Drilling deviated wells is a frequently used approach in the oil and gas industry to increase the productivity of wells in reservoirs with a small thickness. Drilling these wells has been a challenge due to the low rate of penetration (ROP) and severe wellbore instability issues. The objective of this research is to reach a better drilling performance by reducing drilling time and increasing wellbore stability.
In this work, the first step was to develop a model that predicts the ROP for deviated wells by applying Artificial Neural Networks (ANNs). In the modeling, azimuth (AZI) and inclination (INC) of the wellbore trajectory, controllable drilling parameters, unconfined compressive strength (UCS), formation pore pressure, and in-situ stresses of the studied area were included as inputs. The second step was by optimizing the process using a genetic algorithm (GA), as a class of optimizing methods for complex functions, to obtain the maximum ROP along with the related wellbore trajectory (AZI and INC). Finally, the suggested azimuth (AZI) and inclination (INC) are premeditated by considering the results of wellbore stability analysis using wireline logging measurements, core and drilling data from the offset wells.
The results showed that the optimized wellbore trajectory based on wellbore stability analysis was compatible with the results of the genetic algorithm (GA) that used to reach higher ROP. The recommended orientation that leads to maximum ROP and maintains the stability of drilling deviated wells (i.e., inclination ranged between 40°—50°) is parallel to (140°—150°) direction. The present study emphasizes that the proposed methodology can be applied as a cost-effective tool to optimize the wellbore trajectory and to calculate approximately the drilling time for future highly deviated wells.