Field Application of Image Analysis Models to Measure the Drill Cuttings Recovery Rate

Abstract This paper presents the development, rigorous validation, and practical application of an image analysis model combined with a remotely monitored and autonomous camera system to measure the recovery rate of drilled cuttings. The primary objective was to assess the effectiveness of hole cleaning, a crucial factor in preventing well construction issues. The model's reliability and accuracy were successfully demonstrated during the continuous monitoring of five wells in the Permian Basin. An optical sensor system with onboard image analysis software was installed inside the hazardous zone near the shale shaker and used to monitor and analyze returning drill cuttings while drilling 6.75-inch lateral hole sections. The varying cutting loads were measured quantitatively and qualitatively using computer vision techniques, including optical flow. The images were analyzed onsite, and data on actual and estimated shaker loads was delivered. The load was measured on only one of three shakers. The project was managed remotely using a multi-platform, open-source analytics and interactive visualization web application. Before deployment, the image analysis model was calibrated and verified against an experimentally validated cuttings transport model by performing experiments at different cutting load rates, volumes, and fluid types using a drilling research flow loop. This calibrated model was deployed during drilling to measure the cuttings recovery rate (CRR) on a relative basis (0-10) and an actual basis (barrels per hour). Approximately 500 bbl. of drilled cuttings were optically measured and compared with expected volumes. The relative cuttings load demonstrated a credible correlation during controlled, lower rate of penetration (ROP) and sweep circulation periods. The actual CRR showed logical results during daylight periods. Additional observations are reported concerning wet versus dry shaker conditions, performance under low light conditions, and the identification of potential time savings based on measured shaker loads. Implementing an image analysis model to analyze drill cuttings in real-time—including autonomously characterizing a drilled cutting’s size, shape, and volume—represents a significant leap toward drilling faster and reducing risk. This promising technology has the potential to significantly reduce the non-productive time associated with circulating and mitigating pack-off events, effectively addressing inadequacies in current cuttings transport modeling and hole-cleaning practices. The experimental validation work and field examples reported here demonstrate the progress towards applying newly developed image analysis capabilities to successfully measure drill cuttings recovery rates, underscoring the importance of this research in the oil and gas industry.