Rolling Mill Setup Prediction Using Prior Operator-Adjusted Values
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining mill setup values in rolling mills face challenges in reflecting immediate adjustments made by operator intervention, as they rely on large datasets and mathematical models that require recalibration, making it difficult to incorporate manual modifications effectively.
Innovation Solution
A method using trained models that incorporate relations between previous and current mill setup values, allowing for the determination of setup conditions that reflect operator-driven modifications by inputting manufacturing conditions and previous setup values, enabling immediate reflection of adjustments through a neural network-based system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If mill setup values are determined based on a large volume of data on past records using trained neural networks, then automation and productivity are improved, but the ability to immediately reflect operator's manual adjustments in the next mill setup is worsened
Solution Approach 1:
The patent segments the determination of mill setup values into two distinct components: (1) automated determination using trained neural networks based on historical data, and (2) manual adjustment by operators. The system processes these separately and combines them, allowing each component to function optimally without interfering with the other's timing or methodology.
Solution Approach 2:
The system performs preliminary automated calculation of mill setup values using trained neural networks before operator intervention. The automated system prepares initial setup values in advance, and then the operator can immediately adjust these values based on current observations (such as motor load status), and the adjusted values are reflected in the next mill setup without requiring recalibration of the entire system.
2Manufacturing precision
If optimization is performed using sequential quadratic programming on a mathematical model, then manufacturing precision is improved, but the system becomes dependent on model precision and cannot quickly adapt to manual modifications
Solution Approach 1:
The patent introduces an intermediary mechanism that bridges the mathematical model optimization results and the actual mill setup values. The trained neural network acts as this intermediary, learning the relationship between model predictions and actual optimal values from historical data. This intermediary layer allows the system to benefit from both the structured optimization approach and the flexibility to adapt to manual adjustments without being constrained by model precision limitations.
3Reliability
If mill setup values are calculated using conventional automated methods, then consistency and reliability are improved, but the ability to incorporate immediate operator feedback is worsened
Solution Approach 1:
The system implements a feedback mechanism where operator observations and manual adjustments are continuously incorporated into the mill setup determination process. The trained neural network is designed to accept both automated input data and operator-provided adjustments as inputs, ensuring that feedback from operators is systematically integrated into subsequent setup calculations without disrupting the overall consistency and reliability of the automated system.
Data Source
Figure 1
Figure 2
Figure 3(a)~3(b)
AI summary
A setup condition determining method for manufacturing facilities includes: obtaining a setup condition for a target product by inputting, into a trained model, a manufacturing condition for the target product and a setup condition that is for a product manufactured in the same manufacturing facilities before manufacture of the target product and that reflects setup condition modification by an operator's manual manipulation, the trained model having been trained with input being manufacturing conditions for the target product and setup conditions that are for the product manufactured in the same manufacturing facilities before the manufacture of the target product and that reflect setup condition modification by an operator's manual manipulation, and output being setup conditions for the target product.