Quantification of Drillstring-Integrity-Failure Risk Using Real-Time Vibration Measurements

Summary Vibration measurements have traditionally targeted the improvement of downhole-tool reliability. This paper targets the effects of vibration on the complete drillstring. Failures associated with drillstring vibration continue to happen despite the sophistication of today's measurements. These failures represent a very significant amount of lost time, which we target to improve. The industry has a very limited database for evaluating indices to manage or quantify risks of vibration to the complete drillstring. This fact makes the use of the methods in the field heavily depend on the past experience of the drillers and on the rig types. Operators are faced with an unknown quantification of the risk severity when attempting to mitigate vibration. By quantifying the risk, this work demonstrates how the prevention of incidents can be achieved. These incidents include but are not limited to, twistoffs, backoffs, and bottomhole-assembly (BHA) component failures. The proposed solution is based on real-time measurements of drillstring vibration to estimate an ongoing drillstring-integrity risk, which is used as a guideline to improve decisions while drilling the well. This solution has been developed through use of advanced vibration sensors to discriminate between different types of vibration. This was critical to estimating realistic cumulative damage to the drillstring, which is highly dependent on the type of vibration suffered by the assembly and the onset of vibration-mode coupling. This paper shows that approximately 80% of drillstring-integrity failures analyzed can be identified and prevented through use of the proposed risk-quantification solution. This result has been obtained despite unknown fatigue or wear of drillstring components before a run, and vibration sensors were located at a single position in the drillstring. This indicates that the primary contributor to drillstring failures is the drilling conditions for any given run.