Advanced ultrasonic inspection technologies applied to the \nwelded joints of hydraulic turbine runners
Bibliographic record
Abstract
Due to the importance of energy production cost, it is critical to reduce unnecessary or unpredicted halts of power generation equipment. Hydro-Québec, as a major power generation company, uses models to estimate the service life of turbine runners to avoid the aforementioned halts. For these models, the characteristics of flaws in the runners are one of the most influential inputs. Since non-destructive testing (NDT) techniques are used to characterize these flaws, it is important to assess the reliability of these methods and to identify methods that could provide better inspection results. The current project aims to provide IREQ with the performance of NDT methods to supply reliable flaw data (both measured and simulated) for their life estimation model. By increasing the accuracy of life estimations, Hydro-Québec will be able to minimize the number of halts and hence reduce the power generation costs. Despite all the previous studies, there is an essential need to extend the knowledge on the detectability of the welding flaws in weld joints of hydroelectric turbine runners. \n \nThis research is part of a program aimed at better understanding the performance of ultrasonic testing technology for the inspection of high-stress areas in Francis runner weld joints. In this research, we will first try to thoroughly study the capability of advanced ultrasonic inspection technologies for hydraulic turbine runners. Inspection of the T-joint mock-up sample was carried out by various NDT methods, namely conventional pulse-echo, phased array, and total focusing method (TFM). With these results, detection rates were obtained in order to compare the effectiveness of each method. In the second step, the reliability of different NDT methods (UT, RT, PAUT, and TFM) in detecting flaws in welded components was investigated using a statistical approach based on the Probability of Detection (POD). The different inspection techniques could thus be compared based on a 90% POD (a90) to determine what is the flaw size that can be reliably detected. The first and second phases deal with the efficiency of ultrasonic inspection as applied to a mock-up sample made of SS415 plates and welded using the same materials and procedure. Finally, the demonstration on a real turbine runner using the highest POD techniques has been experimented. A dual inspection strategy could be implemented after manufacturing or during the in-service inspection with two pass inspections to improve the fitness-for-service assessment of hydraulic turbine runners. The first pass would consist of the use of PAUT with a conventional array while the second pass would be based on the use of TFM. PAUT has shown excellent sensitivity to volumetric and planar defects, while TFM provides more accurate flaw size dimension. This dual inspection strategy aims to increase the reliability of ultrasonic inspection, leading to reduced costs and improved reliability for the hydraulic turbine runner industry. The data collected on the real turbine opened a new opportunity to improve the NDT procedure development. \n \nDemonstrating how TFM has better sizing accuracy is crucial for optimizing the inspection process. This leads to an improvement in the detectability of flaws and improves the process of generating POD curves based largely on TFM examinations as the ground truth method to get the flaw size information. This makes it possible to draw POD curves without using destructive tests which are expensive to perform, and irreparably destroy the specimens. Our work is centered on a real turbine runner using various ultrasonic array configurations to characterize defects. Developing the inspection methodology for hydraulic turbine runners helps us to achieve better sizing measurements of flaws used in the fatigue assessment model. Also, the outcome of this research would be employed to improve the fitness-for-service assessment of hydraulic turbine runners after manufacturing or during inspection operations performed over the useful life of the part.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".