Fiber Defect Detection via Grid Cell Statistical Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Automated manufacturing of fiber composite components faces challenges in quality assurance, leading to higher reject rates and increased costs due to the computational intensity of defect detection in large components, particularly in safety-critical applications like aerospace and automotive sectors.

Innovation Solution

A method involving a height profile determination device that divides the fiber material surface into grid cells, analyzes statistical deviations to identify potential defect areas, and uses machine learning for precise defect detection, reducing the need for additional detection devices and enabling real-time capability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If thermal cameras are used for preliminary defect detection, then detection speed is improved, but device complexity and cost increase

Engineering Contradiction:
Improvedefect detection speedVSAvoiddetection system complexity
Core Design Contradiction:
SpeedVSDevice complexity

Solution Approach 1:

The fiber material surface is divided into grid cells, and each grid cell is evaluated independently using statistical deviation of height levels. This segmentation allows parallel processing of multiple regions simultaneously, achieving real-time detection speed without requiring complex additional hardware systems.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If depth image recording is used for defect detection, then measurement precision is improved, but computational intensity and processing time increase

Engineering Contradiction:
Improvedefect detection precisionVSAvoiddata evaluation time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

Instead of evaluating the entire depth image data set, the method extracts and evaluates only specific characteristics from each grid cell - namely the statistical deviation of height levels. This extraction of essential features maintains defect detection precision while dramatically reducing computational intensity and processing time.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The evaluation focuses on local statistical properties (standard deviation of height levels) within each grid cell rather than analyzing the complete depth image. This local quality approach enables real-time processing by concentrating computational resources on evaluating specific local characteristics that indicate potential defects.

Inventive Principle:
Principle #3Local quality

3Reliability

If comprehensive quality assurance is implemented, then reliability is improved, but productivity decreases due to higher reject rates

Engineering Contradiction:
Improvequality assurance reliabilityVSAvoidmanufacturing productivity
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system performs preliminary evaluation of all grid cells using statistical deviation of height levels to identify potential defect areas before conducting more detailed analysis. This preliminary action enables targeted inspection, maintaining high reliability by detecting defects early while improving productivity by avoiding unnecessary comprehensive analysis of defect-free areas.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3432266B1Device and method for detecting faulty parts
Publication Date: 2022.11.16 DEUTSCHES ZENTRUM FÜR LUFT UND RAUMFAHRT E V
  • EP3432266B1 patent drawingFigure 1
  • EP3432266B1 patent drawingFigure 2
  • EP3432266B1 patent drawingFigure 3

AI summary

The invention relates to a method for detecting defects in fiber materials laid on a tool, wherein a height profile is determined using a laser light section sensor and defects within the fiber material are then detected based on the height profile. For this purpose, a preliminary analysis is first carried out in which individual sub-areas are statistically evaluated. Potential defect areas are then examined by a detailed pattern recognition analysis.