Digital Shaker Surveillance Uses Computer Vision on a Deepwater Rig

_ This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper OTC 35585, “World-First Digital Shaker Surveillance Using Computer-Vision Technology on a Deepwater Rig,” by Calvin Holt, SPE, DrillDocs, and Paolo Patti and Silvia Haiz, Eni. The paper has not been peer reviewed. _ This paper describes a trial deploying and operating a computer-vision system on a deepwater rig to measure drilled cuttings in real time using a remotely monitored camera system. By autonomously characterizing drilled cuttings, the system aims to enhance drilling efficiency, optimize penetration rates, and prevent packoff events. The successful application of image analysis to digital shaker surveillance represents a promising advancement in cuttings-recovery-rate measurement. Problem Statement Finding the optimal combination of drilling parameters and mud characteristics to avoid the excessive accumulation of cuttings in the wellbore and associated risks can be a major challenge. The situation is worsened if rock fragments of larger dimensions (known as cavings) unexpectedly fall into the wellbore from the borehole wall. Such wellbore instability can be caused by insufficient circulating pressure or excessive circulating pressure while drilling. Both the accumulation of cuttings and cavings have consequences when tripping the drillstring or casing into or out of the well. Actions to mitigate accumulations of cuttings and cavings differ significantly, so correctly identifying the root cause of the problem is critical. Possible causes of wellbore instability include purely geomechanical reasons, overpressure in low-permeability shales, and pre-existing planes of weakness or prefractured rock. The drilling team’s ability to react quickly and appropriately to borehole conditions could be enhanced by a system that can continuously measure the quantity of cuttings being returned and quickly detect cavings. The more quickly and reliably deviations in cuttings-return rates or the presence of cavings can be reported, the more rapidly and effectively the drilling team can respond. Optical sensors were considered because of their advantages over cuttings-flow meters, which mechanically collect and weigh the cuttings on an intermittent basis. A camera is easier to install, provides continuous remote monitoring, reduces risk (e.g., injury caused by moving mechanical parts), and can make different measurements simultaneously. The operator postulated that the ability to cross-reference visual data from the shakers with mud-logging data would provide its drilling team with a more-comprehensive understanding of the operational situation, enhancing their ability to manage wellbore stability and ensure the effectiveness of hole-cleaning operations. Approach The system uses computer-vision algorithms and machine learning to interpret images captured by cameras focused on the shale shakers. This enables real-time monitoring of cuttings frequency, size, and shape, from which cuttings volume can be calculated. When integrated with drilling parameters from the rig’s data system and mud-logging information, this system provides key insights about wellbore stability and hole-cleaning efficiency. The operator recognized the potential of this technique and launched a pilot project to evaluate its application on a deepwater rig. The test was conducted during 2023–24 on a seventh-generation drillship during the drilling of subsea wells offshore Egypt and the Ivory Coast.