Book Description

As well designs become increasingly complicated, a complete understanding of drillstring vibrations is key to maximize drilling efficiency, to reduce drillstring dysfunction and to minimize drillstring, tool, and borehole damage. Torque and drag models exist that seek to quantify the effects of borehole inclination and tortuosity on static friction along the drillstring; however, the effects on dynamic friction remains poorly understood. This dissertation begins with a review of the past fifty years of work on drillstring dynamics models, an overview of the proposed control strategies and a summary deployed vibration mitigation applications within the drilling industry. Derivations from first principles of a series of computationally efficient axial and torsional drillstring models in both the frequency and time domains are then presented and verified with field data. The transfer matrix approach is used to predict the severity of axial vibrations along the drillstring and is verified using a series of case studies using field data. Harmonic axial vibrations within drillstrings are either induced intentionally, in the case of axial oscillation tools midway along the drillstring, or unintentional, in the case of bit bounce. Two case studies of bit bounce are first evaluated to ensure model validity for a harmonic excitation at a the bit and the model is found to accurately predict bit bounce based on surface rotation rates. Induced axial oscillations, generated by axial oscillation tools, are then investigated to quantify friction reduction and drilling efficiency improvements. Optimal placement is found to depend on wellbore geometry, but is usually restricted to periodic regions of the drillstring. These optimizations are then verified using field trials and suggest that improved placement can result in 20% or more reduction in friction along the drillstring. Two applications of torsional drillstring vibrations are then investigated -- stick slip mitigation and drillstring imaging. The time domain form of the torsional drillstring model is used first to evaluate the effectiveness of three types of top drive controllers -- stiff controllers, tuned PI controllers and impedance matching controllers -- in mitigating stick slip oscillations. Then, the transfer matrix method is applied to evaluate the effect of wellbore geometry on drillstring mobility to conclude that higher order modes of stick slip may become dominant in non-vertical wellbores. The feasibility of drillstring imaging using torsional signals from surface is then investigated to identify inputs and methods that show promise in three setups of varying complexity -- a hanging beam, a laboratory drillstring model and a drilling rig. Two techniques show promise -- white noise injection and model fitting of a step response -- in identifying larger features, including drillstring length and BHA location. However, low sampling frequencies and low bandwidth inputs reduce the ability to image small features such as friction points along the wellpath.