Date of Graduation


Document Type


Degree Type



Statler College of Engineering and Mineral Resources


Petroleum and Natural Gas Engineering

Committee Chair

Shahab Mohaghegh

Committee Co-Chair

Samuel Ameri

Committee Member

Ebrahim Fathi

Committee Member

Dengliang Gao

Committee Member

Ali T Borujeni


The preferred common tool to estimate the performance of oil and gas fields under different production scenarios is numerical reservoir simulation. A comprehensive numerical reservoir model has tens of millions of grid blocks. The massive potential of existing numerical reservoir simulation models have gone unrealized because they are computationally expensive and time-consuming. Therefore, an effective alternative tool is required for fast and reliable decision making. To reduce the required computational time, proxy models have been developed. Traditional proxy models are either statistical or reduced-order models (ROM). They were developed to substitute complex numerical simulation with producing a representation of the system at a lower computational cost. However, there are shortcomings associated with these approaches when applied to complex systems.;In this study, a novel proxy-model approach is presented in order to overcome the computational size and the traditional proxy-model challenges. The smart proxy model presented is based on artificial intelligence and data-mining techniques. The objective of this study was to develop two types of smart proxy models at each grid block. The first smart proxy model was generated to identify dynamic reservoir properties (pressure and saturation). The other proxy model was created to determine the production profile of a well. The two smart proxy models can be coupled in order to examine field production performance under different operational and geological realization.;The field of study in this work is the SACROC unit. It is a depleted oil field located in Scurry County, Texas. The production history of this field began back in the late 1940s. Based on the long period of production and the different drive mechanisms employed throughout the fields exploitation, its performance history was divided into three phases in this study. Each phase was investigated and smart proxy models were applied to each.;To develop a smart proxy model, multiple reservoir simulation scenarios are designed for different operational constraints and geological realizations. The geological parameters along with the results from the designed simulation runs are collected to build the spatial-temporal database. The parameters in the database are studied and key performance indicators are measured to select the required data to build the smart proxy model. The smart proxy is trained, calibrated, and validated using a series of neural networks. To validate a smart proxy model, it is deployed to replicate a blind numerical simulation run.;The developed smart proxy models are capable of supplying reservoir properties along with production profiles very quickly (seconds) and with an acceptable range of error compared to numerical reservoir models.