Quantifying the Risks of Wellbore Failure During Drilling Operations Using Bayesian Algorithm

The wellbore integrity plays an important role in petroleum operations like drilling, well completion and production. Caliper, Electrical image logs, Acoustic televiewers (ATV) and Optical televiewers (OTV) are some of the devices currently used in the industries to identify breakouts. However, these techniques are restricted in applications. For instance, caliper sometimes indicates the effect of drill spiral grooves as borehole enlargement zones; poor resolution and complicated processing procedure limit the application of the electrical image logs. The ATV and OTV which give better outputs are not often used due to the high cost of installation and operation. These limitations necessitated a new approach to quantifying the wellbore instability. This research work focuses on data analytics and the development of the Bayesian Algorithm (with code in Python) to predict the wellbore failure probability using real-time pore pressure and fracture gradients data obtained from the wellbore using modified d-exponent.