley logo

Related documents

No related documents