Data Integration for Reservoir Modeling


Conceptual Design and Workflow




• Reservoir envelope: top and base structure




• Internal framework: correlation scheme




• Reservoir compartments: fault geometry




• Reservoir architecture: facies model




• Petrophysical property distribution




• Volumetric Assessment




When planning a reservoir modeling project, the ultimate purpose of the model must be defined. To this end the geological model should address all of the above points. On the first day, participants will learn about the retention of relevant fine-scale detail through upscaling. The process of building a 3D reservoir model will always follow the same general workflow, regardless of the tools available to the modelling team. Each of the steps will be outlined in the course with an appreciation of the required input data, associated uncertainties, and likely deliverable and its use. The database used for the exercises will be from a real field that includes two different clastic reservoir types with different modeling challenges.




Reservoir Framework





• Depth conversion uncertainty




• Model surface selection and quality control




• Fault modelling and compartments




• Stratigraphy and correlation




• Grid construction




In volumetric models the greatest uncertainty is usually the gross rock volume at the top structure map and the hydrocarbon contact. Depth conversion, where well data can be sparse, leads to much of this uncertainty. Structural models that result in over complexity, much of which cannot be modelled, may drive seismic interpretation. Participants will learn how to decide what structural elements to include, which can be a source of much debate. Likewise too much well-to-well correlation can over science a model especially when still in development and drilling surprises are common, so it will be covered on this day as well.




Reservoir Architecture





• Depositional models and facies analysis




• Core-log integration




• Basic statistics




• Objects and indicators




• Seismic conditioning




• Facies modelling




Participants will learn about the different clastic and carbonate depositional environments and how to best characterize them for subsequent modelling. The dependence on statistics will be introduced as a way to demonstrate the different methods of facies modelling available.




Property Modeling in 3D





• Basic petrophysics




• Rock typing




• More basic statistics




• Porosity models




• Saturation models




• Permeability models




Running through the course will be the theme of petrophysics, always calibrated to the geology, as a way to distribute reservoir properties. Participants will be shown simple rock typing methods that are readily applicable to 3D models. Particular attention will be paid to saturation height methods that accurately distribute fluids through the model.




Uncertainty and Upscaling





• Geological model analysis




• Hydrocarbon volumes initially in place




Drainable volumes




• Simulation grid construction




• Property upscaling




• Multiple scenarios, realizations, and ranking




What makes a good static model will be discussed. The methods of interrogating and analyzing the results described before addressing the thorny question of upscaling for dynamic simulation will also be covered.