Metal Quality Prediction Model Using Area-Wise Process Data

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional methods for predicting the quality of metal materials fail to effectively utilize detailed manufacturing condition data collected by sensors, limiting the improvement in prediction accuracy.

Innovation Solution

A quality prediction model generation method that collects and associates manufacturing conditions with quality data for each process area, using machine learning techniques like linear regression and XGBoost to generate a model that predicts quality based on stored data, considering factors such as interchange of ends and faces, and cutting positions, to improve prediction accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional quality prediction methods using average values from record databases are used, then the system is simple to operate, but prediction accuracy is limited

Engineering Contradiction:

Improveprediction accuracyVSAvoiddata processing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the metal material into predetermined areas along the longitudinal direction and processes manufacturing condition data and quality data for each area separately. This segmentation allows the system to utilize detailed sensor data for each region while maintaining a structured approach to data management and model generation.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent transitions from conventional one-dimensional average value representation to a multi-dimensional data structure that preserves spatial information along the longitudinal direction. By organizing data in terms of predetermined areas rather than single averages, the system adds a spatial dimension to the prediction model, enabling more accurate quality predictions.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Measurement precision

If detailed manufacturing condition data collected by sensors is not utilized, then the data processing is simpler, but prediction accuracy cannot be improved

Engineering Contradiction:

Improveprediction accuracyVSAvoidutilization of sensor data

Core Design Contradiction:

Measurement precisionVSLoss of information

Solution Approach 1:

The patent performs preliminary actions by collecting and storing manufacturing condition data from sensors during the manufacturing process itself, rather than relying solely on post-manufacturing measurements. This preliminary data collection ensures that detailed process information is captured and preserved for later use in quality prediction.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent establishes a feedback mechanism where quality data measured after manufacturing is associated with the corresponding manufacturing condition data from predetermined areas. This feedback loop allows the system to continuously improve prediction accuracy by learning from actual quality outcomes and refining the relationship between manufacturing conditions and quality results.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20220261520A1Quality prediction model generation method, quality prediction model, quality prediction method, metal material manufacturing method, quality prediction model generation device, and quality prediction device

Publication Date: 2022.08.18 JFE STEEL CORP

AI summary

A quality prediction model generation method for a metal material manufactured through one or more processes includes: a first collection step of collecting a manufacturing condition of each of the processes for each of predetermined areas of the metal material; a second collection step of evaluating and collecting quality of the metal material manufactured through each process for each of the predetermined areas; a storage step of storing the manufacturing condition of each process and the quality of the metal material manufactured under the manufacturing condition in association with each other for each of the predetermined areas; and a model generation step of generating a quality prediction model that predicts quality of the metal material for each of the predetermined areas based on the stored manufacturing condition for each of the predetermined areas in each process.