Incorporation of Engineering Knowledge in History Matching, Optimization, and Uncertainty Assessment Frameworks with Application to the SAGD Process

Date
2015-04-30
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Sensitivity Analysis (SA), History Matching (HM), Optimization (OP), and Uncertainty Assessment (UA) are the four main steps during a field development plan. The main stimulus in this thesis is to provide promising frameworks within HM, OP, and UA contexts. These workflows are assessed to show their merits for tackling complex problems, mainly the Steam Assisted Gravity Drainage (SAGD) process. The first phase of this research study is devoted to the HM problems, where an optimization algorithm called Differential Evolution (DE) is utilized as one of the most efficient global optimization techniques to solve challenging reservoir engineering problems. Comparative studies were conducted to show the merits of the DE method compared with the Particle Swarm Optimization (PSO), which is a common optimizer used in this arena. The results confirmed the superiority of the DE technique over the PSO scheme from different perspectives. Integration of engineering knowledge in the current HM and OP workflows has not been considered for a long time, where reservoir engineers are not able to monitor the optimization methods in order to control their mechanisms during generating candidate solutions. Therefore, in the second phase, a workflow is developed to facilitate the incorporation of engineering knowledge to deal with constraints in the HM and OP contexts. The proposed framework called Fuzzy Inference System - Differential Evolution (FIS + DE), was applied during different HM and multi-objective optimization (MOO) examples, where the results demonstrated its power to solve reservoir engineering problems in a more realistic way than the current approaches. Eventually, correlations that may exist between parameters during an UA problem are addressed in this thesis. In general, they are assumed to be independent for the sake of simplicity. This entails a modified approach to improve the current UA workflows by considering the interrelationships between the uncertain parameters. The technique was applied to a SAGD case study, in which the outcomes indicated that neglecting the relationships between variables could yield a bias during an UA study due to underestimating the overall uncertainty. However, the integration of the parameters correlations provided reliable solutions to facilitate the decision making process.
Description
Keywords
Computer Science, Engineering--Petroleum
Citation
Mirzabozorg, A. (2015). Incorporation of Engineering Knowledge in History Matching, Optimization, and Uncertainty Assessment Frameworks with Application to the SAGD Process (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/26502