The first step towards drilling optimization and automation is a reliable data acquisition and handling system. This includes receiving and processing different frequency data across multiple platforms and ensuring proper data quality. Once we implement such a platform, we can build different advisory solutions to improve drilling efficiency and move towards drilling automation. The developed Drilling Intelligence Guide (DIG) already enabled us to achieve the aforementioned goal and access different data (from contextual to high frequency) in real-time.
In the next phase, different models (physical or data analytics) can be built in to optimize various stages of drilling operations. The drilling industry has made significant progress on different physical models that can be run offline (either for pre-job design or post-job analysis) in the previous decades. Given the developed data platform, we can now adopt and run these models in real-time for optimization and automation purposes. As an example, a real-time, computationally efficient hydraulic model that can account for drill string rotation and eccentricity would enable us to monitor ECD (to be used for wellbore stability, kick and lost circulation mitigation) and calculate pump pressure corresponding to different operational conditions (to be used for drill string washout/pump failure prediction and sensor calibration).
The aim of this paper is to show the application of the hydraulic model for real-time monitoring and optimization purposes. An analytical hydraulic model including drill pipe rotation and eccentricity effects is used and compared against transient numerical simulations. In the next step, Pressure While Drilling (PWD) data are used to verify the accuracy of the model for real-time applications. After the verification steps, we explain the process to couple real-time data (e.g., flow rate, RPM), contextual data (e.g., mud properties, BHA and wellbore geometry) and the physical model using field examples. In addition to ECD monitoring at the bit and casing shoe, the real-time hydraulic model can also be used to monitor washout and pump failure events. The application of the model is shown using field examples.