288
D. Mishra et al.
Fig. 8.14 Surface plot for the obtained tensile strength values
The discussions above provide an understanding of the behaviour of the process
signals in FSW. In order to predict the weld quality, a predictive model was essential
to build. This would lead to the fulfilment of the decentralization principle. It refers to
the decisions taken by analysing certain data acquired from a manufacturing process.
The decision here also refers to individual decisions at multiple devices. In this case
study, two devices are involved: one is the physical system (FSW machine), and the
other is the cyber component (cloud server). The tasks performed at the physical part
were: (a) welding, (b) acquisition of the signal and (c) transmission of the acquired
signal. The cyber part is held responsible for: (a) analysis of the data, (b) extraction
of useful information and (c) feedback to the physical part.
For the predictive model, several experiments were needed to be performed. These
experiments were performed in the machine, depicted in Fig. 8.11. Several combinations of joining parameters were selected as elaborated in the WPS for FSW.
Workpieces of AA6061 (an alloy of aluminium) were butt welded. The reasons for
selecting this alloy are its versatile applications in various sectors of manufacturing
[115, 116]. This is because of its properties such as high strength, low density, high
electrical and thermal conductivity, and resistance to corrosion. [117].
In order to assess the welded joints (i.e. PQR of FSW joints) obtained with the
selected combinations, ultimate tensile strength of the welds was determined. The
tensile strength is one of the vital mechanical properties, and it refers to the maximum
load a welded joint can bear. The tensile strength was determined in a universal tensile
D. Mishra et al.
Fig. 8.14 Surface plot for the obtained tensile strength values
The discussions above provide an understanding of the behaviour of the process
signals in FSW. In order to predict the weld quality, a predictive model was essential
to build. This would lead to the fulfilment of the decentralization principle. It refers to
the decisions taken by analysing certain data acquired from a manufacturing process.
The decision here also refers to individual decisions at multiple devices. In this case
study, two devices are involved: one is the physical system (FSW machine), and the
other is the cyber component (cloud server). The tasks performed at the physical part
were: (a) welding, (b) acquisition of the signal and (c) transmission of the acquired
signal. The cyber part is held responsible for: (a) analysis of the data, (b) extraction
of useful information and (c) feedback to the physical part.
For the predictive model, several experiments were needed to be performed. These
experiments were performed in the machine, depicted in Fig. 8.11. Several combinations of joining parameters were selected as elaborated in the WPS for FSW.
Workpieces of AA6061 (an alloy of aluminium) were butt welded. The reasons for
selecting this alloy are its versatile applications in various sectors of manufacturing
[115, 116]. This is because of its properties such as high strength, low density, high
electrical and thermal conductivity, and resistance to corrosion. [117].
In order to assess the welded joints (i.e. PQR of FSW joints) obtained with the
selected combinations, ultimate tensile strength of the welds was determined. The
tensile strength is one of the vital mechanical properties, and it refers to the maximum
load a welded joint can bear. The tensile strength was determined in a universal tensile