Regression Model for Microstructure Data Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The reliability of microstructure data in materials science is highly dependent on the skill of the person performing measurements, leading to variability and reduced estimation accuracy in machine learning applications.
Innovation Solution
A material data processing device and method that uses machine learning to create a regression model based on process data, composition data, characteristics data, and microstructure data, including a feature amount difference between magnetization temperature dependence during heating and cooling, for high-precision estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If microstructure data is obtained through measurement, observation, and analysis by personnel, then the data can be acquired, but the reliability varies greatly depending on the skill of the person performing the measurement
Solution Approach 1:
The patent replaces manual measurement, observation, and analysis operations with automated image processing and machine learning algorithms. Specifically, it uses computer vision techniques to automatically analyze microstructure images, extracting features such as phase distribution, grain size, and morphology without human intervention. This substitution of manual mechanical operations with automated computational methods eliminates the variability introduced by operator skill differences while maintaining data acquisition capability
Solution Approach 2:
The system enables self-service by allowing the microstructure analysis process to perform itself automatically. The image processing unit automatically processes microstructure images, and the machine learning model automatically extracts relevant features and creates regression models without requiring skilled operators. The system serves itself by having the data processing pipeline handle all operations from raw image input to predictive model output autonomously
2Productivity
If machine learning is performed using microstructure data with varying reliability, then estimation can be conducted, but the estimation accuracy is reduced
Solution Approach 1:
The patent transforms the microstructure data from raw images to standardized extracted features through systematic parameter changes. It converts visual microstructure information into quantitative parameters such as phase area ratios, grain size distributions, and morphological descriptors. This parameter transformation standardizes the input data for machine learning, ensuring consistent feature representation regardless of the original data quality variations, thereby improving estimation accuracy while maintaining productivity
Solution Approach 2:
The system performs preliminary action by pre-processing and standardizing microstructure data before it enters the machine learning estimation process. It automatically corrects image quality variations, normalizes feature extraction, and prepares standardized input data in advance. This preliminary processing ensures that the machine learning model receives consistent, high-quality input data, improving estimation accuracy without reducing productivity
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables easy data acquisition and high-precision estimation by stabilizing microstructure data quality and improving prediction accuracy in materials development.
Implementation Method 1
a feature amount difference that is a difference between a feature amount during heating as a feature amount based on a magnetization temperature dependence during heating and a feature amount during cooling as a feature amount based on a magnetization temperature dependence during cooling
Data Source
AI summary
A material data processing device is provided with a regression model creation processing unit that performs machine learning using, out of process data including information on manufacturing conditions for manufacturing individual samples, composition data including information on composition of the individual samples, characteristics data including information on characteristics of the individual samples, and microstructure data including information on microstructure of the individual samples, two or more data including the microstructure data, and creates a regression model representing a correlation between respective data; and an estimation processing unit that estimates, by using the regression model, the process data, the composition data, the characteristics data, or the microstructure data, having been used for the machine learning, wherein the microstructure data includes a feature amount difference that is a difference between a feature amount during heating as a feature amount based on a magnetization temperature dependence during heating and a feature amount during cooling as a feature amount based on a magnetization temperature dependence during cooling. A material data processing method includes performing machine learning to create the regression model, and estimating, by using the regression model, the process data, the composition data, the characteristics data, or the microstructure data.