CAD Component Selection via HDBSCAN Shape Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing CAD systems struggle to efficiently select similar components during the design process, relying on user recognition, advance preparation, and limited searches within the CAD application.
Innovation Solution
A computer-implemented method using hierarchical clustering algorithms, such as HDBSCAN, to cluster components based on their shape elements, allowing users to select similar components on-the-fly by optimizing hyperparameter values for clustering validity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If users rely on visual recognition, naming standards, and categorizations to select similar components, then the selection process can be simple for limited designs, but it becomes increasingly difficult as designs become larger and more complex
Solution Approach 1:
The patent replaces manual visual recognition and categorization methods with an automated machine learning system. The clustering algorithm automatically groups components based on shape similarity, substituting the mechanical process of manual inspection with an automated computational approach that scales to complex designs.
Solution Approach 2:
The system enables components to serve themselves by automatically organizing them into clusters based on their inherent shape characteristics. The components' geometric properties are used by the algorithm to self-organize the library, eliminating the need for manual categorization and allowing users to simply query the organized structure.
2Productivity
If users create advance sets of components based on attributes, then selection is streamlined, but the ability to predict and offer alternative similar products during the design process is removed
Solution Approach 1:
The patent implements a dynamic clustering system where the component organization is not fixed in advance but can adapt during the design process. The machine learning model can be retrained or recalculated as new components are added or design requirements change, allowing the system to dynamically offer alternative similar products based on current design context.
Solution Approach 2:
The system allows for parameter changes in the clustering algorithm, such as adjusting similarity thresholds or retraining with new data, to adapt to different design scenarios. This enables the system to maintain high productivity while also being versatile enough to handle various design requirements and offer alternative components.
3Adaptability or versatility
If users manually search through large libraries of components, then comprehensive selection is possible, but significant time is required for the selection process
Solution Approach 1:
The patent applies preliminary action by pre-processing and organizing the entire component library into clusters based on shape similarity before the user needs to make selections. This upfront organization using machine learning algorithms allows users to quickly navigate to relevant clusters and make selections without manually searching through the entire library, significantly reducing selection time while maintaining comprehensive coverage.
Data Source
AI summary
A computer-implemented method of enabling a user to select at least one component from a group including identical and/or non-identical components forming part of a computer-aided design (CAD) model is provided. The method includes: a) receiving a seed component selection from a user via a user input device, where the seed component represents component criteria desired by the user; and b) based on the seed component, generating a selection including at least one component sharing common shape elements with the seed component. The HDBSCAN algorithm is used to cluster together components within a CAD application. The clustered components are then displayed to a user based on the seed component. This enables a user to select similar components quickly and simply.