Data Visualization Workflow an Alternative Procedure to Quality Check Static Grids

Le résumé fourni par la source

Abstract Static grids are commonly an ensemble of millions of reservoir property values. Analyzing this vast dataset is a challenging task and is usually performed via simple statistical parameters (e.g arithmetic mean and standard deviation). The objective of this paper is to present a quick and efficient way of analyzing static model realization results in a more visual, interactive and efficient manner allowing a multidisciplinary team to absorb information and key ideas quickly. The proposed methodology starts with a set of properties from a given static grid realization. The data are prepared with all necessary labeling and categorization to be analyzed (e.g. reservoir and segment indicators, well regions etc.). All kinds of data variations available in the static model are gathered and mobilized into dynamic data dashboards for further analysis. The interactive data visualization templates provide flexibility to filter information either by categories or data ranges. The filtering schemes are immediately propagated to all available plots in the dashboard highlighting common patterns from all the data relationships in the static model. The principle output from this methodology is an interactive data visualization panel that displays all necessary infographics related to reservoir data-trends. Starting from simple rock quality pie-charts, which can be mapped showing field-wide trends, porosity-permeability plots, to more detailed infographics of property ranges per well/reservoir indicator as well as summary tables of average properties per reservoirs/layers/segments etc. The ability to have all plots linked in a single display provides a simple but powerful platform to evaluate and interrogate data from the static grid using different views. Data can be aggregated in different ways and summarized accordingly. With this continuous process of data filtering, it is possible to quickly identify outliers and collect evidence of data inconsistency within the static grid. The additional value of the presented approach is the ability to compare multiple property realizations in a single template and summarize similarities and differences between different static realizations. The implementation of reservoir property dashboards proved to be an effective practice to increase the understanding of distributed properties inside static grids. A more comprehensive and detailed data quality check can be performed in multiple properties using a single view of the structured dataset. Additionally once data is loaded into the data visualization platform it can be easily shared with team members without the need of any specialized modeling software.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.