Implicating of Cation Exchange Capacity in Rate of Penetration Predictive Model for Tanuma Shale Formation/Garraf Oil Field Using Artificial Neural Network

Authors

  • Ahmed Habeeb Alshamy, Faleh Hassan Almahdawi

Abstract

Shale and shaly formations constitute about 70% of the total rock formations drilled worldwide. The rate of penetration (ROP) is commonly decreased due to serious drilling problems occurring in such formations, such as wellbore instability and bit balling problem. Notoriously, the ROP is one of the main drilling parameters that strongly affected the cost of drilling operation. Thus, it is crucial to optimize ROP while drilling to avoid drilling problems. Generally, these problems are caused by the chemical reactive nature of shale formations, leading cuttings of rock to accumulate and stick to the bit. Consequently, ROP decreases due to such an undesirable effects. Cation exchange capacity (CEC) is a shale property which reflects the electrochemical properties of clays and characterizes the potential of shale rocks including clay minerals to exchange ion leading to shale hydration. In this study, CEC was correlated reasonably with ROP using artificial neural networks (ANNs) to develop a model of ROP prediction. The ANNs model was developed as a function of the most critical parameters affecting ROP such as: torque (TQ), weight on bit (WOB), rotation per minute (RPM), flow rate (Q), depth (D), equivalent circulating density (ECD) and (CEC). The CEC values were estimated using common log data for Tanuma shale formation. Training, validation and testing was performed to the developed ANNs model of ROP prediction. The results showed that the correlation coefficient (R2) and absolute percentage error (AAPE) was (0.961) and (0.456%) respectively, and this indicates the high accuracy of ROP-predictive model. Unlike previous studies, this paper includes the influence of CEC on the ROP in shale formation.

Published

2020-11-01

Issue

Section

Articles