Corrugated Board Plant Diagnostics for Predictive Failure Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Corrugated board production plants face significant downtime and maintenance costs due to wear and tear in functional units, leading to production losses and increased expenses, particularly in the wet-end section where stopping the line results in substantial waste and long restart times.
Innovation Solution
A predictive diagnostics method is implemented to monitor operational parameters of functional units by calculating statistical functions based on historicized data within a movable learning temporal window, allowing for real-time comparison and generation of predictive diagnostic information to anticipate and prevent failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If functional units are monitored using traditional methods, then maintenance can be performed, but downtime and production losses increase due to unexpected failures
Solution Approach 1:
The patent applies preliminary action by continuously monitoring operational parameters and calculating statistical functions (mean, variance, skewness, kurtosis) to detect early signs of functional unit degradation. This allows maintenance to be scheduled before failure occurs, preventing unexpected downtime while maintaining high reliability through proactive intervention.
Solution Approach 2:
The system implements feedback by continuously comparing current statistical function values against historical data and predefined thresholds. When deviations indicate potential failures, the system generates alerts that trigger maintenance actions, creating a closed-loop control system that improves reliability while minimizing unplanned downtime.
2Loss of time
If functional units are monitored and maintained proactively, then downtime is reduced, but measurement and detection complexity increases
Solution Approach 1:
The patent replaces complex manual analysis of operational parameters with automated computational methods. Statistical functions (mean, variance, skewness, kurtosis) are automatically calculated from sensor data and compared against historical patterns using computer algorithms, substituting mechanical/manual detection with electronic processing that reduces complexity while improving detection accuracy.
Solution Approach 2:
The system transforms raw operational parameters into meaningful diagnostic information by calculating statistical functions (mean, variance, skewness, kurtosis). This parameter transformation simplifies detection by converting complex time-series data into interpretable metrics that clearly indicate functional unit health status, reducing the difficulty of monitoring while maintaining high responsiveness.
3Loss of information
If statistical analysis is performed on operational parameters, then predictive diagnostic information is generated, but computational resources and system complexity increase
Solution Approach 1:
The patent segments the diagnostic process into distinct computational stages: data collection, statistical function calculation (mean, variance, skewness, kurtosis), historical comparison, and threshold evaluation. This segmentation distributes computational load and organizes complexity into manageable modules, reducing overall system complexity while preserving comprehensive diagnostic information.
Solution Approach 2:
The system calculates multiple statistical functions (mean, variance, skewness, kurtosis) to ensure thorough diagnostic coverage. While this exceeds the minimum single-metric approach, it provides redundant and complementary information that improves diagnostic accuracy without proportionally increasing complexity, as all calculations are performed automatically by the monitoring system.
Data Source
AI summary
A new method is disclosed for monitoring the operation of a corrugated board production plant, the method provides for detecting at least one operational parameter of a functional unit of the plant, for example a current absorbed by a motor. Then, the current value of a statistical function of the operational parameter is calculated in a current temporal window. The maximum value and the minimum value of the same statistical function are calculated based on historicized data of the operational parameter in question. By comparing the current value of the statistical function and the maximum and minimum values, a piece of information of predictive diagnostics is obtained.