Active Region Adaptation for Topology Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Topology optimization computations are resource-intensive and time-consuming, especially for complex design spaces with millions of design elements, making it impractical to perform hundreds or thousands of iterations efficiently.
Innovation Solution
The implementation of active region adaptations, where specific design elements within a design domain are identified and optimized, focusing resources on high-potential regions while keeping others constant, reducing computational time and resources needed for simulations and finite element analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If topology optimization is performed on the entire design domain with millions of design elements, then comprehensive optimization coverage is achieved, but computational time and resource consumption become excessively high
Solution Approach 1:
The design domain is segmented into multiple subdomains or regions of interest. Instead of optimizing all millions of design elements simultaneously, the method divides the large design space into smaller manageable regions, allowing selective optimization of specific areas while maintaining overall design integrity.
Solution Approach 2:
Different regions of the design domain are treated with different optimization strategies. High-priority regions with greater design potential receive intensive optimization, while low-priority regions use simplified or coarser optimization approaches, allocating computational resources according to local importance.
2Measurement precision
If the number of iterations in topology optimization is increased to improve solution accuracy, then optimization quality improves, but computational resource consumption increases
Solution Approach 1:
Instead of performing full optimization iterations across the entire design domain, the method applies partial optimization actions only to selected regions or design elements. This reduces the total number of computational operations while maintaining sufficient accuracy for critical areas.
Solution Approach 2:
The method performs preliminary analysis to identify regions with high design potential or sensitivity before initiating full optimization iterations. This preliminary screening allows the optimization process to focus computational efforts on areas where additional iterations will yield the most benefit, avoiding wasteful computations in regions where design variables are already near optimal.
3Adaptability or versatility
If all design elements are optimized simultaneously, then complete design space exploration is achieved, but computational complexity becomes unmanageable
Solution Approach 1:
The design space is segmented into multiple independent or semi-independent optimization problems. By dividing the large-scale optimization into smaller sub-problems, the computational complexity of each individual problem is reduced, making them more tractable while collectively covering the entire design space.
Solution Approach 2:
The optimization process dynamically adjusts which design elements are active in each iteration based on their potential for improvement. Regions showing significant design potential are activated for optimization, while regions with limited potential are temporarily deactivated or frozen, reducing the active problem size dynamically throughout the optimization process.
Data Source
AI summary
A computing system may include an initial design space engine and an active region adaptation engine. The initial design space engine may be configured to identify a design domain for which to optimize a topology based on an objective function and determine an active region. The active region adaptation engine may be configured to iteratively adapt the active region until an optimization ending criterion is satisfied. Iterative adaptation of the active region may include expanding the design domain to include branch design elements, performing finite element analysis (FEA) on the expanded design domain, and determining an adapted active region by activating some of the branch design elements based on an active sensitivity threshold and deactivating some of the active design element based on design variable value changes.