Hierarchical Multi-Scale Part Design for Manageable Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current design tools struggle to efficiently analyze and optimize complex structures across multiple length scales due to computational limitations and the need for detailed multi-scale analysis, often leading to impractical and numerically unstable solutions, especially in additive manufacturing where high-resolution multi-material designs require consideration of nano-scale to macro-scale phenomena.
Innovation Solution
A hierarchical multi-scale design method that uses a top-down approach with a declarative representation of shape and material distributions, allowing for the optimization of surrogate properties at each scale while enforcing scale-agnostic invariance, enabling the synthesis of complex designs that span multiple length scales without relying on restrictive assumptions about material properties or structural separability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If detailed multi-scale analysis is performed across all length scales, then design precision and physical accuracy are improved, but computational complexity and analysis time increase exponentially
Solution Approach 1:
The design space is segmented into multiple hierarchical levels, each representing a different length scale. The analysis is performed separately at each level, starting from macro-scale and progressively refining to micro-scale only where needed. This segmentation allows detailed analysis to be focused on critical regions while using simplified models for less critical areas, thereby reducing overall computational complexity while maintaining design precision.
Solution Approach 2:
The patent introduces a hierarchical level dimension to the traditional single-scale analysis approach. By adding this temporal/spatial dimension, the analysis transitions from analyzing all details at one scale simultaneously to analyzing different scales sequentially at different hierarchical levels. This dimensional change enables manageable computational steps that collectively achieve comprehensive multi-scale design precision.
2Manufacturing precision
If high-resolution designs are generated at the beginning, then design accuracy is improved, but manufacturing feasibility and numerical stability deteriorate due to connectivity issues
Solution Approach 1:
The patent performs preliminary coarse-scale design and analysis at higher hierarchical levels before finalizing detailed micro-scale geometries. This preliminary action establishes the overall structural framework and identifies critical regions that require detailed analysis. By delaying the generation of high-resolution designs until necessary, the method avoids premature commitment to detailed geometries that may create manufacturing connectivity issues, while still achieving high design accuracy in critical areas.
Solution Approach 2:
High-resolution design details are applied locally only to critical regions identified through coarse-scale analysis, rather than uniformly across the entire structure. This local quality approach maintains manufacturing feasibility by avoiding excessive detail in non-critical areas, while achieving high design accuracy where it matters most. The hierarchical method naturally identifies which regions require local high-resolution treatment based on structural importance and loading conditions.
3Reliability
If comprehensive multi-scale optimization is performed, then design performance is improved, but analysis time and computational resources increase significantly
Solution Approach 1:
The optimization process is segmented into multiple hierarchical stages, each optimizing design variables at a specific length scale. Coarse-scale optimization is performed first to establish the overall structural configuration, followed by progressive refinement at finer scales. This segmentation allows the use of appropriate optimization algorithms and mesh resolutions for each scale, significantly reducing total analysis time while maintaining comprehensive design performance through cumulative optimization effects.
Solution Approach 2:
The patent applies partial optimization action by focusing detailed optimization efforts only on critical regions and scales that significantly impact overall design performance. Rather than performing exhaustive optimization across all scales and regions uniformly, the hierarchical method identifies and concentrates computational resources on the most influential design variables, achieving high design performance with reduced analysis time through strategic partial action.
Data Source
AI summary
The present disclosure is directed to a method and system for hierarchical multi-scale design with the aid of a digital computer. A hierarchical representation of a shape and material distribution is constructed which satisfies a top-level constraint at a top-level of representation. Properties for families of designs at each of the lower levels of representation that satisfy additional constraints link each of the lower levels of representation to at least a next higher level of the representation.