The validation of laser welding of metallic materials is challenging due to its highly dynamic processes and limited accessibility to the weld. The measurement of process emissions and the processing laser beam are one way to record highly dynamic process phenomena. However, these recordings always take place via the surface of the weld. Phenomena on the inside are only implicitly recognizable in the data and require further processing. To increase the validity of the diagnostic process, the multispectral emission data are synchronized with synchrotron data consisting of in situ high-speed images based on phase contrast videography. The welding process is transilluminated by synchrotron radiation and recorded during execution, providing clear contrasts between solid, liquid, and gaseous material phases. Thus, dynamics of the vapor capillary and the formation of defects such as pores can be recorded with high spatial and temporal resolution of <5 μm and >5 kHz. In this paper, laser welding of copper Cu-ETP and CuSn6 is investigated at the Deutsches Elektronen-Synchrotron (DESY). The synchronization is achieved by leveraging a three-stage deep learning approach. A preprocessing Mask-R-CNN, dimensionality reduction PCA/Autoencoders, and a final LSTM/Transformer stage provide end-to-end defect detection capabilities. Integrated gradients allow for the extraction of correlations between defects and emission data. The novel approach of correlating image and sensor data increases the informative value of the sensor data. It aims to characterize welds based on the sensor data not only according to IO/NIO but also to provide a quantitative description of the defects in the weld.

1.
G.
Xiangdong
,
Y.
Deyong
, and
K.
Seiji
, “
The high frequency characteristics of laser reflection and visible light during solid state disk laser welding
,”
Laser Phys. Lett.
12
,
076003
(
2015
).
2.
F.
Vakili-Farahani
,
J.
Lungershausen
, and
K.
Wasmer
, “
Wavelet analysis of light emission signals in laser beam welding
,”
J. Laser Appl.
29
, 022424 (
2017
).
3.
I.
Eriksson
,
J.
Powell
, and
A. F. H.
Kaplan
, “
Signal overlap in the monitoring of laser welding
,”
Meas. Sci. Technol.
21
,
105705
(
2010
).
4.
S.
Shevchik
, “
Supervised deep learning for real-time quality monitoring of laser welding with X-ray radiographic guidance
,”
Sci. Rep.
10
,
3389
(
2020
).
5.
J.
Heilmeier
,
M. K.
Kick
,
S.
Grabmann
,
T.
Muschol
,
F.
Schlicht
,
F.
von Hundelshausen
,
H.-G.
von Ribbeck
,
T.
Weiss
, and
M. F.
Zaeh
, “
Inline failure detection in laser beam welding of battery cells: Acoustic and spectral emission analysis for quality monitoring
,”
J. Laser Appl.
36
,
022007
(
2024
).
6.
A.
Olowinsky
,
A.
Gillner
, and
C.
Spurk
,
Laser Meets Synchrotron
(
Fraunhofer ILT
,
2024
), see https://laser-meets-synchrotron.de
7.
H.
Hotelling
, “
Analysis of a complex of statistical variables into principal components
,”
J. Educ. Psychol.
24
,
417
(
1933
).
8.
G. E.
Hinton
and
R. R.
Salakhutdinov
, “
Reducing the dimensionality of data with neural networks
,”
Science
313
,
504
507
(
2006
).
9.
K.
Hornik
,
M.
Stinchcombe
, and
H.
White
, “
Multilayer feedforward networks are universal approximators
,”
Neural Netw.
2
,
359
366
(
1989
).
10.
S.
Hochreiter
and
J.
Schmidhuber
, “
Long short-term memory
,”
Neural Comput.
9
,
1735
1780
(
1997
).
11.
M.
Schuster
and
K. K.
Paliwal
, “
Bidirectional recurrent neural networks
,”
IEEE Trans. Signal Process.
45
,
2673
2681
(
1997
).
12.
A.
Vaswani
,
N.
Shazeer
,
N.
Parmar
,
J.
Uszkoreit
,
L.
Jones
,
A.
Gomez
,
L.
Kaiser
, and
I.
Polosukhin
, “
Attention is all you need
,” arXiv:1706.03762 [cs.CL] (
2017
).
13.
Q.
Wen
,
L.
Sun
,
F.
Yang
,
X.
Song
,
J.
Gao
,
X.
Wang
, and
H.
Xu
, “Time series data augmentation for deep learning: A survey,” arXiv:2002.12478 (
2020
).
14.
Y.
Kang
,
Y.
Zhao
,
Y.
Li
,
J.
Wang
, and
X.
Zhan
, “
Simulation of the effect of keyhole instability on porosity during the deep penetration laser welding process
,”
Metals
12
,
1200
(
2022
).
15.
M.
Sundararajan
,
A.
Taly
, and
Q.
Yan
, “
Axiomatic attribution for deep networks
,” arXiv:1703.01365 [cs.LG] (
2017
).
16.
P.
Sturmfels
,
S.
Lundberg
, and
S.-I.
Lee
, “
Visualizing the impact of feature attribution baselines
,”
Distill
5
,
e22
(
2020
).
17.
P.
Hase
,
H.
Xie
, and
M.
Bansal
, “
The out-of-distribution problem in explainability and search methods for feature importance explanations
,”
Adv. Neural Inf. Process. Syst.
34
,
3650
3666
(
2021
).
18.
S.
Hooker
,
D.
Erhan
,
P.-J.
Kindermans
, and
B.
Kim
, “
A benchmark for interpretability methods in deep neural networks
,” arXiv:1806.10758 [cs.LG] (
2019
).
19.
C.
Meier
,
Synchrotron-Strahlung zur Visualisierung und Analyse Wellenlängenabhängiger Prozesseigenschaften Beim Laserstrahlschweißen von Kupfer
(
RWTH Aachen
,
Aachen
,
2022
).
You do not currently have access to this content.