Feature Recognition Engine for CAD Design Automation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current CAD tools lack effective integration of artificial intelligence in the post-processing phase of finite element analysis and design modification, particularly in structural analysis, limiting the ability to extract meaningful qualitative design information for design improvements.
Innovation Solution
A rule-based suggestion engine that determines physical characteristics of simulated objects, compares them to stored characteristics, identifies matching objects, and generates lists of objects meeting desired physical properties, enabling intelligent design modifications and parallel processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If AI methods are applied in FEA post-processing phase, then design improvement capability is enhanced, but system complexity increases
Solution Approach 1:
The system segments the design improvement process into distinct modules: FEA post-processing module that extracts qualitative design information, a knowledge base module that stores design rules and constraints, and a design modification module that generates improved designs. This segmentation allows AI methods to be applied systematically without overwhelming system complexity.
Solution Approach 2:
The patent introduces an intermediary knowledge base that acts as a mediator between the FEA analysis results and the design modification system. This knowledge base stores qualitative design information and engineering rules, enabling the system to make intelligent design decisions without requiring complex direct AI algorithms, thus reducing overall system complexity while maintaining automation capability.
2Adaptability or versatility
If qualitative description of engineering analysis results is used, then system generality is improved, but information precision is reduced
Solution Approach 1:
The system transforms quantitative FEA results into qualitative parameters that describe design characteristics (e.g., stress concentration patterns, deformation modes). This parameter transformation allows the system to maintain generality across different applications while preserving essential design information needed for modification decisions.
Solution Approach 2:
The patent creates a qualitative copy or representation of the quantitative analysis results. Instead of directly using complex numerical data, the system generates simplified qualitative descriptions that capture the essential design features. This copying approach enables broader system applicability while retaining sufficient information for design improvement.
3Productivity
If feature recognition is used to identify similar parts, then design reuse is improved, but processing time increases
Solution Approach 1:
The system performs preliminary feature recognition and part classification during the design creation phase, storing extracted features and similarity metrics in advance. When design reuse is needed, the system queries this pre-processed information rather than performing complete feature recognition, significantly reducing processing time while maintaining effective design reuse capability.
Data Source
AI summary
A system or method that includes determining a plurality of physical characteristics of a first simulated object. The system or method includes comparing the plurality of physical characteristics of the first simulated object to a plurality of characteristics of a plurality of objects stored on a storage medium. The system or method includes identifying at least one matching object from the plurality of objects stored on the storage medium. The system or method includes comparing at least one physical property of the at least one matching object to at lease one desired physical property of the first simulated object and generating a list of matching objects that meet the at least one desired physical property.