Loading...

Acoustic simulation of ultrasonic testing and neural network used for diameter prediction of three-sheet spot welded joints

Ghafarallahi, E ; Sharif University of Technology | 2021

398 Viewed
  1. Type of Document: Article
  2. DOI: 10.1016/j.jmapro.2021.03.012
  3. Publisher: Elsevier Ltd , 2021
  4. Abstract:
  5. Ultrasonic Testing (UT) is one of the most common types of nondestructive methods that is being used in various industries, especially in the automotive industry. In this paper, qualitative and quantitative control of resistance spot welds on three-sheet joints was studied. Initially, mathematical model of ultrasonic waves was extracted for triple sheet joints. Then, acoustic simulation of ultrasonic testing on spot welds was performed using Finite Element Method (FEM). Afterwards, A Multilayer Perceptron (MLP) neural network was used to classify spot welds based on their diameter. There was a mean error of 20.9 % between peak amplitudes of numerical and theoretical models which the most reason was due to the dispersion effect. Also, theoretical formula for diameter calculation showed 14 % error. Results of Artificial Neural Network (ANN) had an error of 5.11 % on average which indicated capability of ANN to evaluate quality of spot welds. It was also found that diameter calculation using mathematical formulas are less accurate than estimation by ANN. Finally, the simulation and UT results of real samples were compared which indicated that ANN has over 90 % accuracy in identifying the state of spot welds. © 2021 The Society of Manufacturing Engineers
  6. Keywords:
  7. Automotive industry ; Errors ; Neural networks ; Nondestructive examination ; Quality control ; Spot welding ; Welds ; Acoustic simulations ; Automotives ; Element method ; Finite-element method modeling ; Multilayers perceptrons ; Neural-networks ; Non destructive ; Nondestructive methods ; Resistance spot weld ; Spot-welded joints ; Ultrasonic testing
  8. Source: Journal of Manufacturing Processes ; Volume 64 , 2021 , Pages 1507-1516 ; 15266125 (ISSN)
  9. URL: https://www.sciencedirect.com/science/article/abs/pii/S152661252100178X