8 Industry 4.0 in Welding
279
The application of cloud computing in welding has been demonstrated in few
articles. One of them utilizes a vision camera in arc welding technique for capturing
the weld images in real time and transfers them to a cloud server for classification
of those welds [101]. The acquired images were processed using the “OpenCV”
platform which is another example of an open-source platform. The cloud server had
the ML model trained for classifying the weld either into defective or defect-free.
The potential of this work lies in the use of low-cost resources including computation
in cloud for real-time monitoring of the welded samples. Another article shows the
usage of cloud computing in FSW technique [54]. In this work, the cloud receives
data from multiple sensors in real time and feeds them to ML models for online
prediction and control of the weld quality. The usage of cloud eliminated the use of
hardware for processing the signals.
8.3.5.3 Digital Twin, Virtual Reality, and Augmented Reality
These tools mark the importance of virtualization principle which aims at creating
virtual model of physical system [102]. The digital twin is a replica of a physical
counterpart existing in a digital world. The digital twin gathers information from the
physical part in real time and applies analytics to make useful predictions. This tool
will be useful for planning the welding process. For instance, the weldability of materials along with suitable parameters can be determined through welding simulations
prior to actual welding. This might be useful for determining suitable conditions for
joining dissimilar materials with significant differences in the physical properties.
This may prove beneficial for automobile industry where welding is one of the major
jobs. This sector is constantly exploring newer and advanced materials for manufacturing of the automobile. As such, it becomes crucial to determine weldability of
these materials. The virtual reality (VR) allows users to experience a 360° enclosed
view of the working environment, and the augmented reality (AR) enhances reality
by adding digital information to it. Both VR and AR enable a cooperation between
a user and a virtual world [103]. With these technologies, the design can be tested
in a virtual environment, and the errors can be identified and rectified at the design
level. These tools seem to be promising for imparting training to novice welders.
This will help them understand the physics of the process, such as proper positioning
and movement of the welding torch, importance of joint preparation and importance
of shielding the molten pool in an interactive way. This aims to improve the quality
of welding and productivity.
A digital twin has been proposed for FSW machine where the twin in real time
gathers useful information from the physical FSW machine [1]. The twin evaluates
the health of the machine by determining the error between the instructed and actual
velocities of the spindle and feed motors in real time, classifies the health of the tool
by checking the force and power variation and tracks the temperature of the hydraulic
oil, oil flow rate, and oil turbidity. Similarly, a digital twin has also been proposed for
spot welding which includes virtual simulations for weld designing, optimization,
planning, production and evaluation [104].
279
The application of cloud computing in welding has been demonstrated in few
articles. One of them utilizes a vision camera in arc welding technique for capturing
the weld images in real time and transfers them to a cloud server for classification
of those welds [101]. The acquired images were processed using the “OpenCV”
platform which is another example of an open-source platform. The cloud server had
the ML model trained for classifying the weld either into defective or defect-free.
The potential of this work lies in the use of low-cost resources including computation
in cloud for real-time monitoring of the welded samples. Another article shows the
usage of cloud computing in FSW technique [54]. In this work, the cloud receives
data from multiple sensors in real time and feeds them to ML models for online
prediction and control of the weld quality. The usage of cloud eliminated the use of
hardware for processing the signals.
8.3.5.3 Digital Twin, Virtual Reality, and Augmented Reality
These tools mark the importance of virtualization principle which aims at creating
virtual model of physical system [102]. The digital twin is a replica of a physical
counterpart existing in a digital world. The digital twin gathers information from the
physical part in real time and applies analytics to make useful predictions. This tool
will be useful for planning the welding process. For instance, the weldability of materials along with suitable parameters can be determined through welding simulations
prior to actual welding. This might be useful for determining suitable conditions for
joining dissimilar materials with significant differences in the physical properties.
This may prove beneficial for automobile industry where welding is one of the major
jobs. This sector is constantly exploring newer and advanced materials for manufacturing of the automobile. As such, it becomes crucial to determine weldability of
these materials. The virtual reality (VR) allows users to experience a 360° enclosed
view of the working environment, and the augmented reality (AR) enhances reality
by adding digital information to it. Both VR and AR enable a cooperation between
a user and a virtual world [103]. With these technologies, the design can be tested
in a virtual environment, and the errors can be identified and rectified at the design
level. These tools seem to be promising for imparting training to novice welders.
This will help them understand the physics of the process, such as proper positioning
and movement of the welding torch, importance of joint preparation and importance
of shielding the molten pool in an interactive way. This aims to improve the quality
of welding and productivity.
A digital twin has been proposed for FSW machine where the twin in real time
gathers useful information from the physical FSW machine [1]. The twin evaluates
the health of the machine by determining the error between the instructed and actual
velocities of the spindle and feed motors in real time, classifies the health of the tool
by checking the force and power variation and tracks the temperature of the hydraulic
oil, oil flow rate, and oil turbidity. Similarly, a digital twin has also been proposed for
spot welding which includes virtual simulations for weld designing, optimization,
planning, production and evaluation [104].