Next Article in Journal
Opportunistic Large Array Propagation Models: A Comprehensive Survey
Previous Article in Journal
Image Sensors for Wave Monitoring in Shore Protection: Characterization through a Machine Learning Algorithm
Article

A Spatio-Temporal Ensemble Deep Learning Architecture for Real-Time Defect Detection during Laser Welding on Low Power Embedded Computing Boards

1
Fraunhofer-Institute for Laser Technology ILT, Steinbachstrasse 15, 52074 Aachen, Germany
2
Faculty of Mechanical Engineering, Chair for Laser Technology, RWTH Aachen University, Steinbachstrasse 15, 52074 Aachen, Germany
*
Author to whom correspondence should be addressed.
Academic Editor: Andrea Cataldo
Sensors 2021, 21(12), 4205; https://doi.org/10.3390/s21124205
Received: 12 May 2021 / Revised: 9 June 2021 / Accepted: 15 June 2021 / Published: 18 June 2021
(This article belongs to the Topic New Frontiers in Industry 4.0)
In modern production environments, advanced and intelligent process monitoring strategies are required to enable an unambiguous diagnosis of the process situation and thus of the final component quality. In addition, the ability to recognize the current state of product quality in real-time is an important prerequisite for autonomous and self-improving manufacturing systems. To address these needs, this study investigates a novel ensemble deep learning architecture based on convolutional neural networks (CNN), gated recurrent units (GRU) combined with high-performance classification algorithms such as k-nearest neighbors (kNN) and support vector machines (SVM). The architecture uses spatio-temporal features extracted from infrared image sequences to locate critical welding defects including lack of fusion (false friends), sagging, lack of penetration, and geometric deviations of the weld seam. In order to evaluate the proposed architecture, this study investigates a comprehensive scheme based on classical machine learning methods using manual feature extraction and state-of-the-art deep learning algorithms. Optimal hyperparameters for each algorithm are determined by an extensive grid search. Additional work is conducted to investigate the significance of various geometrical, statistical and spatio-temporal features extracted from the keyhole and weld pool regions. The proposed method is finally validated on previously unknown welding trials, achieving the highest detection rates and the most robust weld defect recognition among all classification methods investigated in this work. Ultimately, the ensemble deep neural network is implemented and optimized to operate on low-power embedded computing devices with low latency (1.1 ms), demonstrating sufficient performance for real-time applications. View Full-Text
Keywords: real-time process monitoring; recurrent neural network; high-speed infrared imaging; convolutional neural network; lack of fusion (false friends); feature importance; AI edge device real-time process monitoring; recurrent neural network; high-speed infrared imaging; convolutional neural network; lack of fusion (false friends); feature importance; AI edge device
Show Figures

Figure 1

MDPI and ACS Style

Knaak, C.; von Eßen, J.; Kröger, M.; Schulze, F.; Abels, P.; Gillner, A. A Spatio-Temporal Ensemble Deep Learning Architecture for Real-Time Defect Detection during Laser Welding on Low Power Embedded Computing Boards. Sensors 2021, 21, 4205. https://doi.org/10.3390/s21124205

AMA Style

Knaak C, von Eßen J, Kröger M, Schulze F, Abels P, Gillner A. A Spatio-Temporal Ensemble Deep Learning Architecture for Real-Time Defect Detection during Laser Welding on Low Power Embedded Computing Boards. Sensors. 2021; 21(12):4205. https://doi.org/10.3390/s21124205

Chicago/Turabian Style

Knaak, Christian, Jakob von Eßen, Moritz Kröger, Frederic Schulze, Peter Abels, and Arnold Gillner. 2021. "A Spatio-Temporal Ensemble Deep Learning Architecture for Real-Time Defect Detection during Laser Welding on Low Power Embedded Computing Boards" Sensors 21, no. 12: 4205. https://doi.org/10.3390/s21124205

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop