Year-end Sale % OFF 
Abstract
Volume change of oil between reservoir condition and standard surface condition is called oil formation volume factor (FVF), which is very time, cost and labor intensive to determine. This study proposes an accurate, rapid and cost-effective approach for determining FVF from reservoir temperature, dissolved gas oil ratio, and specific gravity of both oil and dissolved gas. Firstly, structural risk minimization (SRM) principle of support vector regression (SVR) was employed to construct a robust model for estimating FVF from the aforementioned inputs. Subsequently, an alternating conditional expectation (ACE) was used for approximating optimal transformations of input/output data to a higher correlated data and consequently developing a sophisticated model between transformed data. Eventually, a committee machine with SVR and ACE was constructed through the use of hybrid genetic algorithm-pattern search (GA-PS). Committee machine integrates ACE and SVR models in an optimal linear combination such that makes benefit of both methods. A group of 342 data points was used for model development and a group of 219 data points was used for blind testing the constructed model. Results indicated that the committee machine performed better than individual models.
Highlights
Oil formation volume factor (FVF) is defined as the ratio of the volume of oil at the prevailing reservoir temperature and pressure to the volume of oil at standard conditions (Ahmed 2000)
A simple curve fitting between output and sum of transformations produces functional form for estimating formation volume factor from PVT data
This study proposed an accurate, cheap and rapid way for estimating oil formation volume factor from available PVT data
Summary
Oil formation volume factor (FVF) is defined as the ratio of the volume of oil (plus the gas in solution) at the prevailing reservoir temperature and pressure to the volume of oil at standard conditions (Ahmed 2000). The quest for higher accuracy forced researchers not to satisfy themselves with individual intelligent systems but to develop integrated models such as committee machines for enhancing precision of final prediction (Asoodeh and Bagheripour 2012b, Asoodeh 2013, Asoodeh et al 2014a, b; Gholami et al 2014c, d; Bagheripour et al 2014). Results of SVR and ACE models were combined by means of genetic algorithm-pattern search technique in an optimal linear combination of committee machine. This strategy was successfully applied to open source oil samples. GHOLAMI the committee machine significantly enhanced accuracy of final prediction compared with individual ACE and SVR models
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
