Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/517163
Title: | Surface Flaw Detection In Plug Valves Through Infrared Thermography And Fuzzy Deep Learning Algorithm |
Researcher: | JACINTHA VIJAYAKUMAR |
Guide(s): | Karthikeyan S |
Keywords: | Engineering Engineering and Technology Engineering Electrical and Electronic |
University: | Sathyabama Institute of Science and Technology |
Completed Date: | 2022 |
Abstract: | This Thesis addresses the detection and identification of flaws newlinein Plug valves. The Plug valve thermal images is acquired using Thermal newlineFluke camera (TiS20). Thermal images of plug valve is used for newlineidentification of flaws such as Crack, Porosity, Corrosion, Internal newlinedefects. The thermal images detects the surface flaws and never for newlinesubsurface flaws in Plug valves. The subsurface flaws detection is a newlinechallenging problem in valve inspection. In this Thesis, the thermal newlineimages obtained after dye penetrating the surface valve detects the surface newlineflaws more efficiently after applying the fuzzy deep learning algorithms. newlineDye-Penetrating Test (DPT) combined with Infrared Thermography to newlineidentify heat flux changes and flaws in the faulty metal surface of Plug newlinevalves is proposed. In DPT, thinned paint is employed on the metal newlinesurface that displays metal porosity and even fine cracks. After DPT, newlinethermal images of plug valve process through Fuzzy Deep Learning newlineAlgorithm to evaluate flaws. The Fuzzy Algorithm utilize prior to Deep newlineLearning to simplify and speed up the classification task. The flaws are newlineidentified using Slicing, Accuracy, Mathew s Correlation coefficient newline(MCC), local self-similarity descriptor (LSS). The parametric quantities newlinedepict corresponding variation with regard to surface coarseness and newlinemetal flaws. The DPT and Fuzzy Deep Learning Algorithm identify metal newlinedefect with 85.45% accuracy. newlineKeywords: Infrared Thermography, NDT (DPT), Fuzzy, Deep Learning, newlineSurface texture newline |
Pagination: | vi, 170 |
URI: | http://hdl.handle.net/10603/517163 |
Appears in Departments: | ELECTRONICS DEPARTMENT |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
10.chapter 6.pdf | Attached File | 1.57 MB | Adobe PDF | View/Open |
11.chapter 7.pdf | 310.56 kB | Adobe PDF | View/Open | |
12.annexure.pdf | 5.36 MB | Adobe PDF | View/Open | |
1.title.pdf | 190.62 kB | Adobe PDF | View/Open | |
2.prelim pages.pdf | 1.06 MB | Adobe PDF | View/Open | |
3.abstract.pdf | 311 kB | Adobe PDF | View/Open | |
4.contents.pdf | 343.41 kB | Adobe PDF | View/Open | |
5.chapter 1.pdf | 1.27 MB | Adobe PDF | View/Open | |
6.chapter 2.pdf | 165.21 kB | Adobe PDF | View/Open | |
7.chapter 3.pdf | 899.7 kB | Adobe PDF | View/Open | |
80_recommendation.pdf | 190.62 kB | Adobe PDF | View/Open | |
8.chapter 4.pdf | 1.07 MB | Adobe PDF | View/Open | |
9.chapter 5.pdf | 1.13 MB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: