Planar Material Variation Diagnosis via Periodic Sensor Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing web inspection systems for planar materials, such as moving webs of films, paper, and metals, are inefficient in real-time identification of subtle variations and root cause analysis due to high speeds and complexity in manufacturing lines, making it difficult to detect faults or features like holes, spots, and coating issues effectively.
Innovation Solution
A method and apparatus for diagnosing subtle variations in planar materials moving along a manufacturing line by obtaining datasets from sensors, statistically decomposing variations into mechanical and fluid mechanical causes, and identifying periodic and non-periodic variations to attribute them to specific machine components or processing conditions, allowing for real-time analysis and corrective actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If continuous imaging is used for inspection, then the entire web can be inspected for faults or features, but the inspection cannot be performed in real time and requires later analysis
Solution Approach 1:
The system uses periodic sampling of the web at specific locations rather than continuous imaging. Sensors take measurements at discrete points along the web as it moves through the manufacturing line, analyzing variations at periodic intervals. This allows real-time processing while maintaining inspection effectiveness by focusing on critical measurement points rather than capturing every portion of the web continuously.
2Productivity
If the web moves at high speed through the manufacturing line, then productivity is improved, but the web moves too fast for human inspection or accurate analysis
Solution Approach 1:
The system replaces human visual inspection with automated sensor systems and computational analysis. Optical sensors, capacitive sensors, or other measurement devices automatically capture data from the moving web at high speeds, and algorithms analyze the variations in real time. This substitution of mechanical/optical systems for human inspection enables accurate measurement even when the web moves too quickly for human perception.
3Loss of information
If statistical decomposition is performed to identify root causes of variations, then the complexity of analysis increases, but the ability to identify specific machine component issues improves
Solution Approach 1:
The system decomposes the total variation in web properties into distinct components attributed to different machine elements. By segmenting the analysis into contributions from individual rollers, conveyors, and processing sections, the system identifies which specific component is causing variations. This segmentation transforms a complex overall variation problem into manageable discrete source identification, reducing analytical complexity while improving root cause detection.
Data Source
AI summary
The present disclosure provides for identifying subtle variations in a planar material and in real time associating the subtle variations with a cause. Data from gauging or optical inspection of the planar material on a manufacturing line is analyzed in real time for certain root causes of identified variations in the planar material. The data is analyzed at a predetermined longitudinal frequency, averaged and compared to an estimated effect of a known variation source to identify a residual variation. The process is iterative to identify all statistically significant causes of the subtle variations in the planar material.