A Robust Correlation Improves Well Drilling Performance

Abstract

Utilization of drilled wells operations’ records is required to perform improvement of performance to minimize drilling cost of planned drilling of new and re-entry wells (workover – wells). Many operators are always interested in finding optimum ways to save on drilling cost. Optimization of Rate of Penetration (ROP) has direct effects on cost reduction. Different Techniques were used to optimize ROP such as regression technique, multiple linear regression technique, neural network, artificial neural network methods, and basic reference of Bayesian networks. There are several factors that will limit application of ROP optimization models in different hole sections with high degree of accuracy. It is the authors’ opinion that modeling on smaller selected section with controlled parameters will give better optimization and validation. In this paper an empirical correlation of rate of penetration (ROP) is presented for a particular hole section. The data selected are from same hole size, formation type and mud type. It is based on monitoring and controlling simultaneously the applied weight on bit (WOB), drill-string's rotation (RPM), Torque (TRQ) and rig pump's flow rate (GPM). During this study will demonstrate the use of this empirical correlation to improve drilling in a hole section by more than 50%. The developed model also has high potential to be automated in real time operating environment to improve drilling performance.