Machine Learning Part Assessment Beyond Nominal Tolerances
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for assessing the quality of manufactured gas turbine engine components, such as integrally bladed rotors, rely solely on geometric tolerances, leading to the rejection of functionally tolerant parts that fall outside nominal drawing limits.
Innovation Solution
A method and system utilizing machine learning and physical testing to evaluate components at multiple locations, comparing generated data to training data to determine functional tolerance, ensuring components meet functional dimensions before acceptance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If manual inspection of dimensions is used to assess component quality, then manufacturing precision can be verified against nominal tolerances, but functionally tolerant parts outside nominal limits are incorrectly rejected
Solution Approach 1:
The patent transforms the assessment from checking against fixed nominal tolerances to evaluating functional performance parameters. The machine learning model assesses whether parts meet functional requirements rather than strict geometric limits, allowing functionally tolerant parts to be accepted even if outside nominal dimensions.
Solution Approach 2:
The patent replaces manual dimensional inspection with an automated machine learning-based assessment system. This substitution enables more sophisticated evaluation that considers functional tolerance rather than just geometric compliance, improving the accuracy of quality assessment.
2Productivity
If machine learning assessment is implemented, then evaluation speed and accuracy are enhanced, but system complexity increases
Solution Approach 1:
The patent uses machine learning models that have been trained on historical assessment data to create a virtual assessment system. This copied knowledge from training data enables rapid evaluation without requiring complex physical testing infrastructure, achieving high evaluation speed while managing system complexity.
Solution Approach 2:
The patent performs preliminary training of machine learning models using historical assessment data before actual quality assessment. This preliminary action creates a ready-to-use assessment system that can rapidly evaluate parts without requiring complex real-time analysis, improving evaluation speed while the complexity is managed during the offline training phase.
Data Source
AI summary
A method of assessing the quality of a manufactured component includes the steps of manufacturing a component and generating data at each of a plurality locations through the component. The generated data is passed to a machine learning branch wherein the generated data is compared to training data at each of the plurality of locations to determine whether the component is of functionally tolerant dimensions at each of the plurality of locations. The manufactured component is accepted should the component be within functionally tolerant dimensions at each of the plurality of locations, and the component is rejected if the component fails to be within functionally tolerant dimensions respectively, at each of the plurality of locations. A system is also disclosed.