Automated Classification of Numerical Simulation Results
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual examination of numerous graphical simulation results, such as crashworthiness assessments, is tedious and error-prone, necessitating improved methods for characterizing and classifying these results.
Innovation Solution
A training database is created with graphical representations from multiple simulations, each associated with textual descriptions, using an autocorrelation technique to calculate a quality index, allowing new simulations to be characterized or classified with textual descriptions and confidence scores, and enabling database updates based on predefined criteria.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual examination of graphical simulation results is performed, then detailed analysis can be conducted, but the process becomes tedious and error-prone
Solution Approach 1:
The system enables self-service by allowing the database to automatically update itself through unsupervised learning. The application module continuously learns from new graphical representations and textual descriptions without requiring manual intervention, automatically adding learned patterns to the training database and improving its own classification capabilities over time.
Solution Approach 2:
The patent replaces the manual mechanical examination process with an automated computer-based system. The application module uses machine learning algorithms to automatically compare new graphical representations against the training database, eliminating the need for manual visual inspection while maintaining or improving analysis accuracy.
2Reliability
If every graphical result is manually examined, then comprehensive assessment is achieved, but the process is tedious and error-prone
Solution Approach 1:
The system performs self-improvement through continuous learning. As new simulation results are processed, the application module automatically updates the training database with newly learned patterns and characteristics, enabling the system to become progressively more reliable without additional manual effort.
Solution Approach 2:
The training database acts as an intermediary between the raw simulation data and the final assessment. It stores learned patterns and characteristics that mediate the comparison process, allowing the system to reliably assess new results by matching them against established patterns rather than requiring direct manual analysis.
3Measurement precision
If a large training database is created with multiple graphical representations, then classification accuracy improves, but database maintenance becomes complex
Solution Approach 1:
The application module automatically manages the training database through unsupervised learning. It independently identifies new patterns, determines when to add new graphical representations, and updates the database without requiring manual cur curation or intervention, thereby simplifying database maintenance while improving accuracy.
Solution Approach 2:
The system dynamically adjusts the composition and characteristics of the training database based on learned parameters. As the application module processes new data, it modifies the database structure and content parameters automatically, optimizing the database for improved classification accuracy while managing complexity through adaptive parameter changes.
Data Source
AI summary
Methods of characterizing or classifying graphical representation of numerical simulation results are disclosed. A training database is created in a computer system by including a plurality of graphical representations of respective results obtained from a plurality of numerical simulations. Each graphical representation is associated with a textual description of a pertinent feature related to the numerical simulations by user. A quality index with respect to the associated textual description is calculated for each graphical representation by application module using an autocorrelation technique of correlating all graphical representations with one another in the training database. A new graphical representation obtained from another numerical simulation can then be characterized with one of the textual descriptions and a corresponding confidence score by comparing the new graphical representation with all graphical representations in the training database. The training database may be improved by adding or removing appropriate graphical representations in accordance with predefined criteria.