Automated analysis of Baffle Bolts
The Baffle bolts (BB) are components of the lower internals of the vessel of Pressurized Water Reactor (PWR) nuclear power plants. They are subjected to ultrasonic testing (UT) inspection as part of the Long-Term Operation (LTO) of the PWR plants worldwide. Unlike other areas undergo inspection, us...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | deu |
| Published: |
NDT.net
2023-08-01
|
| Series: | e-Journal of Nondestructive Testing |
| Online Access: | https://www.ndt.net/search/docs.php3?id=28217 |
| _version_ | 1797356297787539456 |
|---|---|
| author | Javier de la Morena Amador Sillero |
| author_facet | Javier de la Morena Amador Sillero |
| author_sort | Javier de la Morena |
| collection | DOAJ |
| description |
The Baffle bolts (BB) are components of the lower internals of the vessel of Pressurized Water Reactor (PWR) nuclear power plants. They are subjected to ultrasonic testing (UT) inspection as part of the Long-Term Operation (LTO) of the PWR plants worldwide. Unlike other areas undergo inspection, usually there are more than one thousand bolts in a typical PWR reactor. The goal of the paper is to present how the Artificial Intelligence (AI) can be integrated into the inspection of the baffle bolts described in the introduction of the present document. The scope of the project was limited to the analysis of the signals acquired during the inspection to discriminate between healthy BB and those showing signals that need to be evaluated by an expert. Some different independent algorithms are used by the automated analysis system, which counts with a support decision system to achieve the most robust result possible. First inspection of its kind was carried out in winter 2021 with promising results, being scheduled the next one in Fall 2022. The expectation is to get the system qualified according to the criteria stablished by the normative. AI can be applied to UT inspections being the major limitation the amount of data, in a dual sense: train the data needed for the algorithms and obtain a confident improvement of the productivity. A key point of the project has been to create working teams of professionals that bring together knowledge in UT and AI.
|
| first_indexed | 2024-03-08T14:24:36Z |
| format | Article |
| id | doaj.art-d534be7f34bd4f439aa2c4a1bbdf8c4b |
| institution | Directory Open Access Journal |
| issn | 1435-4934 |
| language | deu |
| last_indexed | 2024-03-08T14:24:36Z |
| publishDate | 2023-08-01 |
| publisher | NDT.net |
| record_format | Article |
| series | e-Journal of Nondestructive Testing |
| spelling | doaj.art-d534be7f34bd4f439aa2c4a1bbdf8c4b2024-01-13T11:06:28ZdeuNDT.nete-Journal of Nondestructive Testing1435-49342023-08-0128810.58286/28217Automated analysis of Baffle Bolts Javier de la MorenaAmador Sillero The Baffle bolts (BB) are components of the lower internals of the vessel of Pressurized Water Reactor (PWR) nuclear power plants. They are subjected to ultrasonic testing (UT) inspection as part of the Long-Term Operation (LTO) of the PWR plants worldwide. Unlike other areas undergo inspection, usually there are more than one thousand bolts in a typical PWR reactor. The goal of the paper is to present how the Artificial Intelligence (AI) can be integrated into the inspection of the baffle bolts described in the introduction of the present document. The scope of the project was limited to the analysis of the signals acquired during the inspection to discriminate between healthy BB and those showing signals that need to be evaluated by an expert. Some different independent algorithms are used by the automated analysis system, which counts with a support decision system to achieve the most robust result possible. First inspection of its kind was carried out in winter 2021 with promising results, being scheduled the next one in Fall 2022. The expectation is to get the system qualified according to the criteria stablished by the normative. AI can be applied to UT inspections being the major limitation the amount of data, in a dual sense: train the data needed for the algorithms and obtain a confident improvement of the productivity. A key point of the project has been to create working teams of professionals that bring together knowledge in UT and AI. https://www.ndt.net/search/docs.php3?id=28217 |
| spellingShingle | Javier de la Morena Amador Sillero Automated analysis of Baffle Bolts e-Journal of Nondestructive Testing |
| title | Automated analysis of Baffle Bolts |
| title_full | Automated analysis of Baffle Bolts |
| title_fullStr | Automated analysis of Baffle Bolts |
| title_full_unstemmed | Automated analysis of Baffle Bolts |
| title_short | Automated analysis of Baffle Bolts |
| title_sort | automated analysis of baffle bolts |
| url | https://www.ndt.net/search/docs.php3?id=28217 |
| work_keys_str_mv | AT javierdelamorena automatedanalysisofbafflebolts AT amadorsillero automatedanalysisofbafflebolts |