Topology Optimization of Load-Bearing Trusses
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for topology optimization of load-bearing structures, such as voxel-based and Ground Structure methods, face limitations in computational efficiency and require extensive post-processing to convert solid geometry into discrete members, and struggle with creating localized areas of higher beam density.
Innovation Solution
A computer-executed method that combines topology and geometry optimization using a free-form arrangement of linear beams, allowing for iterative insertion and removal of nodes and elements, and adjustment of node positions based on finite element analysis, mimicking natural cell-based growth patterns to optimize structural models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If voxel-based topology optimization is used, then the optimization process is automated and can be executed on a computer, but the resulting structure is a solid volume that requires extensive post-processing to convert it into discrete linear beams
Solution Approach 1:
The design domain is divided into a regular grid of voxels that can be independently added or removed during optimization. The final solid volume is then segmented into discrete linear beam elements through systematic post-processing that identifies beam paths within the voxel structure.
Solution Approach 2:
The patent introduces an intermediate representation where the optimized solid volume serves as a bridge between the automated voxel-based optimization process and the final discrete beam structure. This intermediate form allows automated optimization while providing a clear pathway for conversion to constructable beam elements.
2Ease of manufacture
If Ground Structure method is used with a large amount of beams filling the design domain, then linear beams can be directly used without post-processing, but underutilized beams must be iteratively removed based on FEA which increases computational complexity
Solution Approach 1:
Instead of starting with a dense Ground Structure and removing beams, the patent extracts only the necessary beam elements from the optimized solid volume. The post-processing identifies and extracts critical beam paths while removing redundant elements, reducing the initial beam count and computational complexity.
Solution Approach 2:
The patent performs preliminary optimization using the voxel-based approach to define the optimal material distribution before converting to beams. This preliminary action establishes the structural layout, so that subsequent beam extraction only needs to follow the pre-determined optimal paths rather than evaluating all possible beam configurations.
3Extent of automation
If voxel-based topology optimization is used, then automated structural design can be achieved, but it is difficult to create local areas of higher beam density than the initial global density
Solution Approach 1:
The voxel-based optimization naturally produces local variations in material density, with some voxels being solid and others void. The post-processing converts this to local variations in beam density by concentrating beam elements in regions where the solid volume is densest, creating localized areas of higher beam density without requiring manual intervention.
Solution Approach 2:
The patent transitions from the 3D voxel grid to a 1D beam network representation. This dimensional change allows the structure to achieve higher local beam density along critical load paths while maintaining overall sparsity, as beams can be concentrated in specific regions without the constraints of the original voxel grid uniformity.
Data Source
AI summary
The present invention is for a computer-executed method for the generation or topology optimization of load-bearing trusses. The trusses consist of joints that are connected by linear structural elements, with at least one support and one load that the truss is to support. Based on a finite element analysis of the current state of the truss, the truss is iteratively improved by adjusting its topology, its geometry, and optionally the sizing of its members.