Component Model Validation Using ML Discrepancy Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Complex technical systems with multiple intricately linked components are difficult to validate or verify due to their stochastic and complex nature, making it hard to predict and guarantee desired behavior, especially in systems of systems where complexity grows rapidly.
Innovation Solution
A method involving obtaining models for system components, training machine learning models to predict outputs, determining discrepancies between model and machine learning model outputs, and verifying system behavior by maximizing the probability of meeting desired criteria under constrained discrepancies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional validation methods are used on complex technical systems, then validation accuracy may be maintained, but the complexity of the validation process and data requirements increase significantly
Solution Approach 1:
The patent segments the complex technical system into individual components, obtaining models for each component separately. This allows validation to be performed at the component level rather than requiring full-system validation, thereby reducing the overall complexity of the validation process while maintaining accuracy through component-level precision.
Solution Approach 2:
The patent creates virtual copies (digital twins) of system components through machine learning models that replicate component behavior. These models serve as substitutes for physical components during validation, eliminating the need for extensive real-world data collection and reducing validation complexity while preserving validation accuracy.
2Reliability
If extensive real-world data collection is performed for validation, then validation completeness improves, but the time and resources required increase significantly
Solution Approach 1:
The patent performs preliminary actions by training machine learning models on available data before actual validation is needed. These pre-trained models capture component behaviors and relationships in advance, so that during validation, the models can be queried directly without requiring additional real-world data collection, thus improving validation completeness while reducing validation time.
Solution Approach 2:
The patent replaces the mechanical process of collecting real-world data with a computational approach using machine learning models. Instead of physically gathering data from the system during validation, the trained models generate predictions and behaviors computationally, significantly reducing validation time while maintaining reliability through the models' ability to replicate real-system behavior.
3Quantity of substance
If machine learning models are used to predict component outputs, then data requirements are reduced, but model training complexity increases
Solution Approach 1:
The patent applies segmentation by training machine learning models for individual components rather than attempting to model the entire complex system at once. This divides the training task into smaller, more manageable pieces, reducing the data requirements for each model while the overall system maintains comprehensive coverage through the collection of component-level models.
Data Source
AI summary
A method for verifying and/or validating whether a technical system fulfills a desired criterion. The technical system emits output signals based on input signals supplied to the technical system. The method includes: obtaining models for a plurality of components comprised by the technical system; obtaining a plurality of validation measurements; for each component, training a machine learning model to predict measurement outputs of the respective component based on inputs of the respective component; obtaining first test outputs from a last model based on test inputs; determining second test outputs from the machine learning model corresponding to the last model and based on the test inputs of the models; determining a discrepancy; verifying and/or validating whether the technical system fulfills the criterion.