Optimizing Conditioning Data in Multiple Point Statistics Simulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional Multiple Point Statistics (MPS) simulation methods require excessive computation time due to the large number of conditioning data used, which is necessary for accurate geological feature reproduction, but reducing this data degrades simulation quality, necessitating a method to optimize the number of conditioning data for minimized computation time while preserving pattern reproduction quality.
Innovation Solution
A method that estimates the value of additional conditioning data for inferring local facies probabilities and defines a threshold beyond which additional data do not significantly modify these probabilities, using a search tree to store patterns and compute facies probabilities efficiently, allowing for a reduced template size that balances computation time and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the template size (number of conditioning data) is increased to improve geological feature reproduction quality, then the accuracy of facies probability estimation is improved, but the computation time increases exponentially
Solution Approach 1:
The patent applies partial action by determining the minimum sufficient template size required to achieve accurate facies probability estimation. Instead of using excessively large templates that guarantee accuracy but incur exponential computation costs, the method identifies and uses only the necessary number of conditioning data points. This is achieved through iterative testing where template sizes are gradually increased until the facies probability estimates converge, at which point the process stops, avoiding unnecessary computation with larger templates.
2Productivity
If the template size is reduced to decrease computation time, then the simulation speed is improved, but the training pattern reproduction quality deteriorates
Solution Approach 1:
The patent applies preliminary action by pre-determining the optimal template size through a calibration phase before actual simulation. During this preliminary stage, the method tests various template sizes to identify the point where facies probability estimates stabilize. This pre-established optimal template size is then used in subsequent simulations, ensuring both adequate pattern reproduction quality and efficient computation speed without needing to re-optimize for each simulation run.
3Reliability
If a conservative high template size is used to ensure reasonable pattern reproduction, then the geological feature accuracy is maintained, but the simulation computation time becomes excessively long
Solution Approach 1:
The patent applies dynamics by making the template size adaptive rather than fixed. The method dynamically adjusts the template size based on the specific characteristics of the training image and the local geological complexity. For simple geological features, smaller templates are used, while for complex features, larger templates are employed. This dynamic adjustment ensures reliable pattern reproduction only where necessary, significantly reducing overall computation time compared to using a uniformly conservative large template size throughout the entire simulation.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer system and a computer-implemented method for optimizing the number of conditioning data used in multiple point statistics simulation. The method includes inputting a training image representative of subsurface geological heterogeneity; and inputting an initial conservative number of conditioning data. The method further includes selecting a geometry of a template wherein a size of the template is defined by the conservative number of conditioning data; building a search tree using the template by scanning the training image with the template and storing data patterns present in the training image in the search tree to obtain a plurality of patterns; and determining a threshold number of conditioning data smaller than the initial conservative number of conditioning data beyond which estimated facies probabilities are not significantly modified by additional number of conditioning data.