Rolling Line Energy Prediction via Parameter Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing energy consumption prediction methods for rolling lines lack accuracy in determining the influence of individual parameters, such as roll torque and roll velocity, leading to uncertainties in prediction errors, and fail to account for energy consumption in facilities like winders and conveying tables.
Innovation Solution
An energy consumption predicting device that includes a setting calculation unit, actual value collection unit, actual energy consumption acquisition unit, energy consumption learning value calculation unit, pseudo-actual energy consumption calculation unit, and correction learning value calculation unit, which analyze and correct prediction values by dividing actual and pseudo-actual energy consumption calculations to refine the model expression inputs for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all various actual operating values are input into the model expression to calculate energy consumption, then the calculation covers all parameters, but it is unclear which parameter exerts more influence on prediction error
Solution Approach 1:
The patent segments the calculation process into multiple stages: first calculating a preliminary energy consumption value using all operating values, then calculating correction values for each individual parameter (roll torque, roll velocity, etc.) separately. This segmentation allows identification of which parameters have greater influence on prediction errors while maintaining comprehensive parameter coverage.
Solution Approach 2:
The patent changes parameters by calculating correction values that represent the difference between actual operating values and set operating values for each parameter. By analyzing these parameter changes and their corresponding correction values, the system identifies which parameters exert more influence on energy consumption prediction errors.
2Measurement precision
If the model expression includes multiple parameters such as roll torque and roll velocity, then comprehensive energy consumption calculation is achieved, but the error influence of individual parameters cannot be distinguished
Solution Approach 1:
The patent segments the error analysis by calculating separate correction values for each parameter (roll torque correction value, roll velocity correction value, etc.). This allows the complex multi-parameter model to be analyzed in terms of individual parameter contributions to prediction errors, reducing the perceived complexity through structured breakdown.
Solution Approach 2:
The patent applies local quality by treating each parameter differently through individual correction value calculations. Instead of uniform treatment of all parameters, the system identifies and analyzes specific parameters (such as roll torque versus roll velocity) that have different influences on prediction errors, allowing targeted refinement of the model.
3Measurement precision
If learning values are calculated by comparing actual energy consumption with calculated values, then prediction accuracy improves, but the influence of individual operating parameters on errors remains unclear
Solution Approach 1:
The patent segments the learning value calculation into parameter-specific correction values. Instead of calculating a single aggregate learning value, the system calculates separate correction values for roll torque, roll velocity, and other parameters, preserving information about which parameters contribute most to prediction errors.
Solution Approach 2:
The patent implements feedback by using the calculated correction values to refine the model expression. The correction values provide feedback information about parameter influence that is fed back into the system to improve future predictions, maintaining both accuracy improvement and parameter influence information.
Data Source
AI summary
The present invention includes (1) inputting, into a model expression that defines relation between various operating values of a facility operating on a material to be rolled and energy consumption of the facility, various actual operating values as the various operating values, to calculate an actual calculation value of the energy consumption; (2) dividing the actual value of the energy consumption by the actual calculation value to calculate a reference learning value of the energy consumption; (3) inputting the set operating value defined by the setting calculation unit only in one operating value, among various operating values of the model expression, while inputting the actual operating values collected by the actual value unit in other operating values to calculate a pseudo-actual calculation value of the energy consumption; (4) dividing the actual calculation value by the pseudo-actual calculation value to calculate a correction learning value; and (5) inputting the various set operating values as the various operating values of the model expression to calculate a prediction value of the energy consumption for the material to be rolled, which is scheduled to be conveyed to the rolling line next time or later, and multiplies the prediction value by the reference learning value and the correction learning value to calculate a corrected prediction value of the energy consumption.