Coiled Tubing Drilling ROP Optimisation
Challenge
A major Australian oil and gas company seeked to increase its drilling efficiency by optimising the rate of penetration (ROP) of its coiled tubing drilling operations while minimising downtime.
The company had identified that an improvement to ROP would have a significant improvement on the overall cost across their operations.
The company had identified key drilling parameters that drive ROP through engineering approaches. However, they had not succeeded in modelling ROP or downtime risk as a function of these parameters due to the complexity of the underlying relationships.
The company identified that a sophisticated approach may be required to model ROP and downtime risk, and generate real-time recommendations on drilling parameterisation in an operational setting. The company engaged us to provide our machine learning and statistical capabilities to support them in optimizing ROP.
Utilizing machine learning and statistical methodologies, we developed an algorithm which suggests the optimal differential pressure, focusing on maximizing the rate of penetration for the upcoming drilling segment while minimizing downtime risk.
This algorithm was implemented as part of a web application that provides real-time decision support to drilling engineers and operators on site on the basis of live drilling data. The web application recommends the optimal drilling parameters to achieve the best ROP.
The recommendations generated by the AI algorithm resulted in significantly faster overall drilling time.
The company is now using the real-time recommendations generated by the algorithm to optimise their ROP in their drilling operations.