Geostatistical Model Constrained by Process-Based Simulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing geostatistical models lack integration of process information, leading to approximate and impractical conditioning, while process-based models fail to accurately incorporate conditioning data, resulting in a disconnect between geological processes and reservoir properties.
Innovation Solution
A method is developed to generate a geostatistical model by integrating process-based modeling with conditioning information, where a process-based model is constrained to match criteria derived from conditioning data, allowing for stochastic and deterministic parameter determination, and subsequent generation of statistics that reflect both process and geostatistical properties, enabling accurate representation of geological volumes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If process-based models are used to represent geological processes, then process information is captured, but accurate incorporation of conditioning data becomes impractical
Solution Approach 1:
The model is segmented into distinct process-based components (deposition, erosion, non-deposition) that can be independently conditioned. Each process module handles specific geological mechanisms, allowing conditioning data to be applied to individual processes rather than the entire model at once, making conditioning practical while preserving process information.
Solution Approach 2:
An intermediary conditioning framework is introduced that translates conditioning data into process-specific constraints. This intermediary layer bridges the gap between raw conditioning information and the process-based model, enabling accurate data incorporation without compromising the process representation.
2Ease of manufacture
If geostatistical models are used to incorporate conditioning data, then conditioning is improved, but process information is lost
Solution Approach 1:
The model uses parameter changes to represent different geological processes. By varying process parameters (deposition rate, erosion rate, non-deposition duration) while maintaining the underlying process framework, the model incorporates geostatistical conditioning without losing process information. The parameters are adjusted based on conditioning data while preserving the process-based structure.
3Loss of information
If process-based models are used, then geological processes are represented, but accurate matching to conditioning data becomes approximate
Solution Approach 1:
The model employs dynamic process parameters that can be adjusted during conditioning. Rather than static process representations, the model allows parameters such as deposition and erosion rates to be dynamically modified to match conditioning data, improving accuracy while maintaining process representation.
Solution Approach 2:
A feedback mechanism is implemented where the model compares its output against conditioning data and adjusts process parameters accordingly. This iterative feedback loop enables the process-based model to accurately match conditioning data by continuously refining process parameters based on the difference between model predictions and observed data.
Data Source
Figure 1
Figure 2
AI summary
A process-based model of a geological volume of interest is generated. The process-based model is conditioned with conditioning information associated with the geological volume of interest. Statistics are generated from the process-based model that represent parameters of the geological volume of interest locally. These statistics are used to constrain one or more geostatistical models of the geological volume of interest.