Neural Network Rotor Alloy Design System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Selecting appropriate materials for gas turbine engine components to withstand extreme stresses and loads is challenging due to the need for specific microstructural features that ensure high strength, resistance to stress corrosion, and optimal workability, which existing methods often require costly prototyping and testing.
Innovation Solution
A neural network is trained to correlate microstructural features of alloys with material properties using image data and machine learning, determining non-linear relationships to identify suitable microstructural features for achieving desired properties, allowing for the design and manufacturing of optimized alloys without prototyping.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional prototyping and testing methods are used to select materials for gas turbine engine components, then material reliability and performance can be ensured, but development time and cost increase significantly
Solution Approach 1:
The patent performs preliminary computational actions by training machine learning models on existing material data before actual material selection. The system pre-processes and analyzes microstructural features, mechanical properties, and chemical compositions to create predictive models that can rapidly evaluate new alloy compositions without requiring physical prototyping, thus reducing development time while maintaining reliability assessment capabilities
Solution Approach 2:
The patent creates virtual copies of material behavior through computational models. Instead of physically testing each alloy composition, the system uses machine learning models trained on existing material data to predict the performance of new compositions. These virtual models replicate the relationship between microstructural features and material properties, allowing rapid evaluation without time-consuming physical experiments
2Reliability
If traditional prototyping and testing methods are used to select materials for gas turbine engine components, then material performance can be validated, but development cost increases significantly
Solution Approach 1:
The patent replaces expensive physical prototyping with computational models that copy material behavior. Machine learning models trained on existing material databases predict the performance of new alloy compositions, eliminating the need to manufacture and test physical samples for each composition evaluation, thus significantly reducing development costs while maintaining performance validation capability
Solution Approach 2:
The patent substitutes mechanical testing systems with computational prediction systems. Instead of using physical testing apparatus to measure mechanical properties of alloy samples, the system uses machine learning models that computationally predict mechanical behavior based on microstructural features and chemical composition, replacing costly and time-consuming mechanical testing with low-cost computational analysis
3Strength
If complex microstructural features are designed to achieve desired material properties, then material performance improves, but manufacturing complexity increases
Solution Approach 1:
The patent systematically varies microstructural parameters such as grain size, phase distribution, and precipitate characteristics to achieve desired material properties. The machine learning models identify optimal parameter combinations that deliver high strength while considering manufacturability, allowing systematic exploration of the parameter space to find solutions that balance performance and manufacturing complexity
Solution Approach 2:
The patent applies different microstructural features to different regions or aspects of the material to optimize specific properties. For example, certain microstructural characteristics may be targeted for high-strength requirements while other regions maintain features that facilitate manufacturing, allowing localized optimization of both performance and manufacturability through controlled variation of microstructural parameters
Data Source
AI summary
A method for designing a material for an aircraft component according to one example includes training a neural network to correlate microstructural features of an alloy with material properties of the alloy by at least providing a set of images of the alloy. Each of the images in the set of images has varied constituent compositions and at least one patch of corresponding data is embedded into the image. The method also includes determining non-linear relationships between the microstructural features and corresponding empirically determined material properties via a machine learning algorithm, receiving a set of desired material properties of the alloy for aircraft component, and determining a set of microstructural features capable of achieving the desired material properties of the alloy based on the determined non-linear relationships.