Winder Defect Inspection via Dynamic Sensor Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing break detection and inspection systems for winders lack sensitivity, configurability, and integration, leading to unnecessary culling of saleable products and limited defect profiling, particularly in flex winders with increased defect toleration and masking requirements.
Innovation Solution
A method and system for inspecting defects in winders that involve scanning with sensors to detect defects, associating them with winding process parameters, and classifying into defect profiles using segmentation of roll build and sensor segments, with adjustable sensitivity and control of winding modules based on detected defects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing break detection and inspection systems are used in winders, then basic web break detection can be achieved, but sensitivity is insufficient leading to limited defect profiling and unnecessary culling of saleable products
Solution Approach 1:
The inspection system segments the web inspection process into multiple scanning zones (head box region, forming section, etc.) with different sensitivity levels and defect profiles. Multiple sensors are positioned at different locations to detect defects at various stages of web formation, allowing precise defect classification without requiring a single overly complex inspection system.
Solution Approach 2:
The system changes detection parameters dynamically by adjusting sensor sensitivity and inspection criteria based on the specific region being scanned and the type of defect detected. This allows the system to maintain high sensitivity for critical defects while tolerating minor variations in non-critical areas, improving overall measurement precision without proportionally increasing complexity.
2Adaptability or versatility
If existing inspection systems with fixed sensitivity are used, then simple defect detection can be achieved, but configurability is limited leading to unnecessary culling of saleable products in flex winders with increased defect toleration
Solution Approach 1:
The inspection system transitions from fixed sensitivity to dynamic, adjustable sensitivity levels. Operators can configure defect toleration thresholds based on product specifications and customer requirements, allowing the system to adapt to different production scenarios. This dynamic configurability prevents unnecessary culling of saleable products while maintaining quality standards.
Solution Approach 2:
The system incorporates feedback mechanisms where defect detection data is analyzed and used to adjust inspection parameters and defect profiles. This allows continuous optimization of the inspection process, reducing false positives and unnecessary product culling while maintaining high quality standards through adaptive configurability.
3Loss of information
If basic sensor scanning is used for defect detection, then simple defect identification can be achieved, but defect profiling capability is limited
Solution Approach 1:
The system segments defect detection into multiple categories and profiles based on defect location, type, and severity. By dividing the web into different scanning zones and assigning specific defect profiles to each zone, the system captures comprehensive defect information without requiring overly complex automated analysis, thus reducing information loss while maintaining manageable automation levels.
4Measurement precision
If high sensitivity defect detection is implemented, then precise defect identification can be achieved, but unnecessary culling of saleable products occurs due to lack of configurability
Solution Approach 1:
The system applies different quality standards and sensitivity levels to different regions of the web and different types of defects. Critical defects in key areas trigger culling, while minor defects in non-critical areas allow the product to pass. This local quality approach maintains high detection precision while preventing unnecessary waste of saleable products through region-specific tolerance levels.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Registration and inspection systems and methods for use on a winder are disclosed. The systems and methods include scanning a region proximate a winding module with one or more sensors to determine a product defect. The product defect is then associated with a winding process parameter, such as percent roll build information or the identity of the particular sensor to first detect the defect. The winding process parameter is used to classify the defect into one or more defect profiles. The defect profiles can be based at least in part on segmenting the roll build of a web onto a winding module into a plurality of inspection windows and segmenting a plurality of sensors into a plurality of inspection sensor segments.