Grid-Free Turbulence Simulation Using Vortex Filaments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computational fluid dynamics (CFD) simulation systems face inefficiencies and inaccuracies in modeling high Reynolds number turbulence, particularly in complex geometries and scenarios involving moving boundaries, synthetic jets, and exascale computing, due to limitations in vortex methods and grid-based simulations.
Innovation Solution
The implementation of a CFD simulation system that utilizes vortex sheets and tubes, combined with the Fast Multipole Method (FMM) and exascale computing, to efficiently model fluid flow and turbulence, reducing computational complexity and improving accuracy by using a grid-free approach and adaptive meshing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If grid-based simulations (RANS/LES) are used to model fluid flow, then the fluid flow domain can be covered with a mesh, but the simulation efficiency and accuracy deteriorate due to failure to address real physics in high Reynolds number turbulence
Solution Approach 1:
The patent replaces the traditional grid-based mechanical mesh system with a vortex-based computational approach. Instead of using fixed or moving grids to discretize the flow domain, the invention uses vortex elements (vortex particles, vortex rings, vortex filaments) to represent and track turbulent structures directly, eliminating the need for mesh generation and associated computational overhead while capturing true turbulent physics
Solution Approach 2:
The invention changes the fundamental parameter representation from grid coordinates and velocity fields to vortex strength, position, and circulation parameters. This parameter transformation allows direct modeling of high Reynolds number turbulence without the diffusive effects inherent in grid-based methods, improving both accuracy and computational efficiency
2Measurement precision
If vortex methods are used to track individual vortical structures, then high Reynolds number turbulence effects can be represented with high numerical accuracy, but the number of vortices grows to unmanageable levels
Solution Approach 1:
The patent extracts and removes vortex elements that have become negligible or redundant in the flow field. By continuously identifying and eliminating vortices with minimal impact on the overall flow dynamics, the system maintains numerical accuracy while preventing uncontrolled growth in the number of tracked vortex elements
Solution Approach 2:
The invention implements a vortex lifecycle management system where vortices are discarded when they merge, dissipate, or become insignificant, and new vortices are generated only when physically necessary. This dynamic creation and destruction process maintains computational tractability while preserving essential turbulent structures
3Reliability
If Biot-Savart law evaluation is used to compute velocities from vortex elements, then the vortex method formulation can be maintained, but the computational time increases beyond reasonable levels
Solution Approach 1:
The patent segments the computational domain and vortex elements into groups or clusters, allowing the Biot-Savart law to be applied efficiently to subsets of vortices. This segmentation enables parallel computation and reduces the computational burden of evaluating velocity fields from all vortex elements at each time step
Solution Approach 2:
The invention introduces intermediary data structures and computational representations that mediate between the vortex elements and the Biot-Savart law evaluation. These intermediaries pre-compute or approximate velocity contributions, reducing the computational complexity from O(Nv²) to a more manageable level while maintaining formulation fidelity
4Ease of operation
If random walk diffusion model is used for traditional random vortices, then the vortex dynamics can be simulated, but noise is introduced that exceeds levels found in real turbulent flows
Solution Approach 1:
The patent changes the diffusion parameterization from random walk models to physically-based diffusion coefficients that match real turbulent flow characteristics. By using diffusion parameters derived from actual turbulence statistics rather than stochastic random walk assumptions, the invention eliminates excessive numerical noise while maintaining computational simplicity
Data Source
AI summary
A system and method for simulation of fluid flow. The system being configured to remove loops in a vortex filament in a simulation model and reconnect the filament. The system may also be configured to model fluid flow in relation to a moving object and to correct errors in surface vorticity.