Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Automated Defect Detection and Decision-support for internal hole inspection of critical aerospace parts: Feasibility study
University West, Department of Engineering Science, Division of Industrial Engineering and Management, Electrical- and Mechanical Engineering.
2021 (English)Independent thesis Advanced level (degree of Master (One Year)), 10 credits / 15 HE creditsStudent thesisAlternative title
Automatiserad upptäckt av defekter och beslutsstöd för inspektion av inre hål i kritiska flygplansdelar (Swedish)
Abstract [en]

This project was tasked with answering the question: “Can inner wall surfaces of a critical aerospace part be inspected by placing a camera on the outside and making use of an active positioning system, such that images can be acquired and defects can be detected on the image?” The reason why inspection is important is to ensure that the predetermined quality is achieved and no defects are present.

The inspection system is set up by using a lens, camera, and robotic arm. Which was mounted to the side of the GKN inspection module, that is used to rotate the part. To ensure all wall surfaces are clearly visible and in focus, the lens must be centered infront of the hole. To accomplish this the camera is fixed on the robot which allows it to be positioned precisely. The camera was calibrated at a specific distance from the part. In the image, the distance between the center of the hole and the center of the image is measured. This measurement is sent to the robotic arm to move those distances and position the lens accurately in front of the hole. 

Defect detection was performed by using the built-in surface defect-detection function in the Cognex software. A quantitative method was evaluated where the number of detected flaws returned by the software was used as a basis for determining the compliance of the hole. During the experimental phase, images were taken of 144 different holes present in the parts. The diameter and depth of each hole was around 5 and 10 mm respectively. All interior surfaces of holes were clearly visible and in focus. Then in 12 of the holes defects were introduced. These defects could be clearly seen in the images. However, the flaw detection algorithm was unable to correctly differentiate between the holes with defects and those without. This was due to the high level of noise present in the clean images. However, this visual inspection system allows for easier access to the tiny holes by displaying the image on a screen, greatly assisting the operator in determining if a defect is present in the hole or not

Place, publisher, year, edition, pages
2021. , p. 35
Keywords [en]
Automation, data acquisition, Image processing, Inspection, Robotics
National Category
Production Engineering, Human Work Science and Ergonomics
Identifiers
URN: urn:nbn:se:hv:diva-17398Local ID: EXP800OAI: oai:DiVA.org:hv-17398DiVA, id: diva2:1588535
Subject / course
Mechanical engineering
Educational program
Produktionsteknik, magister
Supervisors
Examiners
Available from: 2021-09-01 Created: 2021-08-27 Last updated: 2021-09-01Bibliographically approved

Open Access in DiVA

No full text in DiVA

By organisation
Division of Industrial Engineering and Management, Electrical- and Mechanical Engineering
Production Engineering, Human Work Science and Ergonomics

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 115 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf