Workflow of Managing Reservoir Uncertainties Through Merging Diagnostic Fracture Injection Test, Wireline Formation Tester and Petrophysical Logs: Application in Clastic Reservoirs in Multiple Wells

Abstract

Abstract Well potential evaluation is key to reservoir development which can be estimated from petrophysical logs, mobility from wireline formation tester, Diagnostic Fracture Injection Test, or measured directly from drill stem tests, production well tests, or through extended well testing. All these data are having different coverage of the reservoir volume which represent the scale of the data from low scale core and log data to large scale from DFIT or DST. All of the extracted mobility and permeability data are correct in their scale and the objective of this paper is address the data scale through a comprehensive workflow to adopt all of them into a reservoir dynamic model to maximize the data utilization. Wide range of available permeability estimation from different source imposed the question of data reliability toward different analyses. The established workflow guides to analyze all individual data sources including NMR (nuclear magnetic resonance), pressure transient analysis from single probe points and sampling stations, after closure analysis from DFIT, and pressure transient analysis from pressure build-up tests. It also details out the scale of the data and their radius of investigation. The derived permeability from each scale is used in static model with their relevant scale. Finally, it explores an approach to use the highest data scale such as DFIT upscale for improve static model and dynamic model reliability and predictability. Additionally, the workflow addresses the permeability transformation from effective permeability for few data sources such as DFIT, DST to absolute permeability as a reference for comparison for all permeability measurements. Defining right tested interval for wireline formation tester and vertical extension of DFIT is one of the main challenges that is tackled through integration with image logs and implementing a single well model. The workflow was implemented in four wells that were all completed with frack-pack as sand control application. The DFIT analysis was showing a lower effective permeability in most of the wells due to saturated reservoir condition compared to core and log results which was showing very high absolute permeability. Calculated permeability from DFIT has been transformed to absolute permeability using post-production saturation from petrophysical logs with integration with relative permeability dataset from SCAL data and close estimation of absolute permeability is made from DFIT to core and log driven permeability. A comparison of all permeability data sources in one single plot on each well is made to confirm vertical permeability contrast and reservoir heterogeneity to build reliable dynamic model and this narrow down the uncertainty in the dynamic model and de-risk the project. No measured permeability data is wrong. As long as the right analysis is done, they are all correct and their difference lies under the nature of reservoir property variation laterally and vertically due to rock type change. The reflection of data scale in 3D dynamic model is an essential step to boost dynamic model reliability, as the conventional approach of using permeability from core/logs to populate the permeability is more statistical than measurement driven. This work illuminates embedding the big scale data specifically DFIT results into dynamic model to enhance model reliability and predictability.