MPD choke control considering the thixotropic gelation behavior of drilling fluids

Abstract

This study evaluates control algorithms that manage choke pressure during managed pressure drilling applications (MPD) while also taking the thixotropic behavior of drilling fluids into account. This evaluation analyzes proportional-integral (PI) feedback control and proposed preview-based feedback control for choke pressure. A pressure spike can be observed in standpipe and bottomhole pressure after making a drillstring connection or experiencing a prolonged break in drilling operations. This is caused by the gelation of drilling fluids. It is vital to be able to predict the magnitude of such pressure spikes when drilling with active annular pressure control, such that the choke can be effectively controlled to mitigate an induced non-productive time event (i.e., lost circulation events, pump or pipe failures, or well control situations). In this study, a shear-rate and time-dependent thixotropy model is integrated into a rigorously validated transient hydraulics engine with MPD control capabilities. Drilling scenarios are simulated with choke controllers and a case study is presented. When using a preview-based feedback control algorithm, a physics-based model is executed while setpoints are established proactively. The combined capabilities of the physics-based model and the improved MPD controller are demonstrated. Accurately modeling and controlling pressure during drilling operations is essential, especially for wells drilled in narrow-margin environments. The results of our study show that a reactive PI controller is not fast enough to account for transient pressure behavior related to fluid thixotropy. Moreover, the results show that a preview-based feedback controller is more robust and efficient. We demonstrate how a physics-based model can be used in a preview-based control scenario to keep the bottomhole pressure constant with strongly attenuated pressure fluctuations, leading to safer, more optimized operations.