Bayesian MCMC Reservoir History Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current history matching methods for reservoir simulation models are inefficient due to high computational costs, uncertainty quantification challenges, and inconsistency, particularly relying on human interpretation and being affected by prior selection and likelihood calculations.
Innovation Solution
The implementation of a Bayesian Markov Chain Monte Carlo (MCMC) workflow that includes selecting a reservoir simulation model, identifying relevant mathematical models and history matching parameters, constructing low-fidelity models to update priors, and generating posteriors to determine the accuracy of the reservoir simulation model, thereby reducing computational costs while maintaining physics consistency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional trial-error approach is used to adjust multipliers, then human interpretation is required, but computational cost is reduced and ease of operation is improved
Solution Approach 1:
The patent replaces the traditional mechanical trial-error adjustment process with an automated Bayesian MCMC workflow. The system automatically performs prior selection, likelihood calculation, and posterior sampling without requiring manual human intervention in each iteration, thereby maintaining ease of operation while significantly improving productivity through automation.
Solution Approach 2:
The Bayesian MCMC system performs self-service by automatically selecting priors, calculating likelihoods, and generating posteriors without requiring continuous human guidance. The workflow manages its own iterations and converges on history-matched models autonomously, eliminating the need for manual multiplier adjustment while maintaining operational simplicity.
2Productivity
If modern computerized algorithm approach is used, then productivity is improved, but computational cost increases and device complexity increases
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting the complexity of the Bayesian MCMC workflow based on the specific reservoir simulation model being analyzed. The system adapts the number of iterations, prior distributions, and likelihood calculations to match the computational requirements of each model, thereby improving productivity while controlling computational costs through parameter optimization.
Solution Approach 2:
The workflow is designed to be dynamic, automatically adjusting its computational intensity based on the complexity of the reservoir model and the available data. The MCMC process adapts its sampling frequency and iteration count to achieve convergence efficiently, thereby maintaining high productivity while optimizing computational resource usage rather than requiring fixed high computational costs.
3Productivity
If modern computerized algorithm approach is used, then productivity is improved, but device complexity increases
Solution Approach 1:
The patent segments the complex history matching process into distinct modular components: prior selection, likelihood calculation, and posterior sampling. Each component is handled by dedicated modules within the Bayesian MCMC workflow, which simplifies the overall system architecture and reduces device complexity while maintaining high productivity through automated processing of each segment.
Solution Approach 2:
The patent introduces an intermediary layer in the form of the Bayesian MCMC workflow that mediates between the reservoir simulation model and the history matching process. This intermediary framework automatically handles the complexity of parameter adjustment and model calibration, thereby improving productivity while containing device complexity within a standardized, manageable workflow structure.
4Measurement precision
If history matching is performed with high accuracy, then measurement precision is improved, but computational cost increases
Solution Approach 1:
The patent implements feedback mechanisms within the Bayesian MCMC workflow where the posterior distributions from each iteration are fed back into the likelihood calculation for the next iteration. This feedback loop continuously refines the history matching precision while the automated convergence criteria prevent unnecessary computational iterations, thereby achieving high measurement precision without proportionally increasing computational costs.
Data Source
AI summary
A method for history matching utilizing Bayesian Markov Chain Monte Carlo (MCMC) workflow may include selecting a reservoir simulation model of interest, identifying a mathematical model relevant to the reservoir simulation model, and identifying a plurality of history matching parameters as initial priors. The method may include constructing a first model, utilizing the initial priors, to obtain updated priors. The method may include constructing a second model to obtain posteriors. The method may include determining history matching accuracy of the reservoir simulation model by comparing medians of the posteriors and a plurality of measured data. The method may further include, upon determining accuracy of the reservoir simulation model, performing a plurality of predictions of a reservoir.