Multi-Variable Predictive Control for Cold Rolling Mill Transients
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current metal rolling mills face challenges in achieving robust and high-performance thickness control due to time delays, non-linearities, and internal disturbances, leading to suboptimal gauge and flatness control, especially during acceleration and deceleration states.
Innovation Solution
A model-based multi-variable predictive control system is implemented for 4-hi non-reversible single-stand metal rolling mills, utilizing a model predictive controller (MPC) to regulate roll gap, roll force, mill speed, and strip tension for automatic gauge and flatness control, incorporating sensors and actuators to adjust parameters in real-time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If PID based closed loop control is used, then the control system is simple to implement, but the product optimization time is high and control accuracy is insufficient
Solution Approach 1:
The control system performs preliminary actions by predicting future gauge and flatness deviations using a process model before they actually occur. The model-based predictive controller calculates optimal control moves in advance based on predicted disturbances and plant response, allowing the system to proactively compensate for deviations rather than reactively correcting them after PID control fails to prevent them
Solution Approach 2:
The control system dynamically adapts to changing operating conditions by continuously updating its predictions based on actual measurements and model parameters. The multi-variable predictive controller adjusts control actions in real-time based on predicted disturbances and plant dynamics, enabling the system to maintain high control accuracy across varying rolling speeds and material properties where fixed-gain PID control fails
2Reliability
If traditional PID control is used, then the control strategy is easy to operate, but it cannot effectively handle time delays and non-linearities in the rolling process
Solution Approach 1:
The control system introduces an intermediate predictive model that acts as a mediator between the measured disturbances and the control actuators. This model-based intermediary predicts the plant response to disturbances and calculates optimal control moves, bridging the gap between simple PID control and the complex multi-variable rolling process with time delays and non-linearities
Solution Approach 2:
The control system replaces the mechanical PID control approach with an information-based predictive control system that uses process models and measurements to compute optimal control actions. This substitution of control mechanism enables the system to handle time delays and non-linearities by using predictive calculations rather than reactive error correction
3Manufacturing precision
If PID control is used during acceleration and deceleration states, then the control system maintains simplicity, but gauge and flatness control performance deteriorates
Solution Approach 1:
The control system dynamically adapts to different rolling states by continuously updating its predictions based on actual measurements and model parameters. The multi-variable predictive controller adjusts control actions in real-time based on predicted disturbances and plant dynamics, enabling the system to maintain high control accuracy across varying rolling speeds and material properties where fixed-gain PID control fails
Solution Approach 2:
The control system changes its control parameters and predictions based on the current rolling state. By using a process model that captures the dynamic behavior of the rolling mill, the controller adapts its predictions and control moves to match the actual plant response during acceleration, deceleration, and steady-state operation, rather than relying on fixed PID gains that cannot accommodate transient conditions
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
A control system employs a model-based multi-variable predictive control for cold rolling mills to improve sheet thickness uniformity to meet or exceed specifications in flatness. Sheet metal thickness and flatness deviations from standard requirements are significantly reduced with attendant improved control accuracy as compared to traditional control approaches that use PID based closed loop controls. The control system is particularly suited for control of 4-hi non-reversible single-stand metal rolling mills. The mill stand has a first work roll and a second work roll respectively positioned between a first back up roll and a second back up roll. A plurality of sensors measures and acquires property data of the sheet of material. A model predictive controller manipulates actuators to regulate thickness and flatness. The controller executes automatic gauge control, which is machine direction metal control, and automatic flatness control, which is cross direction metal sheet control, as metal sheet is rolled.