Abstract
Abstract This study investigates the application of high-speed videography as a novel, non-invasive method to analyze drill string vibration dynamics under real-world drilling conditions. Utilizing a laboratory-scale drilling rig, experiments were conducted to simulate operational scenarios, varying parameters such as rotary speed, rock type, and drilling fluid. Camera captured axial, lateral, and torsional vibrations with high spatial and temporal resolution. Video data were processed using MATLAB (on-line) and OpenCV, employing optical flow algorithms to quantify vibration frequencies (5–50 Hz) and amplitudes (0.1–2.5 mm). Results revealed distinct vibration characteristics influenced by formation type and drilling parameters, with sandstone inducing higher amplitude lateral vibrations (up to 2.5 mm) compared to shale (1.2–1.5 mm). Vibrations correlated with a 25% reduction in rate of penetration and 30% increase in tool wear in high-vibration regimes. Compared to traditional accelerometers, videography offered spatial coverage and comparable accuracy (3–8% error), highlighting its potential for real-time monitoring. Study demonstrates high-speed videography’s efficacy in enhancing drilling safety and efficiency, proposing its integration into monitoring systems and future automation via machine learning. Recommendations include optimizing rotary speeds and fluid properties to mitigate vibrations, with applications extending to unconventional drilling operations like geothermal wells.