Steel Production Specification Control for Disturbance-Robust Properties
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for producing high-grade steel materials with desired crystal grain and hardness are prone to variations due to disturbances in chemical composition, dimensions, or temperature, requiring repeated experiments to establish production guidelines, and lack robustness against production disturbances.
Innovation Solution
A production specification determination method and apparatus that acquires performance data and uses a prediction model to optimize production specifications, ensuring that material characteristics asymptotically approach desired values through back analysis and feedback control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If semi-empirical guidelines are established using laboratory experiments and trial production, then desired steel material characteristics can be obtained under ideal conditions, but the method lacks robustness when disturbances occur during actual production
Solution Approach 1:
The patent performs back analysis using a prediction model before actual production to determine optimal production specifications. This preliminary calculation predicts material characteristics based on proposed production parameters, allowing disturbances to be anticipated and compensated for in advance, eliminating the need for repeated trial productions.
Solution Approach 2:
The patent implements a feedback control mechanism where the prediction model's estimated material characteristics are compared with desired target values, and production specifications are iteratively adjusted until the estimated characteristics converge to the desired values. This closed-loop approach ensures robustness against disturbances by continuously correcting production parameters.
2Manufacturing precision
If high-grade steel material with greater strength and ductility is produced using conventional methods, then desired material characteristics can be achieved, but repeated experiments on laboratory level and trial production on actual machines are required
Solution Approach 1:
The patent replaces the mechanical trial-and-error experimentation process with a computational prediction model. The model calculates optimal production specifications and predicts material characteristics through information processing, substituting physical experiments with virtual simulations that provide the same design guidance without requiring actual material production and testing cycles.
3Reliability
If production guidelines are established semi-empirically, then some production disturbances can be handled, but variations in chemical composition, dimensions, or temperature cannot always produce the desired steel material
Solution Approach 1:
The patent uses a prediction model that accepts production specifications as input parameters and outputs estimated material characteristics. By systematically varying input parameters (chemical composition, dimensions, temperature, processing conditions) in the model, the optimal combination of parameters can be identified that guarantees desired material characteristics even when disturbances occur, providing a comprehensive control strategy rather than handling individual variations separately.
Data Source
AI summary
A production specification determination method, a production method, and a production specification determination apparatus that can increase robustness against disturbances during production of a metal material are provided. Included are the steps of acquiring at least one piece of performance data established after a predetermined process during production of a metal material, performing back analysis based on the at least one piece of performance data and a prediction model that relates production specifications and material characteristics, and searching for production specifications for after the predetermined process such that an estimated value for the material characteristics asymptotically approaches a desired value.