Shelled Generative CAD Models for Faster Crash Simulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing generative design methods for crash simulations result in unwieldy solid state designs that are computationally expensive and heavy, lacking an efficient mechanism for producing lighter, better-performing designs suitable for CAE crash simulations.
Innovation Solution
A lightweight shelled generative design with internal support structures is developed, which can be used in additive build simulations and vehicle crash simulations without significantly impacting simulation run time or mass scaling, utilizing a workflow that includes creating a shelled generative design with internal support structures and associative CAE crash simulation models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Strength
If a solid state generative design is used for crash simulations, then the design provides structural strength and完整性, but the design becomes heavy and computationally expensive with approximately 1,500,000 mesh elements
Solution Approach 1:
The patent applies shell elements to represent the generative design geometry instead of solid elements. This transforms the modeling approach from volumetric solid state design to surface-based shell design, significantly reducing the number of mesh elements from approximately 1,500,000 to a manageable size while maintaining structural accuracy and computational efficiency in crash simulations
Solution Approach 2:
The patent creates an associative crash simulation model that automatically updates when the generative design geometry changes. This copying mechanism ensures the simulation model remains synchronized with the CAD model without requiring manual remeshing, maintaining accuracy while reducing computational overhead
2Reliability
If a solid state generative design is used for crash simulations, then the design provides structural完整性, but the simulation processing time becomes excessively long
Solution Approach 1:
By using shell elements instead of solid elements, the patent reduces the computational complexity of the crash simulation model. This approach maintains the accuracy needed for reliable crash simulation results while dramatically reducing processing time by eliminating the need to model volumetric interiors with millions of elements
Solution Approach 2:
The patent substitutes the traditional solid mechanics-based finite element model with a shell-based model. This substitution replaces the computationally intensive solid element formulation with a more efficient shell element formulation that achieves comparable accuracy for crash simulation purposes with significantly reduced processing time
3Productivity
If a shelled generative design is used instead of solid state design, then weight is reduced and processing time decreases, but the design may lack structural完整性
Solution Approach 1:
The patent demonstrates that shell elements can accurately represent the structural behavior of generative designs for crash simulation purposes. The shell model captures the essential structural integrity and load paths needed for reliable crash analysis while avoiding the computational burden of solid element models, thus maintaining both accuracy and efficiency
Data Source
AI summary
A method and system provide the ability to generate models. A generative shelled base is created as a hollow computer-aided design (CAD) design. A t-spline mid-surface shell is created from the generative shelled base, which is then used to create a shell mesh model. A t-spline solid body is created from the generative shelled base, which is used to create an internal support structure that is converted into a shell CAD geometry, which is used to create a support structure mid-surface shell. The support structure mid-surface shell is combined with the shell mesh model into a generative mid-surface mesh that is used in a computer-aided engineering (CAE) crash simulation. The generated shelled base is combined with the shell CAD geometry into a generative shelled solid that is utilized in an additive build simulation.


