PCA-Based Heat Treatment Data Analysis for Process-Property Links
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing temperature profile setting systems for metal workpieces do not effectively output relationships between performance, structure or physical properties, and processing settings, limiting user understanding of how these factors interact during heat treatment.
Innovation Solution
An information processing device and method that performs principal component analysis on time series state data combined with performance, structure, or process data, generating principal component values and outputting information on relationships, correlations, or regression equations to visually and quantitatively represent these interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If existing temperature profile setting systems are used to set heat treatment parameters, then temperature profiles can be established for metal workpieces, but the systems cannot output relationships between performance, structure/physical property, and processing setting data
Solution Approach 1:
The patent introduces an information processing device as an intermediary between the temperature profile setting system and the user. This device includes an acquisition section that collects performance, structure/physical property, and processing setting data; an analysis section that processes this data using machine learning models; and an output section that presents the relationships. The intermediary transforms raw data into meaningful relationship information without requiring direct modification of the core heat treatment system.
2Manufacturing precision
If machine learning is applied to generate temperature profiles based on workpiece parameters, then temperature profiles can be optimized for heat treatment, but the systems cannot provide comprehensive relationship analysis between multiple data types
Solution Approach 1:
The information processing device is designed with multi-functionality to handle various data types and analysis requirements. The acquisition section can collect performance data, structure/physical property data, and processing setting data. The analysis section uses machine learning models that can process multiple data types simultaneously. The output section can present different types of relationship analyses, making the system versatile for comprehensive data analysis while maintaining temperature profile accuracy.
3Loss of information
If comprehensive data collection is performed to analyze relationships between performance, structure, and process settings, then complete relationship information can be obtained, but the computational load and system complexity increase
Solution Approach 1:
The system performs preliminary actions by pre-collecting and organizing performance data, structure/physical property data, and processing setting data in the acquisition section before analysis. The data is structured and prepared in advance, which reduces the computational load during the actual analysis phase. This preliminary data preparation enables complete relationship information to be obtained while managing computational requirements efficiently.
Data Source
AI summary
An information processing device that acquires a data set including a combination of state data representing a time series of states when a target object is worked combined with at least one of performance data representing a performance of the target object, data representing a structure or physical property of the target object, or process data representing a process setting value when the target object is worked. The information processing device generates a principal component value of each of a plurality of items of the state data in the acquired data set by executing principal component analysis on the plurality of items of state data. The information processing device outputs information representing a relationship between principal component values of the plurality of items of state data, and the at least one of the performance data, the data representing the structure or physical property, or the process data.