Open Access
Cai, Yuzhe
Graduate Program:
Energy and Mineral Engineering
Master of Science
Document Type:
Master Thesis
Date of Defense:
April 17, 2017
Committee Members:
  • Luis F. Ayala H., Thesis Advisor
  • Shimin Liu, Thesis Advisor
  • unconventional gas
  • shale gas
  • tight gas
  • linear flow;
  • well performance
Unconventional natural gas resources have become an important energy supply in North America. Shale gas and tight gas provides over half of the natural gas production in the United States. Linear flow is the most common flow type in tight and shale gas reservoirs. A density-based analytical approach was proposed by Vardcharragosad and Ayala (2014) that could predict well production performance in the linear flow regime. However, this approach requires reservoir and fluid properties for the prediction. As matter of fact, most of these reservoir properties are unknown and undetermined. The thrust of this study is to predict the long-term (Boundary-dominated period) well production behavior with the reservoir properties as inputs through an updated density-based approach. The BDF model proposed in this study is based on Vardcharragosad and Ayala’s density-based approach (2015). A novel and key task for the BDF production prediction is the utilization of historical production data, which is often termed as Production Data Analysis (PDA). A PDA technique is used to estimate the characterization ratio, which is applied to replace reservoir properties in the prediction model. In addition, transition time needs to be determined to estimate reservoir size because it determines the starting point of the boundary-dominated flow. By using the estimated characterization ratio and the transition time, the novel density-based approach could be re-constructed without inputting specific reservoir properties. The proposed model was initially validated with numerical simulation results. This thesis presents a method to predict boundary-dominated flow behavior in tight and shale gas reservoirs without the knowledge of reservoir properties. The proposed transition time determination approach shows advantages over traditional end of half slope method. Thus, original gas in place can be directly calculated from results of transition time determination and early data analysis more accurately. Also, cost of techniques on determining reservoir properties such as well logging could be reduced.