Publication: Study of Holes Detection via Vibrational Analysis & Machine Vision in Aero-Composites Structure
dc.contributor.affiliation | #PLACEHOLDER_PARENT_METADATA_VALUE# | en_US |
dc.contributor.author | MOHD FIRDAUS BIN SHAMSUDDIN | en_US |
dc.contributor.supervisor | MUHAMMAD HANAFI BIN AZAMI,Assistant Professor | en_US |
dc.date.accessioned | 2024-10-08T03:18:27Z | |
dc.date.available | 2024-10-08T03:18:27Z | |
dc.date.issued | 2022 | |
dc.description.abstract | Currently, most of the processes within advanced composite manufacturing still being performed manually; such as pre-impregnated material manual hand layup, drilling and trimming. The same condition is true for its associated quality control check, where the inspection is also still performed manually either by visual or via measurement tools. In the shop floor of manual drilling process, the quantity of holes drilled are counted manually and consequently leading to the possibility of miscounting. This study explores a novel approach to automate the counting process of holes after the drilling process on aero-composites structure via two methods; vibrational analysis and machine vision. For vibrational analysis, the vibrational pattern of the pneumatic drill gun was measured and analyzed in order to identify and isolate signature vibrational pattern that could be associated with the drilling action. The second method, relies on an image processing system called machine vision; built on top of deep learning algorithm in order to visually detect the existence of drilled hole on the aero-composites structure. Testing were performed in the actual working environment and the result were used to evaluate and validate the feasibility of the vibrational analysis and/or machine vision architectures for deployment in the actual manufacturing floor. While vibrational analysis achieved a higher detection percentage of 71.4% compared to machine vision, 33.33%; the latter approach was found to be easier to implement due to the image processing algorithm. The lower detection score for machine vision may be attributed to the convolutional operation used within the deep learning algorithm; thus leading to difficulties in detecting the drilled hole due to color similarity within the aero-composites structure. For vibrational analysis, it was found that a major problem to the score improvement lies in the uncertainties within the manual drilling process since the accuracy of the vibrational analysis is highly sensitive to the handling of the pneumatic drill gun. | en_US |
dc.description.cpsemail | cps2u@iium.edu.my | en_US |
dc.description.degreelevel | Master | |
dc.description.email | firdaus.shamsuddin7@gmail.com | en_US |
dc.description.identifier | THESIS :Study of Holes Detection via Vibrational Analysis & Machine Vision in Aero-Composites Structure/MOHD FIRDAUS BIN SHAMSUDDIN | en_US |
dc.description.identity | G2010399 | en_US |
dc.description.kulliyah | Kulliyyah of Engineering | en_US |
dc.description.nationality | MALAYSIA | en_US |
dc.description.programme | Master of Science in Engineering | en_US |
dc.identifier.uri | https://studentrepo.iium.edu.my/handle/123456789/7140 | |
dc.language.iso | en | en_US |
dc.publisher | Kuala Lumpur : International Islamic University Malaysia, 2022 | en_US |
dc.rights | OWNED BY STUDENT | |
dc.title | Study of Holes Detection via Vibrational Analysis & Machine Vision in Aero-Composites Structure | en_US |
dc.type | Master Thesis | en_US |
dspace.entity.type | Publication |