Optimizing Micro Friction Stir Spot Welding (mFSSW) of Aluminum Alloy AA1100 Using Neural Network Model

Abstract:

This study explores the optimization of Micro Friction Stir Spot Welding (mFSSW) by investigating the influence of tool profiles on welding outcomes, using aluminum alloy AA1100 with a 0.42 mm thickness as the specimen material. Monitoring temperature and RPM during welding with thermocouples and tachometers, mechanical properties are assessed through tensile shear tests, microhardness measurements, and macrostructural observations. The findings serve as the basis for developing Neural Network models using Rapidminer software, marking a transformative development that positions Neural Networks as potent tools for optimizing welding processes, potentially leading to achieving optimal weld quality. The investigation also delves into three welding tool configurations – the two-stage pin, one-stage pin, and pinless mFSSW probes – highlighting their distinct impacts on tensile shear test values and overall welding quality. Notably, the two-stage pin configuration emphasizes the significance of larger pin diameters and controlled heat generation for enhanced weld strength, while the one-stage pin configuration underscores the pivotal role of pin diameter and elevated temperatures in improving weld quality. The pinless mFSSW probe configuration, on the other hand, emphasizes the importance of shoulder diameter and temperature control for superior tensile shear test results. Leveraging Neural Network modeling for optimization, this study advances our understanding of parameter interactions and underscores the efficacy of Neural Networks in achieving superior tensile shear test values and welding quality in mFSSW, offering valuable insights for future endeavors in the field..

• Periodical: Materials Science Forum (Volume 1122)
• Pages: 35-42
• DOI: https://doi.org/10.4028/p-BbosI6
• Online since: May 2024
• Authors: Tri Haryanto Soleh Atmaja*, Laksita Aji Safitri, Pathya Rupajati, Ario Sunar Baskoro
• Keywords: Micro Friction Stir Spot Welding, Neural Network, Tool Profiles, Welding Optimization

References

[1] R.S. Mishra, Z.Y. Ma, Friction stir welding and processing, Mater. Sci. Eng. R. 50 (1-2) (2005) 1-78.

[2] N. Khan, S. Rathee, M. Srivastava, C. Sharma, Effect of tool rotational speed on weld quality of friction stir welded AA6061 alloys, Mater Today: Proc., 47(19) (2021) 7203-7207.

DOI: 10.1016/j.matpr.2021.07.496

[3] A.S. Baskoro, S. Hadisiswojo, G. Kiswanto, Winarto, M.A. Amat, Z.W. Chen, Influence of Welding Parameter on Macrostructural and Thermomechanical Properties in Micro Friction Stir Welded Under High Speed Rotation, The Int. J. Adv. Manuf. Technol., 106 (1) (2019) 163-175.

DOI: 10.1007/s00170-019-04490-8

[4] H. Atharifar, Optimum parameters design for friction stir spot welding using a genetically optimized neural network system. Proc IMechE, Part B: J Engineering Manufacture 224(3) (2010) 403–418.

DOI: 10.1243/09544054jem1467

[5] S. Lathabai, M.J. Painter, G.M.D. Cantin, V.K. Tyagi, Friction Spot Joining of an Extruded Al–Mg–Si alloy. Scr. Mater. 55(10) (2006) 899-902.

DOI: 10.1016/j.scriptamat.2006.07.046

[6] L.A. Safitri, I.F. Fakhrurrozi, M.A. Amat, A.S. Baskoro, P. Rupajati, G. Kiswanto, Effect of Tool Profile on Mechanical Properties, Temperature, and RPM Using Micro Friction Stir Spot Welding (mFSSW) on Aluminum Alloy AA1100: submitted to IOP Proceeding (2023)

DOI: 10.15587/1729-4061.2023.276484

[7] ISO 14273, Resistance Welding – Destructive Testing of Welds – Specimen Dimensions and Procedure for Tensile Shear Testing Resistance Spot and Embossed Projection Welds, ISO, (2016).

DOI: 10.3403/30266453

[8] M. Ilham, Suparjon, Y. Sanjiwani, Optimization of Cutting Speed and Feed Rate on Surface Roughness and Vibration using Taguchi Method: A Review, MECHTA Int. J. of Mech. Eng. Tech. and Applic. 1 (2) (2020) 43-49

DOI: 10.21776/mechta.2020.001.02.2