Alleviating Directional Well Trajectory Problems via Data Analytics

Abstract

A consistent leading cause of drilling non-productive time (NPT) is the inability to steer the planned well trajectory trouble-free. Separate from downhole tool and drill bit failures, an unplanned trip to change the Bottom Hole Assembly (BHA) is required for up to one in every seven drilling runs. Root cause analyses indicate potentially a quarter of all drilling NPT has poor planning or field execution as the failure mechanism, signifying scope for improvement. This paper aims to help guide optimal selection of RSS/motor and bit, to ensure challenging wellpaths will be achieved with minimal NPT associated with BHA trips. Directional drilling analysis typically compares dogleg severity (DLS) for planned and actual trajectory. This metric is fundamentally direction-blind; absolute tortuosity is represented whether planned or unintentional. Without full context, DLS analysis can mask many steering issues. Typically, industry software does not measure how closely the steering inputs match their anticipated responses during a run. Strategic management and identification of zones with erratic toolface control, or strong formation/BHA tendencies is critical. The proposed ‘derived steering’ analytics method was applied to plan demanding 3D trajectories for an Extended Reach offshore campaign in Australia. Existing minimum curvature equations were repurposed to plot previous runs steering inputs and then infer efficiencies for each formation. Supervision was essential to counteract strong consistent right-hand BHA walk tendency for all the variety of wells studied. Multiple NPT events on previous campaigns had resulted from poor steering response in the shallow interbedded geology. In view of quantifiable field-specific risks, wellplans were refined to minimize tortuosity and maximize the design safety factor. The combination of highest anticipated dogleg response rotary steerable technology and bit selection was selected for steering assurance. Modelled tendencies per lithology were shared with wellsite supervisors, and recent drilling results essentially mimicked data analytics. Others operating in this field in the 21st century had drilled total meterage of 36,740m MD from 83 runs. Bit Gradings showed two ‘Lost in Holes’, one ‘Drill String Failure’, six trips for ‘Downhole Tool Failures’, seven for ‘Penetration Rate’, six to ‘Change BHA’, two for ‘Hole Problems’ and one for ‘Downhole Motor Failure’. The current campaign's improved directional drilling offset analysis contributed towards significant avoidance of well delivery NPT to drill 28,061m in 34 runs. No trips were required to change BHA or bit because of inability to follow the trajectory, and field teams were able to pre-empt lithology-specific challenges.