Custom Cellular Lattice Kernel Design via Genome Engine
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for designing cellular lattice materials focus primarily on geometric optimizations rather than topology, limiting the ability to customize lattice kernels according to specific material properties.
Innovation Solution
A computer-aided design system that uses a genome engine to generate custom cellular lattice kernels through an iterative process involving a prediction engine and fitness evaluator, allowing for the optimization of both geometry and topology based on targeted material properties, utilizing a machine learning-based model to approximate lattice structure properties and evaluate kernel candidates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If pre-defined lattice kernel library is used, then design process is simplified, but customization capability according to material properties is limited
Solution Approach 1:
The lattice kernel design is segmented into fundamental features (nodes, beams, geometric orientations, topological configurations) that can be independently controlled through the genome engine, allowing systematic generation of customized kernels while maintaining a structured design process
Solution Approach 2:
The system changes parameters by using a genome represented as binary code where each bit controls a specific feature, enabling systematic variation of geometric and topological parameters to achieve customized material properties while following a standardized design framework
2Measurement precision
If full-scale lattice simulations are performed for each kernel candidate, then prediction accuracy is improved, but computational time and process efficiency are reduced
Solution Approach 1:
The machine learning prediction model is trained in advance on lattice structure properties, enabling rapid prediction of kernel candidate performance without performing full-scale simulations for each candidate, thus saving computational time while maintaining accuracy
Solution Approach 2:
The machine learning prediction model acts as an intermediary between the genome engine and fitness evaluator, providing accurate predictions of lattice properties without requiring direct full-scale simulations, thereby reducing computational cost and time
3Adaptability or versatility
If topology optimization is performed, then material property customization is improved, but design process complexity is increased
Solution Approach 1:
The topology optimization process is segmented into discrete genome features that can be independently optimized, transforming the complex continuous topology optimization problem into a manageable discrete optimization problem with clear control over each feature
Solution Approach 2:
The fitness evaluator provides feedback on kernel candidate performance based on predicted material properties, enabling iterative optimization of the genome sequence to achieve customized material properties through systematic evaluation and selection
Data Source
AI summary
Methods and systems are disclosed for generation of cellular lattice kernels optimized by multiple objectives for highly specific targeted properties of geometry and topology rather than state of the art methods that rely on a predefined kernel library. Using a characterization of virtual kernel features, bulk material properties can be predicted using approximations from the virtual kernel rather than having to rely solely on experimental finite element simulations of lattice structures.