Physics-Based Model Assisted Age Prediction of Real- Serviced Thermal Barrier Coated Samples Using Infrared Thermography
- In this study, an age prediction model is developed for the real serviced thermal barrier coated (TBC) samples. TBC is a multilayer coating applied on metallic structures exposed to high temperatures, such as gas turbine blades and aeroengine parts, to extend the operational life. One of the main challenges for developing an age prediction model is the unavailability of serviced samples and labelled datasets. So, in this study, experimentally validated numerical models are used for data generation. The study considers samples with three different service times, i.e., 0 (newly coated), 500 hours, and 1000 hours. The age prediction is done in two stages: thermal diffusivity prediction for blind real serviced samples with a trained 1D-CNN model and classification of samples based on the service hours into different classes using AI classification models. The results demonstrate that our approach provides reliable age estimations with a high correlation between actual and predicted agesIn this study, an age prediction model is developed for the real serviced thermal barrier coated (TBC) samples. TBC is a multilayer coating applied on metallic structures exposed to high temperatures, such as gas turbine blades and aeroengine parts, to extend the operational life. One of the main challenges for developing an age prediction model is the unavailability of serviced samples and labelled datasets. So, in this study, experimentally validated numerical models are used for data generation. The study considers samples with three different service times, i.e., 0 (newly coated), 500 hours, and 1000 hours. The age prediction is done in two stages: thermal diffusivity prediction for blind real serviced samples with a trained 1D-CNN model and classification of samples based on the service hours into different classes using AI classification models. The results demonstrate that our approach provides reliable age estimations with a high correlation between actual and predicted ages of samples. This method provides a non-destructive, efficient, and accurate method of evaluating the life of TBCs, which is a substantial improvement over traditional techniques that rely on complex destructive methods or microstructural analysis for the age evaluation.…


| Autor*innen: | Sruthi Krishna Kunji PurayilORCiD |
|---|---|
| Koautor*innen: | K. Balasubramaniam |
| Dokumenttyp: | Posterpräsentation |
| Veröffentlichungsform: | Präsentation |
| Sprache: | Englisch |
| Jahr der Erstveröffentlichung: | 2024 |
| Organisationseinheit der BAM: | 8 Zerstörungsfreie Prüfung |
| 8 Zerstörungsfreie Prüfung / 8.3 Thermografische Verfahren | |
| Veröffentlichende Institution: | Bundesanstalt für Materialforschung und -prüfung (BAM) |
| DDC-Klassifikation: | Technik, Medizin, angewandte Wissenschaften / Ingenieurwissenschaften / Ingenieurbau |
| Freie Schlagwörter: | AI for NDT; Infrared Thermography; NDE 4.0; Thermal barrier coatings |
| Themenfelder/Aktivitätsfelder der BAM: | Infrastruktur |
| Infrastruktur / Leichtbau | |
| Veranstaltung: | 17th Quantitative InfraRed Thermography Conference (QIRT) |
| Veranstaltungsort: | Zagreb, Croatia |
| Beginndatum der Veranstaltung: | 01.07.2024 |
| Enddatum der Veranstaltung: | 05.07.2024 |
| Verfügbarkeit des Dokuments: | Datei für die Öffentlichkeit verfügbar ("Open Access") |
| Datum der Freischaltung: | 28.01.2025 |
| Referierte Publikation: | Nein |
| Schriftenreihen ohne Nummerierung: | BAM Präsentationen |

