Nuclear Techniques ›› 2020, Vol. 43 ›› Issue (2): 20201-020201.doi: 10.11889/j.0253-3219.2020.hjs.43.020201

• ACCELERATOR, RAY TECHNOLOGY AND APPLICATIONS • Previous Articles     Next Articles

Study of the measurement of the thickness of diffusion aluminizing layer based on X-ray fluorescence

Jichao LIU,Cheng WANG(),Pengli DAI   

  1. Air Force Engineering University, Xi’an 710038, China
  • Received:2019-11-15 Revised:2019-12-06 Online:2020-02-15 Published:2020-02-24
  • Contact: Cheng WANG E-mail:warrant_74@126.com

Abstract: Background

Aluminizing is often used as a protective coating for aero-engine turbine blades, but the current nondestructive testing equipment can not achieve accurate measurement of the coating thickness. The uniformity of aluminizing thickness has a great impact on the performance of turbine blades and the safety and stability of engine.

Purpose

This study aims to develope a nondestructive measurement method for the thickness of diffusion aluminized layer based on X-ray fluorescence absorption.

Methods

First of all, according to the theoretical calculation formula of X-ray fluorescence absorption, the linear relationship between the logarithm of the fluorescence intensity ratio (lnR) and the thickness of the fluorescence penetrating material (x) was obtained. Then a unitary linear regression model related to the main elements of K403 alloy and a multiple regression model based on the forward selection variable method were established. Finally, the comparative study of the prediction results of the two models was carried out.

Results

The experimental results show that the multi regression calculation model is relatively stable in comparison with the unitary regression calculation model, and the average relative error of measurement results under different thicknesses is only 3.2%.

Conclusions

The proposed method provides a convenient and feasible guiding idea for solving the problem of measuring the thickness of diffusion permeable layer.

Key words: Aluminizing, X-ray fluorescence, Forward selection variable, Multiple regression

CLC Number: 

  • TL99