Multi-instanced Physics Simulations via Reduced Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for simulating large-scale physics scenarios are inefficient due to the need to re-mesh, re-apply boundary conditions, and re-calculate solutions for each environmental change, leading to high computational costs and resource usage.
Innovation Solution
A computer-implemented method that creates a database of local simulation instances, computes reduced models for each instance, and reuses these models to simulate a global physics scenario, reducing the need for recalculation and minimizing computational resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional numerical simulation methods are used for each environmental change, then physical accuracy is maintained, but computational cost and simulation time increase exponentially
Solution Approach 1:
The global simulation domain is divided into multiple local simulation instances that can be independently computed and stored. When environmental changes occur, only the affected local instances need to be updated rather than re-computing the entire global simulation, thus maintaining physical accuracy while reducing simulation time.
Solution Approach 2:
Local simulation instances are pre-computed and stored in a database before global simulation is needed. These pre-computed local models can be quickly retrieved and assembled for global simulation, avoiding the need to re-calculate everything from scratch when environmental conditions change.
2Measurement precision
If traditional numerical simulation methods are used for each environmental change, then simulation accuracy is maintained, but computational resources and data storage requirements grow exponentially
Solution Approach 1:
The simulation data is segmented into discrete local instances that can be stored independently. This segmentation allows for efficient storage and retrieval of only the necessary local models rather than storing complete global simulation data for every possible environmental condition.
Solution Approach 2:
Pre-computed local simulation instances are stored as reusable models in a database. These local models act as copies that can be instantiated multiple times in different global simulation contexts without requiring additional computation, reducing both storage requirements and computational resources.
3Adaptability or versatility
If the scene disposition changes frequently, then adaptability is improved, but the need to re-mesh and re-calculate increases computational complexity
Solution Approach 1:
The system dynamically assembles global simulations from pre-computed local instances based on the current scene configuration. When scene disposition changes, the system adapts by selecting and combining different local instances rather than re-computing the entire simulation, maintaining adaptability while reducing computational complexity.
Solution Approach 2:
Local simulation instances are designed to be universal and reusable across multiple global simulation contexts. A single local instance can be instantiated multiple times in different positions and configurations within the global scene, reducing the need to create and store separate models for each scenario.
Data Source
AI summary
The disclosure notably relates to a computer-implemented method for simulating together a plurality of physics simulation instances included in a global physics simulation. The method includes creating a database of local simulation instances. The creating includes providing a set of local simulations. The set of local simulations includes at least two local simulations. A local simulation is a physics simulation that is part of the global physics simulation and that can be computed alone and independently of the multi-physics simulation. Each local simulation of the set of local simulations is already computed. The creating further includes, for each local simulation of the set of local simulations, computing a respective reduced model of the local simulation. The creating further includes, for each local simulation of the set of local simulations, storing in the database a respective local simulation instance. The respective local simulation instance includes the respective computed reduced model.