digilib@itb.ac.id +62 812 2508 8800

ABSTRAK Muhammad Fajar
PUBLIC Suharsiyah

Terbatas  Suharsiyah
» Gedung UPT Perpustakaan

Managed Pressure Drilling (MPD) is an adaptive drilling process used to precisely control the annular pressure profile throughout the wellbore. Managed pressure drilling is commonly used in drilling with a narrow pressure window. Managed pressure drilling uses lower mud densities than conventional drilling to maintain the wellbore pressure close to formation pressure (slightly overbalanced). This is why Managed pressure drilling is not only used in drilling with a narrow pressure window. The use of lower mud densities can increase the Rate of Penetration (ROP) in drilling, thus making drilling more efficient. In this study, Managed pressure drilling will be used to allow drilling with lower mud densities to increase rate of penetration and, this study will also predict rate of penetration when managed pressure drilling is used in wells with the same formation and section, namely in the carbonate formation and section 6 1/8” by creating a machine learning-based program using the light GBM algorithm. This study used 32086 clean data from 4 wells with the same formation in a field for training in machine learning models. The algorithm used as a machine learning model is light GBM with an R2 Score of 0.9368 with a predicted RMSE value that is smaller than the standard deviation value of the actual data used. This study resulted in an increase in rate of pentration when lower mud densities were used as a result of the application of managed pressure drilling.