The project will through experiment and analysis with advanced mathematical and numerical methods achieve a better understanding of the processes and mechanisms related to fluid transport in the well. Such understanding may be used to build better simple real-time models with longer predictive capabilities, and will also help in building relevant and reliable models for calibration and tuning purposes.
The AdWell CFD framework will need validation to build confidence in the modelling results. Experiments will be performed to generate data sets for analysis and model verification. Already existing process data will be applied where available. For this purpose flow laboratories at the University of Stavanger and at NTNU will be used. Adaptations of the laboratories will be made to enable the testing required. Experimental data made available through described collaboration will also be applied where relevant.
Measurements at UIS UiS Flow Loop
A set of experiments will be performed with-in the project, and some data are available from other sources.Necessary measurements will be performed using UiS Small Scale Flow Loop ( Small Scale as well as Medium Scale Flow Loop)
NTNU Stretch Rig
The stretch rig at NTNU is intended to model normal vs. deviatory hook load behaviour, and characterisation of hook load when the drill string is pulled out from the well-bore. The horizontal tripping experiment wil also a possible CFD use-case. In that context, it is important with good measurements of the flow rates of mud and particles.
CFD MATHEMATICAL AND NUMERICAL METHODS
Advanced numerical methods, including 3D Computational Fluid Dynamics (CFD) methods and also Direct Numerical Simulation (DNS) where applicable, will be applied to study the problems in detail.CFD Use Cases
A use case, in our context, is a description of as physical system that should be modelled using CFD. Use cases are selected based on their ability to shed light on fundamental transport processes taking place during drilling operations. We are, of course, particularly focusing on use cases that may provide new insights into the underlying physics, thus to obtain understanding that can enable construction of improved real time models.
SIMPLE REAL-TIME MODELS
Improved and new simplified models for use in real-time drilling applications may be generated based on the developed advanced models and understanding gained. Once the advanced models are developed, recommendations are to be made for real- time models for automation, based on experimental and modeling results.