Date of Graduation


Document Type


Degree Type



Statler College of Engineering and Mineral Resources


Petroleum and Natural Gas Engineering

Committee Chair

Shahab D. Mohaghegh

Committee Co-Chair

Samuel Ameri

Committee Member

Khashayar Aminian

Committee Member

Peter Dzurman

Committee Member

Razi Gaskari

Committee Member

Thomas H. Wilson


Reservoir simulation and modeling is utilized throughout field development in different capacities. Sensitivity analysis, history matching, operations optimization and uncertainty assessment are the conventional analyses in full field model studies. Realistic modeling of the complexities of a reservoir requires a large number of grid blocks. As the complexity of a reservoir increases and consequently the number of grid blocks, so does the time required to accomplish the abovementioned tasks.;This study aims to examine the application of pattern recognition technologies to improve the time and efforts required for completing successful history matching projects. The pattern recognition capabilities of Artificial Intelligence and Data Mining (AI&DM;) techniques are used to develop a Surrogate Reservoir Model (SRM) and use it as the engine to drive the history matching process. SRM is a prototype of the full field reservoir simulation model that runs in fractions of a second. SRM is built using a small number of geological realizations.;To accomplish the objectives of this work, a three step process was envisioned:;• Part one, a proof of concept study: The goal of first step was to prove that SRM is able to substitute the reservoir simulation model in a history matching project. In this part, the history match was accomplished by tuning only one property (permeability) throughout the reservoir.;• Part two, a feasibility study: This step aimed to study the feasibility of SRM as an effective tool to solve a more complicated history matching problem, particularly when the degrees of uncertainty in the reservoir increase. Therefore, the number of uncertain reservoir properties increased to three properties (permeability, porosity, and thickness). The SRM was trained, calibrated, and validated using a few geological realizations of the base reservoir model. In order to complete an automated history matching workflow, the SRM was coupled with a global optimization algorithm called Differential Evolution (DE). DE optimization method is considered as a novel and robust optimization algorithm from the class of evolutionary algorithm methods.;• Part three, a real-life challenge: The final step was to apply the lessons learned in order to achieve the history match of a real-life problem. The goal of this part was to challenge the strength of SRM in a more complicated case study. Thus, a standard test reservoir model, known as PUNQ-S3 reservoir model in the petroleum engineering literature, was selected. The PUNQ-S3 reservoir model represents a small size industrial reservoir engineering model. This model has been formulated to test the ability of various methods in the history matching and uncertainty quantification. The surrogate reservoir model was developed using ten geological realizations of the model. The uncertain properties in this model are distributions of porosity, horizontal, and vertical permeability. Similar to the second part of this study, the DE optimization method was connected to the SRM to form an automated workflow in order to perform the history matching. This automated workflow is able to produce multiple realizations of the reservoir which match the past performance. The successful matches were utilized to quantify the uncertainty in the prediction of cumulative oil production.;The results of this study prove the ability of the surrogate reservoir models, as a fast and accurate tool, to address the practical issues of reservoir simulation models in the history matching workflow. Nevertheless, the achievements of this dissertation are not only aimed at the history matching procedure, but also benefit the other time-consuming operations in the reservoir management workflow (such as sensitivity analysis, production optimization, and uncertainty assessment).