Web Defect Root Cause Identification via Pattern Recognition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current web inspection systems in papermaking machines cannot effectively identify the root causes of repetitive defects, leading to equipment damage and costly repairs due to their inability to analyze the nature, location, and source of defects within the papermaking process.

Innovation Solution

The implementation of diagnostic patterns associated with selected papermaking machine components, using pattern recognition techniques to analyze and classify dynamic paper defects, allowing for the identification of their root causes and enabling corrective actions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If web inspection systems capture and analyze all process images and data, then quality data is delivered to support decisions, but root cause analysis for repetitive defects is not provided

Engineering Contradiction:
Improvedefect root cause informationVSAvoidsystem complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The patent segments the continuous web inspection data into discrete defect events, each with specific characteristics (location, type, timing). By dividing the overall inspection process into individual defect analysis units, the system can focus root cause analysis on specific repetitive defects rather than processing all data uniformly, thereby providing root cause information without proportionally increasing overall system complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the analysis parameters from general quality inspection to specific defect pattern recognition. By transforming raw inspection data into standardized defect parameters (repetitiveness, location, characteristics), the system enables root cause identification through pattern matching while maintaining manageable system complexity through parameter standardization.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If existing WIS systems display video sequences for detected defects, then visual inspection is improved, but identification of specific equipment sources causing defects is not achieved

Engineering Contradiction:
Improvedefect location precisionVSAvoidequipment source information
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The patent implements a feedback loop where defect detection results are fed back into the inspection system with enhanced metadata including equipment source identification. The system uses the detected defect characteristics to query and return information about potential equipment sources, creating a closed-loop system that provides both precise location data and equipment source identification without requiring separate inspection systems.

Inventive Principle:
Principle #23Feedback

3Productivity

If papermaking machines operate continuously, then productivity is maintained, but equipment damage from undetected defects leads to costly repairs and shutdowns

Engineering Contradiction:
Improvemachine run timeVSAvoidequipment reliability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent applies preliminary action by identifying and flagging repetitive defects before they cause equipment damage. The system continuously monitors for patterns of recurring defects and alerts operators to potential equipment issues before they escalate to failures requiring shutdowns, thereby maintaining both productivity through continuous operation and reliability through preventive detection.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3425112B1Process and system for monitoring characteristics of defects in a web which is moving in a web-making machine and for identifying the causes of the web defects
Publication Date: 2020.08.05 HONEYWELL LTD(CA)
  • EP3425112B1 patent drawingFigure 1
  • EP3425112B1 patent drawingFigure 2
  • EP3425112B1 patent drawingFigure 3

AI summary

Monitoring continuous repetitive web sheet defects to determine the causes of such defects by creating unique diagnostic patterns associated with selected components of a web-making machine and employing pattern recognition techniques to analyze and classify web-sheet dynamic and deterministic defect patterns, and identify their root causes. Corrective actions can be effected thereby enhancing machine run-time to minimize late deliveries and improves overall product quality. The technique which is integrated with quality control system can be applied to the manufacturer of paper, packaging, rubber sheets, plastic film, metal foil, and the like.