Aircraft Alloy Microstructure Selection Using Neural Property Mapping

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Selecting appropriate materials for gas turbine engine components that can withstand extreme stresses and loads is challenging due to the complexity of microstructural features and their correlation with material properties, requiring a more efficient method than conventional approaches.

Innovation Solution

A deep learning-based neural network is trained on microstructural images and corresponding material properties to determine non-linear relationships, enabling the identification of microstructural features necessary for achieving desired material properties in aircraft components, such as Young's modulus, yield strength, and thermo-elastic strain, and optimizing alloy composition and grain refinement for improved workability and plasticity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional material selection methods are used for gas turbine engine components, then the design process is straightforward and easy to understand, but the ability to accurately correlate complex microstructural features with material properties is insufficient

Engineering Contradiction:

Improvecorrelation accuracy between microstructural features and material propertiesVSAvoidcomplexity of analysis system

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces conventional mechanical/visual inspection methods with an automated image processing and neural network system. The system uses digital images of microstructural features, processes them through convolutional neural networks, and automatically correlates them with material properties, eliminating the need for manual analysis while significantly improving accuracy.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces an intermediary computational system that acts as a bridge between microstructural images and material properties. The neural network serves as an intermediary that learns complex non-linear relationships from training data, enabling accurate prediction of material properties from microstructural features without direct human intervention.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If prototyping is used to verify material properties, then empirical data can be obtained, but the development time and resource consumption increase significantly

Engineering Contradiction:

Improveempirical validation of material propertiesVSAvoiddevelopment time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent performs preliminary action by training the neural network on existing empirical data from previous prototyping and testing. Once trained, the model can predict material properties for new compositions without requiring physical prototyping, thereby reducing development time while maintaining reliability through the use of empirically-derived training data.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent creates a virtual copy of the material testing process through the neural network model. Instead of physically manufacturing and testing prototypes, the system uses digital copies (images) of microstructural features and processes them through the trained model to obtain predicted material properties, significantly reducing time and resource consumption.

Inventive Principle:

Principle #26Copying

3Loss of information

If the neural network uses full-size RGB images, then the microstructural feature information is complete and detailed, but the computational processing time and memory requirements increase

Engineering Contradiction:

Improvemicrostructural feature information retentionVSAvoidcomputational energy consumption

Core Design Contradiction:

Loss of informationVSUse of energy by moving object

Solution Approach 1:

The patent applies segmentation by dividing the image processing task into distinct stages: preprocessing (resizing to standardized dimensions), feature extraction through convolutional layers, and property prediction. This segmentation allows the system to process images efficiently while retaining essential microstructural information needed for accurate material property prediction.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS11485520B2Material selection and optimization process for component manufacturing

Publication Date: 2022.11.01 RTX CORP

AI summary

A method for designing a material for an aircraft component 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 to the neural network. Each of the images in the set of images has varied constituent compositions. The method further includes providing the neural network with a set of determined material properties corresponding to each image, associating the microstructural features of each image with the set of empirically determined data corresponding to the image, and 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.