Double-Sided Surfel Interface for Multi-Resolution Lattice Boltzmann
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for connecting mesh regions with different resolutions in fluid flow simulations face challenges such as spatial uncertainty, misplacement of particles, and artificial wave reflections, particularly at scales near the lattice spacing, which affect the accuracy of fluid dynamic properties and their spatial propagation.
Innovation Solution
A surface-based approach using double-sided surfels (facets) is employed to connect mesh regions with different resolutions, where each surfel interfaces with one resolution on each side, allowing precise spatial location definitions and accurate propagation of particle distributions across resolution boundaries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If volumetric overlap based VR methods are used to connect mesh regions, then computational cost is reduced and mass-momentum-energy conservation is achieved, but spatial uncertainty and particle misplacement occur leading to artificial wave reflections
Solution Approach 1:
The interface between coarse and fine regions is segmented into discrete surfel elements rather than using continuous volumetric overlap. Each surfel is a two-dimensional surface element that separates coarse and fine regions, allowing precise control over particle transfer at the interface without the spatial uncertainty inherent in volumetric methods.
Solution Approach 2:
The approach transitions from three-dimensional volumetric overlap to two-dimensional surface-based connection. By using surfels as interfaces, the method eliminates the third dimension of uncertainty in volumetric overlap while maintaining computational efficiency through surface-level particle transfer operations.
2Ease of manufacture
If volumetric overlap is used to connect regions with different resolutions, then the geometric weight construction is simple, but it creates misplacement of particles in the stream-wise flow distribution
Solution Approach 1:
Surfels act as intermediary elements between coarse and fine regions, mediating the particle transfer process. Instead of direct volumetric overlap causing misplacement, the surfel interface provides a controlled intermediate zone that preserves particle distribution accuracy while maintaining simple geometric weight construction through explicit surface-based transfer rules.
3Ease of operation
If overlapping volumetric domain is used, then the connection between regions is straightforward, but it creates artificial wave reflections that affect fluid dynamic properties near lattice spacing
Solution Approach 1:
The harmful volumetric overlap region is extracted and replaced with a surface-based surfel interface. This removes the source of artificial wave reflections while maintaining the simplicity of region connection. The surfel approach preserves fluid dynamic properties by providing precise particle transfer control at the interface.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This method improves the accuracy of fluid flow simulations by reducing computational complexity, conserving mass and momentum, and enhancing the precision of fluid dynamic properties while maintaining exact flux definitions between regions with varying resolutions.
Implementation Method 1
Particle distributions are advected from neighboring voxels to the surface elements. Surface dynamics occur on the surface elements and particle distributions are propagated from the first side of the surface element to the second side of the surface element and vice versa.
Data Source
AI summary
Systems and methods include simulating, in a digital representation of a simulation space, a fluid flow across a boundary including double-sided facets abutting a first region and a second region by: determining first particle distributions for first facets of the double-sided facets based on particle distributions of first resolution voxels in the first region and second particle distributions for second facets of the double-sided facets based on particles of the second resolution voxels in the second region; performing surface interactions on the double-sided facets; combining particle distributions from the second facets to the first facets based on the surface interactions; determining particle distributions to be advected from the first facets to the second facets based on the surface interactions; and advecting particle distributions from the first facets to the first resolution voxels and from the second facets to the second resolution voxels.


