Reduced-Model Simulation Visualization With Accuracy-Controlled Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for compressing and visualizing simulation data, such as those from physics simulations, often result in inaccurate representations due to loss of precision and lack of control over numerical accuracy, and require expensive supercomputers for post-processing, while the increasing computational power of high-performance computers is outpaced by data generation.
Innovation Solution
A computer-implemented method that computes a full simulation and a reduced model using Reduced Order Modeling techniques, representing each state as a linear combination of basis elements, allowing for controlled accuracy and data compression based on physics behavior, enabling efficient display on lower bandwidth networks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If compression strategies are applied to reduce data size for network transfer, then data transfer time is reduced, but numerical accuracy and precision are lost
Solution Approach 1:
The patent changes the fundamental parameter of data representation from raw simulation values to coefficients in a reduced basis expansion. This transformation allows the same physical information to be encoded with fewer parameters while maintaining numerical accuracy, as the reduced basis captures the essential physics of the system. The coefficients are computed to preserve the physical behavior rather than simply compressing the data.
Solution Approach 2:
The patent extracts the essential physical behavior from the full simulation data by identifying a reduced basis that spans the dominant modes of the system. Instead of transmitting all simulation data points, only the essential basis functions and their coefficients are transferred, separating the core physical information from redundant details while maintaining accuracy.
2Quantity of substance
If subsampling is used to reduce data size by storing only certain regions or variables, then data storage size is reduced, but physical accuracy is compromised
Solution Approach 1:
Instead of subsampling spatial or temporal data points, the patent transforms the data representation parameters entirely. The full simulation data is projected onto a reduced basis set, changing from a high-dimensional representation to a low-dimensional one that preserves physical accuracy. This allows storing and transferring only the essential coefficients while maintaining the ability to reconstruct physically accurate results.
Solution Approach 2:
The reduced basis acts as an intermediary between the full simulation data and the compressed representation. Rather than directly subsampling or compressing the raw data, the basis functions serve as a mediating mathematical structure that captures the essential physics, allowing accurate reconstruction from fewer parameters.
3Productivity
If video compression techniques are applied to simulation data, then data transfer efficiency is improved, but control over numerical accuracy is lost
Solution Approach 1:
The patent fundamentally changes the data representation from pixel-based or frame-based video formats to a physics-based reduced basis expansion. This parameter transformation allows the data to be compressed according to physical significance rather than visual compression algorithms, maintaining numerical accuracy while improving transfer efficiency.
Solution Approach 2:
The patent replaces video compression mechanisms with a physics-based mathematical transformation. Instead of using empirical compression algorithms designed for visual data, the system uses reduced basis methods from computational physics that inherently preserve numerical accuracy and physical consistency.
4Measurement precision
If full simulation data is transferred without compression, then numerical accuracy is maintained, but network bandwidth requirements increase
Solution Approach 1:
The patent transforms the data from a high-dimensional full simulation representation to a low-dimensional reduced basis expansion. This parameter reduction maintains numerical accuracy by preserving the essential physical behavior in fewer parameters, thereby reducing the data volume that needs to be transferred over the network.
Solution Approach 2:
The patent extracts and retains only the essential physical information from the full simulation data by projecting it onto a reduced basis set. This extraction process separates the critical numerical information from redundant data, maintaining accuracy while minimizing the data volume for network transfer.
Data Source
AI summary
The disclosure notably relates to a computer-implemented method for displaying a simulation. The method includes computing a full simulation. The full simulation includes states. The method further includes computing a reduced model of the computed full simulation. The reduced model includes a basis with elements. Each state of the full simulation is represented by a respective linear combination of basis elements. The method further includes displaying, for at least one state of the full simulation, a part of the respective linear combination. This constitutes an improved method for displaying a simulation.