Particle Discretization for High-Fidelity Simulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Complex computer simulations for physical phenomena, such as object impacts, require high accuracy but often take significant time to run, and existing fast simulations compromise on fidelity and reliability.
Innovation Solution
A computer-implemented method that discretizes objects into a reduced number of particles using a subdivision algorithm and smooth particle hydrodynamics (SPH) to generate data for simulations, allowing for faster execution while maintaining high fidelity by optimizing particle distribution and interaction analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complex computer simulations are used to achieve high accuracy, then simulation fidelity is improved, but simulation run time increases significantly
Solution Approach 1:
The patent segments objects into discrete particles with specific properties (mass, density, velocity, pressure, stress, energy) rather than using continuous complex models. This particle-based segmentation allows the simulation to maintain physical accuracy while reducing computational complexity and execution time.
Solution Approach 2:
The patent changes the fundamental parameters of object representation from continuous geometric models to discrete particle properties. By representing objects as collections of particles with specific physical parameters, the simulation achieves high fidelity while running significantly faster than traditional complex simulations.
2Productivity
If fast computer simulations are used to reduce run time, then simulation speed is improved, but fidelity and reliability decrease
Solution Approach 1:
The patent creates a simplified copy of physical objects using particles that replicate key physical properties (mass, density, velocity, pressure, stress, energy). This particle-based copying maintains the essential physics and reliability of the simulation while enabling faster execution speeds.
Solution Approach 2:
The patent changes the representation parameters from complex continuous geometries to discrete particle properties. This parameter transformation allows the simulation to maintain physical reliability through conserved quantities while achieving superior simulation speed.
3Measurement precision
If more particles are used to represent objects, then simulation accuracy is improved, but data quantity and computational complexity increase
Solution Approach 1:
The patent optimizes the particle representation by assigning multiple physical parameters to each particle (mass, density, velocity, pressure, stress, energy). This allows high simulation accuracy to be achieved with fewer particles, as each particle carries comprehensive physical information rather than relying on large numbers of simple geometric elements.
Data Source
AI summary
A computer-implemented method is provided for generating data representing an object involved in a computer-implemented simulation of a physical experiment. Data describing geometric dimensions and material for the object is received as input. The data may be broken down into data representing distinct components of the object. Data is generated that describes a base parent volume region that extends at least beyond extents of the geometric dimensions of a portion of the object. Next, data is generated that describes child volume regions produced by subdividing the base parent volume region into the child volume regions each having dimensions that are a fraction of the dimensions of the base parent volume region. The data describing each child volume region is examined to determine whether vertices of each of the child volume regions are in the interior or exterior of the portion. The subdivision and examining process is repeated for each child volume region that is not entirely in the interior or entirely in the exterior of the portion to further subdivide each such child volume region to produce data describing a set of child volume regions that do not satisfy the criteria for further subdivision. Data is generated that uniformly distributes particles at positions of the portion of the object based on dimensions of the child volume region in the set, wherein the data for each particle describes a mass density, velocity, pressure, stress and energy at a position and a collection of the particles represent the object portion for use in the computer-implemented simulation.