Management of Drilling Fluid Circulation

Abstract

To maintain the safety of well control operation, especially for the prevention of uncontrolled blowouts, early kick detection is extremely important in the drilling rigs. From very early stage of rig operation, a sudden increase in pit level has been interpreted as an effective kick signal detection mechanism. However, pit level fluctuates at different period of the operation, which may mask a kick. The objective of this project is to develop a method to determine the nature of active pit level variation anytime during the operation so that it would be possible to track down any kind of undesired behaviour of the pit level.To achieve the convictable objective, a traditional and fundamental method was employed to develop mathematical models for different circulation condition and generate pit level data from these models. Pump conditions and tripping are known to be most responsible factors for pit level fluctuation. Therefore, appropriate mathematical model has been developed for these processes. Later,the developed models are modified to take into account different factors concerned with the processes. All the developed models are coded into FORTRAN, which made it possible to generate effective and reliable pit level data for any kind of operation. Data generated from the above procedure are plotted using MATLAB for visual representation of the behaviour of the pit at a specific rig condition. The developed model has been compared with the ideal system in order to gain knowledge about the significance of different processes and factors that affect the pit level variation. Finally, mathematically generated data are compared with the real-time pit data in order to test the effectiveness and reliability of the models. It has been observed that the developed models can successfully determine the nature of fluctuation and predict the pit level behaviour in an average manner. However, the developed models are failed to generate 100% reliable data for different period of circulation. The possible reason for the failure might be the assumptions made to simply the models.Therefore, more work is needed to take into account different pit conditions and circumstances for establishing the best model that can accomplish the task of making the well control operation safer.