| Abstract Scope |
The welding fabrication must keep in mind the importance of the Industry 4.0 philosophy and, consequently, the digital transformation required for the modern manufacturing. Not only in order to not lose competitiveness, but also due to more specific requirements of standards, regarding the monitoring and documentation of the process. It is known that welding is a special process (ISO 9001, 2015), Cls. 7.5.2, which requires constant monitoring by specialists (technical responsibility via ISO 14731, 2019) with quality requirements (ISO 3834, 2005) and specific demand for monitoring and process documentation aiming at improvements, traceability, auditorship, corrective actions, and staff review (ISO 17662, 2016). In the Brazilian welding sector, Item 4.5.4 of N-133 since 2017 underlines this demand. In this scenario, the fourth industrial revolution offers major impacts to the industry's businesses, among them, the quality verification and product manufacturing tracking through data monitoring. It is not a matter of looking for specific faults, but of using data that can alert the inspector or automated systems, for example, when the weld bead is outside its normal manufacturing parameters. Therefore, the objective of this work is to present the development of a computer program (software) that generates different analyses after welding and makes it possible to compare the values of welding energy required by the welding procedure specification (WPS), even for manual or semiautomatic arc welding processes, by current, voltage and travel speed measurement with developed instrumentation. In addition, encrypted files are provided by the software in order to simplify with security the storage of this information on database servers in the context of Industry 4.0 (for instance, Amazon Web Service, Microsoft Azure, Google Cloud, among others). The software for the supervision of the manufacture of the welded joint was implemented in C Sharp (C #) language. It was used to develop the HMI (human-machine interface) and, during welding, acquires and processes data from the acquisition system and signal conditioners developed by the Laprosolda/UFU research Group every one second. The software validation was performed, firstly, by comparing the response signals of the implemented software with a reference one, at the 5 kHz acquisition rate, during the manufacture of weld beads. Subsequently, bench tests were carried out and weld beads were performed using the covered electrode (stick) process to validate the developed software. After welding, the platform provides the welding energy and travel speed graphs delimiting the permitted values and a text file containing all the data at the requested acquisition rate. For final validation of the software, manual welding was performed using the covered electrode process. As an example, the values read from the WPS and typed in the HMI were: Current = 102 A; Travel Speed = 2.97 ± 20% mm/s and Reference Welding Energy = 703 ± 20% J/mm. During the assessment of the system, the arc was deliberately extinguished in two points of the workpiece. In the first extinction, in which the welder raised the torch too much, the system indicated to lower the electrode. In the second extinction, the welder lowered the electrode holder so much that it almost stuck on the plate, at which point the system indicated that it would lift the electrode. The travel speed fluctuated around the reference (2.97 ± 20% mm/s). It is noticed that in the extinctions the value of the travel speed tended to zero and the system indicated to increase. This test showed that the software correctly responds to the arc height variation and the travel speed. The average current measured was 91 A, the voltage was 21.9 V, the travel speed was 3.17 mm/s and the welding energy was 1622 J / mm (there was extinction of the arc and saturation of values). It is possible to assert that these results have the potential to be included, online or offline, in inspection reports and/or procedure qualification records (PQR), evaluate the efficiency o
f the welder and indicate possible locations of discontinuities along the weld bead, among others. It can be concluded, by means of statistical analysis and of different waveforms, that the implemented software is suitable for use in an operational environment to assist in tracking the fabrication of the welded joint in manual, mechanized or automatic welding processes, based on the data requested by an WPS. Finally, the developed software can be used for the development of incremental technologies and for the digital transformation of the welding manufacturing sector. Keywords: Data monitoring; Documentation; Registry; Analytics; Big Data; Data Cloud. |