CFD Super-Resolution Prediction With Uncertainty-Aware Mesh Upscaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional Computational Fluid Dynamics (CFD) simulations at high mesh resolutions are computationally intensive and time-consuming, and existing super-resolution techniques fail to accurately reflect real-world scenarios due to the lack of high-resolution mesh data and limitations in training data generation, leading to noise and artifacts in predictions.
Innovation Solution
A method and system for high-resolution simulation prediction in CFD that utilizes a rule-engine technique to select primary and secondary features, identifies a network architecture, and performs uncertainty analysis using aleatoric and epistemic uncertainty analysis to generate a trained model that predicts fine mesh data from coarse mesh data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high mesh resolution simulations are conducted to improve prediction precision, then measurement precision is improved, but computational intensity and time consumption increase significantly
Solution Approach 1:
The patent uses machine learning models to create a computational shortcut that copies the essential physics relationships from high-resolution simulations to predict fine-mesh results from coarse-mesh inputs, avoiding the need to perform computationally intensive high-resolution simulations while maintaining prediction accuracy
Solution Approach 2:
The patent replaces the mechanical computation of high-resolution CFD simulations with a machine learning-based prediction system that uses neural networks to map coarse-mesh inputs to fine-mesh outputs, substituting complex numerical computations with data-driven inference
2Measurement precision
If conventional super-resolution techniques are applied to reconstruct high-resolution data from low-resolution data, then measurement precision is improved, but the predictions contain noise and artifacts due to insufficient training data
Solution Approach 1:
The patent changes the approach from traditional image-based super-resolution to a physics-aware approach where the machine learning model is trained on physically consistent relationships between coarse and fine mesh data, using physics-informed loss functions and carefully constructed training datasets that respect the underlying fluid dynamics physics
Solution Approach 2:
The patent introduces physics-informed constraints and domain knowledge as intermediaries between the machine learning model and the CFD simulations, using physics-based loss functions and validation mechanisms to ensure that the trained model produces physically consistent predictions without noise or artifacts
Data Source
AI summary
This disclosure relates generally to high resolution simulation prediction for Computational Fluid Dynamics (CFD). CFD plays a crucial role in comprehending intricate physical phenomena spanning across scientific and engineering domains, hence it is essential to conduct simulations at high mesh resolutions for the governing equation of fluid flow. The current state-of-the-art super-resolution techniques involve reconstructing high-resolution data from down sampled low-resolution is limited to single scenario and does not accurately reflect real-world scenarios. The disclosed techniques enable prediction of fine-resolution data from low-resolution inputs from a variety of real-world CFD scenarios. Further the disclosed technique also identifies the most relevant network architecture for any CFD scenario and enabling accurate prediction of high-resolution data from low-resolution inputs by training the network architecture. Furthermore, the disclosure ensures also the robustness of the disclosed system through uncertainty analysis, encompassing both aleatoric and epistemic uncertainty analyses.