Artificial Neural Network Model to Predict Formation Penetration Rate in "T" Field

Tio Prasetio (PT Schlumberger)
Sonny Irawan (Petroleum Engineering Department, School of Mining and Geoscience, Nazarbayev University, Kazakhstan)
R. Hari Kariadi Oetomo (Usakti)

Journal of Earth Energy Science, Engineering, and Technology
Vol. 3 No. 3 (2020): JEESET VOL. 3 NO. 3 2020

Drilling is a costly activity with high risk. Time is a key variable to minimize costs and risks and increase the overall efficiency of drilling activities. An important factor related to the drilling time is the rate of penetration (ROP). The rate of penetration varies widely and is influenced by many factors. In this research, the correlation is derived using Artificial Neural Network (ANN) Model to predict the penetration rate by considering 11 parameters including formation conditions, drilling bit, drilling fluid, and drilling operations to validate the penetration rate data that are obtained from the surrounding wells. Determination of the neural network structure is carried out to obtain the best ANN model. This model produces an equation that can predict the penetration rate of the 'T' field with an error percentage of ± 20%. The existing model is used to optimize the next well drilling activity. Data processing using the ANN method which is relatively fast and precise shows that the application of this method is interesting to discuss and develop.