[CCoE Notice] Thesis Defense Announcement: Application of Data Analytics to Prediction of Initial Production in Tight Oil Reservoir

[Grayson, Audrey A]

MS Thesis Defense

Petroleum Engineering Department



Application of Data Analytics to Prediction of Initial Production in Tight Oil Reservoir

Nathan Tran



Date: Tuesday, December 4, 2018



Location: ConocoPhillips Petroleum Engineering Building 9, Rm 104

Energy Research Park



Time: 11:00 am



Committee Chair:

Dr. Ganesh C. Thakur



Committee Members:

Dr. Guan Qin

Dr. Adwait Chawathe





Fields J, K and N have been discovered and mostly produced in primary recovery stage. Infill drilling has been a strategy to provide incremental oil production for the fields. Predicting Initial production (IP) rate for new infill wells is important because it is used for decline curve analysis to forecast future production hence helpful for economic analysis. The main objective of this study is to develop a quick predictive tool to forecast IP for wells in LT reservoir in the three fields J, K and N.

In conclusion, PCA and linear regression can be used to develop a prediction model for IP. In this case, components with the most correlation to IP contribute to a better prediction model than do the first components with greatest variance. Thorough analysis and quality check on data are important before incorporating them into the prediction model. The model can be used to predict future wells’ initial production which is a very important parameter for economic analysis. Missing data should be handled with understanding the behavior of the parameters (e.g, handling missing pressure data and skin data). As expected, the more data samples in the training set, the more robust the prediction model becomes.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://Bug.EGR.UH.EDU/pipermail/engi-dist/attachments/20181129/641959d6/attachment-0001.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Abstract - Nathan Tran.docx
Type: application/vnd.openxmlformats-officedocument.wordprocessingml.document
Size: 14691 bytes
Desc: Abstract - Nathan Tran.docx
Url : http://Bug.EGR.UH.EDU/pipermail/engi-dist/attachments/20181129/641959d6/attachment-0001.bin