8 Industry 4.0 in Welding
271
8.3.2.2 Arc Sound
The sound generated during arc welding has been found and reported to be containing
useful information relating to the molten metal, arc column, and the metal transfer
modes [58–61]. Conventionally, a welder recognizes the arc sound and typically gets
used to with the different sounds which occur with varying modes of metal transfer.
The primary source of sound is the arc itself, which is followed by the metal that gets
transferred from the electrodes to workpiece or at times sticking of the electrode to
the workpiece, and the spatter occurring during the welding. This marks importance
of the sound signature. Studies have reported analysing the arc sound signal in time
domain, where “kurtosis” has been found to be the feature having strong correlation
with the deposition efficiency [62, 63]. In the frequency domain analysis, using fast
Fourier transform (FFT), the arc sound was able to be differentiated between the
continuous and pulsed-mode of metal inert gas welding processes [62]. This signal
has also been utilized for monitoring of the defects in the welding [64]. Further study
on the utilization of the arc sound signal for monitoring can be found from Ref. [58].
The sound signature has also been utilized in monitoring of the RSW technique for
identifying the lifespan of the electrode and prediction of the joint strength [65].
A major drawback of this sensor is its utility in a machine shop floor where
machines run simultaneously. The sound which will be arising from other sources
will also get reflected over the sound signature of the welding process, and then,
pre-processing of the signal would become difficult.
8.3.2.3 Acoustic Emission Signal
Following the arc sound, acoustic emission (AE) signal has also been successful
in determining the weld quality. AE refers to the elastic transient waves which are
generated within a material undergoing plastic deformation or fracture. The signal
is known for its ability to detect the flaws and discontinuities within a material [66].
The intensity of the AE signal forms an indicator of these flaws. The signal carries
a practical approach for implementation, as it does not get affected by the range of
audible frequencies. However, the AE sensors are required to be mounted over the
job to capture the emissions, which pose a limitation in the case of mass production.
Further, as compared to the sonic signal, a number of AE sensors may be required to
be mounted over the job for capturing the AE signature, which is only a microphone
in case of sonic signal [65].
An earlier research reports the importance of AE signal for monitoring GTAW,
submerged arc welding and RSW techniques [67]. The AE signal successfully
captures the different phases of the spot welding technique, and thus, has been
utilized to extract information about the nugget quality [68]. Features such as “AE
count” and “positive peak” have been reported to have good correlation with the
joint strength. AE signals have been successfully utilized in monitoring the FSW
technique. This is because FSW involves plastic deformation of the workpieces to
be joined caused by a rotating tool which plunges inside. Researchers have captured
271
8.3.2.2 Arc Sound
The sound generated during arc welding has been found and reported to be containing
useful information relating to the molten metal, arc column, and the metal transfer
modes [58–61]. Conventionally, a welder recognizes the arc sound and typically gets
used to with the different sounds which occur with varying modes of metal transfer.
The primary source of sound is the arc itself, which is followed by the metal that gets
transferred from the electrodes to workpiece or at times sticking of the electrode to
the workpiece, and the spatter occurring during the welding. This marks importance
of the sound signature. Studies have reported analysing the arc sound signal in time
domain, where “kurtosis” has been found to be the feature having strong correlation
with the deposition efficiency [62, 63]. In the frequency domain analysis, using fast
Fourier transform (FFT), the arc sound was able to be differentiated between the
continuous and pulsed-mode of metal inert gas welding processes [62]. This signal
has also been utilized for monitoring of the defects in the welding [64]. Further study
on the utilization of the arc sound signal for monitoring can be found from Ref. [58].
The sound signature has also been utilized in monitoring of the RSW technique for
identifying the lifespan of the electrode and prediction of the joint strength [65].
A major drawback of this sensor is its utility in a machine shop floor where
machines run simultaneously. The sound which will be arising from other sources
will also get reflected over the sound signature of the welding process, and then,
pre-processing of the signal would become difficult.
8.3.2.3 Acoustic Emission Signal
Following the arc sound, acoustic emission (AE) signal has also been successful
in determining the weld quality. AE refers to the elastic transient waves which are
generated within a material undergoing plastic deformation or fracture. The signal
is known for its ability to detect the flaws and discontinuities within a material [66].
The intensity of the AE signal forms an indicator of these flaws. The signal carries
a practical approach for implementation, as it does not get affected by the range of
audible frequencies. However, the AE sensors are required to be mounted over the
job to capture the emissions, which pose a limitation in the case of mass production.
Further, as compared to the sonic signal, a number of AE sensors may be required to
be mounted over the job for capturing the AE signature, which is only a microphone
in case of sonic signal [65].
An earlier research reports the importance of AE signal for monitoring GTAW,
submerged arc welding and RSW techniques [67]. The AE signal successfully
captures the different phases of the spot welding technique, and thus, has been
utilized to extract information about the nugget quality [68]. Features such as “AE
count” and “positive peak” have been reported to have good correlation with the
joint strength. AE signals have been successfully utilized in monitoring the FSW
technique. This is because FSW involves plastic deformation of the workpieces to
be joined caused by a rotating tool which plunges inside. Researchers have captured