Physical constraints-based terahertz thickness measurement method of thermal barrier coating

Due to the uneven microstructure of the thermal barrier coatings (TBCs) and dispersion effect, there is a discrepancy between the measured terahertz (THz) signals and the theory, which leads to the optimization process of the model inversion method is prone to local extremes. This results in a lack...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:NDT & E international : independent nondestructive testing and evaluation Jg. 143; S. 103058
Hauptverfasser: Cao, Binghua, Deng, Tao, Fan, Mengbao, Sun, Fengshan, Ye, Bo
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Ltd 01.04.2024
Schlagworte:
Teaching-learning based optimization algorithm
Terahertz non-destructive testing
Theoretical model
Thermal barrier coatings
Thickness measurement
Teaching-learning based optimization algorithm
Thermal barrier coatings
Terahertz non-destructive testing
Theoretical model
Thickness measurement
ISSN:0963-8695, 1879-1174
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Due to the uneven microstructure of the thermal barrier coatings (TBCs) and dispersion effect, there is a discrepancy between the measured terahertz (THz) signals and the theory, which leads to the optimization process of the model inversion method is prone to local extremes. This results in a lack of reliability of THz thickness measurements based on model inversion. A THz inversion method based on physical constraints is proposed to measure the thickness of TBCs. In this paper, we illustrate the characteristics of the THz signal and propose a fitness function suitable for inversion of TBCs thickness. A feature comparison and adaptive mutation combined with teaching-learning-based optimization algorithm (FCAM-TLBO) is also proposed for inverting the theoretical model to extract ceramic layer thickness. The experiments show that the average relative error is less than 0.3 % for thickness measurements of TBC samples, which has better reliability than the conventional inversion methods.
ISSN:0963-8695
1879-1174
DOI:10.1016/j.ndteint.2024.103058