Automatic Variable Fidelity Simulation for Reducing Iteration Time
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The finite element method for simulating mechanical objects is inefficient due to the high number of iterations required to achieve an accurate behavior simulation model, leading to significant user time and computational resource expenditure.
Innovation Solution
The method involves automatically creating and solving simplified models from complex models, using techniques such as creation of rigid body models from flexible body models, coarse mesh models from fine mesh models, and linear connections from nonlinear connections, to reduce simulation time and identify modeling deficiencies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual creation of multiple models with increasing levels of complexity is used, then modeling accuracy is improved, but user time and computational resources are significantly increased
Solution Approach 1:
The system automatically performs preliminary actions by generating simplified models before the complex model to identify and correct modeling deficiencies early in the process. This preliminary validation step prevents time-consuming iterations later by catching errors before they propagate to the full-fidelity model.
Solution Approach 2:
The system creates simplified copies of the complex model at reduced fidelity levels. These copy models replicate the essential geometry and behavior of the original model but with fewer elements and simplified physics, enabling rapid validation without requiring the full computational resources of the complex model.
2Reliability
If multiple iterations of build-solve-review-fix cycles are performed, then valid simulation model is obtained, but the number of iterations and time expended are significant
Solution Approach 1:
Simplified models are automatically generated and solved before the complex model to perform preliminary validation. This early detection of modeling errors reduces the number of iterations needed in the complex model by addressing issues in advance.
Solution Approach 2:
The system implements feedback loops where results from simplified models inform and guide the validation of complex models. Error messages and validation results from simplified models provide feedback that helps identify and correct issues before they affect the complex model, reducing iterative cycles.
3Loss of time
If simplified models are used to validate complex models, then simulation time is reduced, but computational accuracy may be compromised
Solution Approach 1:
The validation process is segmented into multiple fidelity levels. Simplified models validate basic geometric correctness and boundary conditions, while complex models validate detailed physical behavior. Each segment focuses on specific aspects of model validity, distributing the accuracy requirements across different model types.
Solution Approach 2:
Different levels of model fidelity are applied to different aspects of validation. Simplified models provide local validation of geometry and constraints, while complex models provide local validation of detailed physics. This local quality approach ensures accuracy where needed while maintaining efficiency elsewhere.
Data Source
AI summary
A method, apparatus and computer readable medium for performing a computer simulation a physical object, includes receiving at least one selection by a user with respect to usage of a simpler model or a more complex model to be used to model at least one attribute of the physical object; performing a computer simulation of the physical object based on the at least one selection received from the user; and rerunning the computer simulation a plurality of times using results obtained from earlier run computer simulations, to obtain an accurate representation of the physical object.