Cold Rolling Mill Chattering Detection With Neural Speed Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting chattering in cold rolling mills face challenges such as delayed prediction of chattering occurrence and frequent false positives when setting low vibration intensity thresholds, leading to productivity issues.
Innovation Solution
A method using a neural network model to predict chattering in cold rolling mills by analyzing multidimensional array information from rolling conditions and vibration data, allowing for accurate prediction and adjustment of rolling speed to prevent chattering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a low vibration intensity threshold is set for detecting chattering, then chattering detection accuracy is improved, but false positives increase and productivity deteriorates
Solution Approach 1:
The invention transitions from single-dimensional vibration intensity detection to multi-dimensional analysis by incorporating acceleration, velocity, and position information from acceleration sensors. This dimensional expansion enables more accurate chattering detection without requiring overly sensitive vibration thresholds, thereby reducing false positives while maintaining high detection accuracy.
Solution Approach 2:
The invention changes the detection parameters from simple vibration intensity to a composite analysis including acceleration, velocity, and position parameters. By analyzing multiple parameters simultaneously and identifying specific patterns (sudden acceleration followed by velocity increase and position change), the system achieves accurate chattering detection without triggering false alarms that would reduce productivity.
2Productivity
If vibration intensity threshold is set high to avoid false positives, then productivity is maintained, but chattering detection accuracy deteriorates
Solution Approach 1:
By adding acceleration and position dimensions to the detection analysis, the system can distinguish true chattering events from normal vibrations even at higher thresholds. The multi-parameter pattern recognition enables reliable detection without excessive sensitivity, maintaining both productivity and detection accuracy.
Solution Approach 2:
The system uses feedback from multiple sensors (acceleration, velocity, position) to continuously monitor and identify chattering patterns. This feedback mechanism allows the system to maintain higher detection thresholds while still accurately identifying chattering through pattern recognition, avoiding false positives that would interrupt production.
3Device complexity
If traditional vibration detection methods are used, then system complexity is low, but chattering prediction accuracy deteriorates
Solution Approach 1:
The invention adds acceleration sensing capability as an additional dimension to traditional vibration detection. This relatively simple addition of acceleration sensors and multi-parameter analysis provides significant improvement in chattering prediction accuracy without requiring overly complex system architecture.
Solution Approach 2:
The invention replaces simple mechanical vibration threshold detection with a more sophisticated analysis system that processes acceleration, velocity, and position data. This substitution of detection methodology, while increasing complexity, provides dramatically improved chattering prediction accuracy through pattern recognition of multi-parameter data.
Data Source
Figure 1
Figure 2
Figure 3(a)~3(d)
AI summary
A method of detecting chattering in a cold rolling mill, according to the present invention, includes a step of predicting occurrence of chattering during rolling of a material to be rolled, by inputting second multidimensional data to a prediction model, the second multidimensional data having been generated on the basis of condition data corresponding to array data related to the material to be rolled, the prediction model having been trained with explanatory variables and an objective variable, the explanatory variables being first multidimensional data generated on the basis of one-dimensional array data representing a past rolling record of rolling of rolled materials by means of a cold rolling mill, the objective variable being a past record of occurrence of chattering corresponding to the past rolling record.