Dynamic data driven investigation of petrophysical and geomechanical properties for reservoir formation evaluation

Onalo, David Ojonugwa (2019) Dynamic data driven investigation of petrophysical and geomechanical properties for reservoir formation evaluation. Doctoral (PhD) thesis, Memorial University of Newfoundland.

[img] [English] PDF - Accepted Version
Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission.

Download (8MB)


Petrophysical and geomechanical properties of the formation such as Young’s modulus, bulk modulus, shear modulus, Poisson’s ratio, and porosity provide characteristic description of the hydrocarbon reservoir. It is well-established that static geomechanical properties are good representatives of reservoir formations; however, they are non-continuous along the wellbore, expensive and determining these properties may lead to formation damage. Dynamic geomechanical formation properties from acoustic measurements offer a continuous and non-destructive means to provide a characteristic description of the reservoir formation. In the absence of reliable acoustic measurements of the formation, such as sonic logs, the estimation of the dynamic geomechanical properties becomes challenging. Several techniques like empirical, analytical and intelligent systems have been used to approximate the property estimates. These techniques can also be used to approximate acoustic measurements thus enable dynamic estimation of geomechanical properties. This study intends to explore methodologies and models to dynamically estimate geomechanical properties in the absence of some or all acoustic measurements of the formation. The present work focused on developing empirical and intelligent systems like artificial neural networks (ANN), Gaussian processes (GP), and recurrent neural networks (RNN) to determine the dynamic geomechanical properties. The developed models serve as a cost-effective, reliable, efficient, and robust methods, offering dyanmic geomechanical analysis of the formation. This thesis has five main contributions: (a) a new data-driven empirical model of estimating static Young’s modulus from dynamic Young’s modulus, (b) a new data-driven ANN model for sonic well log prediction, (c) a new data-driven GP model for shear wave transit time prediction, (d) a new dynamic data-driven RNN model for sonic well log reproduction, and (e) an assessment on the ANN as a reliable sonic logging tool.

Item Type: Thesis (Doctoral (PhD))
URI: http://research.library.mun.ca/id/eprint/13839
Item ID: 13839
Additional Information: Includes bibliographical references.
Keywords: Sonic transit time, Well logs, Artificial neural networks, Rock formation properties, Recurrent neural networks and Gaussian processes
Department(s): Engineering and Applied Science, Faculty of
Date: January 2019
Date Type: Submission
Library of Congress Subject Heading: Oil well logging; Oil reservoir engineering

Actions (login required)

View Item View Item


Downloads per month over the past year

View more statistics