University of Twente Student Theses
The Application of automated quality control into a cold forming manufacturing process
Hof, L. (2024) The Application of automated quality control into a cold forming manufacturing process.
Full text not available from this repository.
Full Text Status: | Access to this publication is restricted |
Abstract: | Quality plays an essential role in all products nowadays. To meet high standards, product developers determine specified requirements. Philips produces shaving caps by combining cold forming steps with a heat treatment. Research objectives on cold form defects, inspection methods, inspection requirements, and neural networks are combined into a research plan. A literature review is performed on cold form defects and inspection methods. For the processing of the data, the evolution of traditional ruled based systems into deep neural networks is explained. The total inspection is divided into geometric inspection and visual inspection. It was determined that a sensor that can measure the height deviation of lamellae is required. The measurement should have an accuracy of 200nm in the z-direction. For the visual defects, a test setup was designed and used to acquire images used for the training, tuning and testing of the adjusted Patchcore model. An AUROC score of 89.58 \% was researched. For the geometric inspection, the Heliotis H9 sensor was chosen based on standard deviation calculation and the list of specifications. The slip-through rate is expected to be between 300-400 PPM. The research objectives and company goals were met by proving the feasibility of automated quality control. |
Item Type: | Essay (Master) |
Clients: | Philips, Drachten, Netherlands |
Faculty: | ET: Engineering Technology |
Subject: | 52 mechanical engineering |
Programme: | Mechanical Engineering MSc (60439) |
Link to this item: | https://purl.utwente.nl/essays/100443 |
Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Repository Staff Only: item control page