Using Big Data to Spotlight HFTO

Abstract

Abstract High frequency torsional oscillation (HFTO) is a major contributing factor to the drilling dynamics-related bottomhole assembly (BHA) failures. These failures are costly because they not only damage the drilling tools but also cause significant nonproductive time (NPT) to the operational activity. It is therefore highly desirable to measure and mitigate this dysfunction in real time. We have built a high-performance drilling dynamics module, which enables real-time or recorded mode measurements of this HFTO phenomenon. Using this measurement, we can deliver actionable insights to the rig or the rig crew in real time. With data from more than 3000 runs, we can build a clearer understanding of the real characteristics of HFTO, its drivers, and the effects it has on our drilling systems. We show that with increasing Weight on Bit (WOB) the amplitude of the HFTO will increase. However, at high WOB the dysfunction amplitude will drop, while the ROP continues to rise. As such there is a sweet spot with high ROP and low HFTO. By integrating this database of drilling dysfunctions with our records of tool damage and nonproductive time, we can map the effect of HFTO onto different failure criteria. As a result, we can define new and better operational standards and generate real-time insights into what damage is more likely to occur and how to change drilling parameters, if needed, to prevent it or determine if the vibration will pass without incident. We also show that although the rotary steerable tools we use are sensitive to the effect of HFTO, our measurement while drilling (MWD) tools are not. For this latter group the probability of damage is the same for runs with and without HFTO. This paper discusses the method and results for this study.