arrows arrow-right arrow-left menu search rss youtube linkedin twitter facebook instagram arrow-play linkedIn

Unconventional Resources Technology Conference

  • Dates:
    Jul 24
     - 
    Jul 26, 2017
  • Location: Austin, TX, U.S.
  • Address: Austin Convention Center
Battelle researchers will explore data analytics methods for oil production optimization at the upcoming Unconventional Resources Technology Conference

During the conference, Battelle Research Leader Srikanta Mishra will present:  Application of Data Analytics for Production Optimization in Unconventional Reservoirs: A Critical Review.

Big data analytics and data-driven modeling have become quite the buzzwords in recent years—especially in the context of analyzing the production behavior of tight oil and shale gas formations. The ultimate goal is to develop data-driven insights for understanding and optimizing the performance of unconventional reservoirs, where robust mechanistic models of flow (from nano-pores through a network of natural and induced fractures into a multi-stage hydraulically fractured horizontal well) are still under development. However, despite its success in fields ranging from consumer marketing to cybersecurity to health care, the subject of data analytics remains a mystery to most petroleum engineers and geoscientists because of the statistics-heavy jargon and the use of complex "black-box" algorithms.

This presentation will provide a critical review of the state-of-the-art regarding applications of data analytics (also known as machine learning or data mining) for production optimization in unconventional reservoirs by focusing on: 
  • Premises (easy-to-understand descriptions of the commonly used concepts and techniques)
  • Promises (case studies demonstrating successful practical applications)
  • Perils (honest appraisal of challenges and potential pitfalls)
Key concepts such as machine learning algorithms, predictive model building using automatically tuned parameters, cross-validation and variable importance will be introduced. Examples from the literature will be used to showcase how data analytics has been applied to study unconventional well performance, with an emphasis on the applicability of various techniques and their scalability. Finally, potential issues with over-promising and under-delivering—as is common with any new technology—will be discussed along with recommendations regarding best practices.
 
The date and time for the presentation haven't been set yet. Check the conference site for updates.