272
D. Mishra et al.
the AE signal for identification and localization of the void defect [69]. The analysis
has been performed in the time–frequency domain, in which the “energies” of the
wavelet coefficients were extracted to classify the welds. Further, the AE signal was
utilized to analyse the effect of changing tool pin profiles and process parameters
in FSW [70–72]. The signal processing techniques utilized in these studies included
the analysis of the signal in time domain, and techniques such as Fast Fourier Transform (FFT), Short-Time Fourier Transform (STFT), and Discrete Wavelet Transform
(DWT). The amplitude of the AE signal was found to have a good correlation with
the varying parametric conditions [70].
The AE signal represents high-frequency phenomenon, and thus, has high signalto-noise ratio. Because of the high-frequency nature, it carries a lot of information
about the transient events occurring in manufacturing. However, the limitation is
with the fixation of these sensors as they are of contact-type, and handling of the
huge data.
8.3.2.4 Temperature Signal
The transfer of metal from electrode to workpiece in arc welding technique occurs
in different modes. These modes have different characteristics, and the rate of metal
transfer depends upon the value of electrical parameters as well. Thus, the change in
the temperature during these modes is obvious. Instruments like the thermographic
camera have been utilized for acquiring the temperature signal during arc welding
technique [73]. In order to derive useful information about the welding process,
certain disturbances were introduced, and the corresponding temperature signals
were investigated. The penetration depth has been determined by studying the mean
value of the temperature signal [74]. Likewise, the misalignment and irregularities
in the surface have been detected by studying the variations in the signal [74]. A few
welding defects have also been monitored by analysing this signal. The penetration
depth has been controlled by analysing the radii of the surface isotherms [75], and
size, area enclosed by the surface isotherm [76]. In spot welding technique, temperature signal has been acquired by using an infrared sensor [77]. A good correlation
in between the temperature and current signatures has been found in this welding
technique. Features from multiple signals, namely temperature, current, voltage,
ultrasonic, force and dynamic resistance, were extracted and fed to a ML model for
predicting the weld quality.
Infrared sensing is useful in addressing certain problems as it is non-invasive in
nature, and the signal can be acquired with respect to point, line or area. However,
the sensor requires a line of sight, which if interrupted, creates a disturbance in the
measurements. The interruption could be caused because of introduction of other
objects. The limited distance from the targeted area for fixing this sensor is another
limitation.
The temperature signature has been useful for monitoring of FSW technique as
well. This is because of the frictional interaction occurring between the tool and
the workpieces. The stir zone has been found having the maximum temperature
D. Mishra et al.
the AE signal for identification and localization of the void defect [69]. The analysis
has been performed in the time–frequency domain, in which the “energies” of the
wavelet coefficients were extracted to classify the welds. Further, the AE signal was
utilized to analyse the effect of changing tool pin profiles and process parameters
in FSW [70–72]. The signal processing techniques utilized in these studies included
the analysis of the signal in time domain, and techniques such as Fast Fourier Transform (FFT), Short-Time Fourier Transform (STFT), and Discrete Wavelet Transform
(DWT). The amplitude of the AE signal was found to have a good correlation with
the varying parametric conditions [70].
The AE signal represents high-frequency phenomenon, and thus, has high signalto-noise ratio. Because of the high-frequency nature, it carries a lot of information
about the transient events occurring in manufacturing. However, the limitation is
with the fixation of these sensors as they are of contact-type, and handling of the
huge data.
8.3.2.4 Temperature Signal
The transfer of metal from electrode to workpiece in arc welding technique occurs
in different modes. These modes have different characteristics, and the rate of metal
transfer depends upon the value of electrical parameters as well. Thus, the change in
the temperature during these modes is obvious. Instruments like the thermographic
camera have been utilized for acquiring the temperature signal during arc welding
technique [73]. In order to derive useful information about the welding process,
certain disturbances were introduced, and the corresponding temperature signals
were investigated. The penetration depth has been determined by studying the mean
value of the temperature signal [74]. Likewise, the misalignment and irregularities
in the surface have been detected by studying the variations in the signal [74]. A few
welding defects have also been monitored by analysing this signal. The penetration
depth has been controlled by analysing the radii of the surface isotherms [75], and
size, area enclosed by the surface isotherm [76]. In spot welding technique, temperature signal has been acquired by using an infrared sensor [77]. A good correlation
in between the temperature and current signatures has been found in this welding
technique. Features from multiple signals, namely temperature, current, voltage,
ultrasonic, force and dynamic resistance, were extracted and fed to a ML model for
predicting the weld quality.
Infrared sensing is useful in addressing certain problems as it is non-invasive in
nature, and the signal can be acquired with respect to point, line or area. However,
the sensor requires a line of sight, which if interrupted, creates a disturbance in the
measurements. The interruption could be caused because of introduction of other
objects. The limited distance from the targeted area for fixing this sensor is another
limitation.
The temperature signature has been useful for monitoring of FSW technique as
well. This is because of the frictional interaction occurring between the tool and
the workpieces. The stir zone has been found having the maximum temperature