Unstructured Volumetric Grid Simplification via Sub-Volume Clustering

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The oil & gas industry faces challenges in generating visualizations for complex unstructured volumetric grids due to high computing overhead, especially for three-dimensional and interactive representations, which are exacerbated by the increasing complexity of reservoir models and the use of unstructured grids.

Innovation Solution

A method that clusters cells in an unstructured volumetric grid, simplifies the boundary of each cluster, and generates an output grid by merging adjacent cells based on a clustering criterion, discarding interior nodes and faces, and collapsing low connectivity vertices, thereby reducing computational demands for visualization and numerical simulations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If unstructured volumetric grids are used to model complex subsurface features, then modeling accuracy is improved, but computing overhead for visualization increases

Engineering Contradiction:

Improvemodeling accuracyVSAvoidcomputing overhead

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent divides the unstructured volumetric grid into multiple sub-volumes or clusters, each representing a portion of the reservoir. This segmentation allows the system to process and visualize only relevant portions at high detail while using coarser representations for other areas, thereby maintaining modeling accuracy where needed while reducing overall computing overhead for visualization.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies different levels of detail and processing quality to different regions of the grid based on their importance and the user's view. Foreground regions receive high-quality processing while background regions use simplified representations, optimizing the balance between visualization performance and modeling fidelity in local areas.

Inventive Principle:

Principle #3Local quality

2Measurement precision

If the number of cells in the reservoir model increases to capture more details, then model complexity and accuracy are improved, but visualization performance deteriorates

Engineering Contradiction:

Improvemodel detailVSAvoidvisualization performance

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent implements dynamic level-of-detail adjustment that adapts to the user's current viewpoint and interaction state. As users zoom in or out, or navigate to different regions, the system dynamically adjusts the number and detail of cells rendered, maintaining high model detail where needed while optimizing visualization performance by reducing cell count in less critical areas.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent applies full-detail processing only to the portions of the model currently needed for visualization, rather than processing the entire high-detail model. This partial action approach maintains model detail for visible regions while avoiding the computational cost of processing and rendering all cells at maximum detail.

Inventive Principle:

Principle #16Partial or excessive action

3Adaptability or versatility

If interactive real-time modifications are enabled, then user control and adaptability are improved, but computing overhead increases

Engineering Contradiction:

Improveinteractive controlVSAvoidcomputing overhead

Core Design Contradiction:

Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The patent performs preliminary processing and pre-computation of grid data, organizing it into efficient data structures and pre-calculating properties that will be needed for interactive operations. This preliminary action enables real-time modifications by having essential computations completed in advance, reducing the computing overhead during actual user interactions.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements dynamic optimization that adjusts processing intensity and detail levels based on the current interaction state. During interactive real-time modifications, the system dynamically balances between maintaining adaptability for user control and managing computing overhead by adjusting the level of processing applied to different regions and operations.

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS10303711B2Unstructured volumetric grid simplification using sub-volume clustering

Publication Date: 2019.05.28 SCHLUMBERGER TECH CORP

AI summary

A method, apparatus, and program product cluster a plurality of cells of an input unstructured volumetric grid representative of a subsurface volume into a plurality of clusters, simplify a boundary of each cluster and generate an output unstructured volumetric grid representing at least a portion of the input unstructured volumetric grid by generating in the output unstructured volumetric grid a respective cell for each of the plurality of clusters. The resulting output grid may be used to facilitate the generation of visualizations and/or numerical simulations.