Files and links (1)
CC BY-NC-SA V4.0, Open Access
Thesis
VERIFYING MANUFACTURING PARAMETERS FOR COMPOSITE MATERIAL MACHINING USING LEARNED VIBRATION SIGNATURES
Washington State University
Master of Science (MS), Washington State University
01/2022
DOI:
https://doi.org/10.7273/000004348
Handle:
https://hdl.handle.net/2376/124529
Abstract
Fiber-reinforced plastics, such as those used in the aerospace industry, suffer from a complex manufacturing process that also requires very high levels of precision. Further, modern manufacturing equipment has limited ability to identify when manufacturing parameters deviate from specification. This work begins to bridge that gap, showing how some critical parameters such as tool type and feed rate can be verified during autonomous drilling using only a vibration sensor mounted on the drilling jig. We compare results between Principle Component Analysis + k-Nearest Neighbors and a deep neural network, and between frequency domain data and temporal data. Our analysis shows similar accuracy across both classifiers when using frequency data, but PCA+kNN shows robustness across a variety of hyperparameters suggesting it may be preferable given current data set sizes. Our analysis shows that classification accuracy is similar between summary frequency data and short temporal windows of data across a variety of deep neural network architectures.
Metrics
4 File views/ downloads
21 Record Views
Details
- Title
- VERIFYING MANUFACTURING PARAMETERS FOR COMPOSITE MATERIAL MACHINING USING LEARNED VIBRATION SIGNATURES
- Creators
- Christian Svinth
- Contributors
- Scott Wallace (Advisor)Ben McCamish (Committee Member)Xinghui Zhao (Committee Member)
- Awarding Institution
- Washington State University
- Academic Unit
- Engineering and Computer Science (VANC), School of
- Theses and Dissertations
- Master of Science (MS), Washington State University
- Publisher
- Washington State University
- Number of pages
- 51
- Identifiers
- OCLC#: 1370910274; 99900882930001842
- Language
- English
- Resource Type
- Thesis