Trending Content

Prediction of Geomechanical Parameters in Real-time Using Machine Learning while Drilling

Add to Cart
Course Credit: 0.15 CEU, 1.5 PDH

The geomechanical properties of the subsurface formations are considered essential parameters in the optimization process of the engineering practices in the oil and gas industry. Since it controls the mechanical behavior around the wellbore, it plays a direct role in the drilling operation and the associated wellbore instability issues that may cause many drilling-related incidents, i.e., lost circulation, pack off, and pipe sticking. The mechanical behavior of rocks can be studied by determining several geomechanical parameters i.e., static elastic modulus (Es), static Poisson’s ratio (PR), unconfined compressive strength (UCS), etc. Such data are the cornerstone for developing a representative geomechanical model of the subsurface, whereby a broad suite of problems can be avoided by providing practical solutions to the wellbore instability issues during drilling. These solutions include determining the optimum mud weight, defining the safe drilling window, specifying stable trajectories, determination of casing setting depths, etc.

The most accurate method to obtain such measurements is retrieving core samples from the downhole formations that would be later subjected to different experimental tests, i.e., triaxial tests, to estimate these parameters. However, this process is expensive and laborious which makes such measurements not usually available for all drilled wells. This was the motivation to employ the machine learning (ML) techniques to develop new approaches by which these parameters can be estimated in an economically and timely effective way. Several models were developed using different ML techniques i.e., artificial neural network (ANN), support vector machine (SVM), adaptive neuro-fuzzy inference system (ANFIS), etc., to estimate these parameters using logging data. Furthermore, many researchers exceeded this limit and succeeded in predicting these parameters during the drilling operation in real-time using the drilling parameters with high accuracy. This, in turn, would effectively help take better field decisions regarding the design and the optimization of the drilling process. All content contained within this webinar is copyrighted by Dr. Salaheldin Elkatatny and its use and/or reproduction outside the portal requires express permission from Dr. Salaheldin Elkatatny.

SPE Webinars are FREE to members courtesy of the

Post Tags

 1 chapter
  1 downloadable resource

Course Chapters

  • 1Prediction of Geomechanical Parameters in Real-time Using Machine Learning while Drilling - Chapter 1
    Media Type: Video

Credits

Earn credits by completing this course0.15 CEU credit1.5 PDH credits

Speakers

Dr. Salaheldin Elkatatny