Date of Graduation

2018

Document Type

Thesis

Degree Type

MS

College

Statler College of Engineering and Mineral Resources

Department

Petroleum and Natural Gas Engineering

Committee Chair

Ali Takbiri-Borujeni

Committee Co-Chair

Fatemeh Belyadi

Committee Member

Ebrahim Fathi

Committee Member

Ming Gu

Abstract

For the last few decades multiple business sectors have been influenced by the advancement in Artificial Intelligence (AI). Though the oil and gas sector began to utilize the potential of AI comparatively latter than many other sectors, the appreciable amount of work has been done by researchers to equip the industry with AI tools. This work aims to explore various horizons of petroleum engineering by using different AI tools.;For providing better decision making in reservoir fluid characterization problem, fuzzy logic has been applied, which is an AI method to drive decisions when data is incomplete or unreliable. The second part of the work is the combination of supervised and unsupervised machine learning has provided an automated version of well log analysis, where the generated algorithm is able to distinguish between different lithological zones on the basis of well log parameters.;The majority of the problems such as drilling process optimization, production forecasting, comes under the umbrella of statistical regression. The supervised learning regression algorithm was generated to predict the drilling performance in terms of rate of penetration. The similar model was used for producing regression analysis of reservoir that has been treated by steam assisted gas drainage. The accuracy of both cases were investigated by comparing the prediction with available real time data.;The work has been concluded by providing conclusion gathered from comparing different methods and limitations of methodologies derived from Artificial Intelligent (AI) tools.

Download

Share

COinS