Resumen de Tribiological investigation and optimization of friction stir spot welding of dissimilar metals by LSSM-ANN method
Ajit Kumar, Kamal Pal, Debadutta Mishra
In manufacturing firms, welding technology plays a supreme role in joining purposes. The joining of dissimilar metals is complex in several conventional type welding. Friction stir spot welding (FSSW) is a solid-state welding technique and it is a derivative of conventional friction stir welding (FSW) to join the metals. The recent demands of advanced engineering materials such as aluminum alloys and stainless steel in the automobile sectors are forced to facilitate this research work, at the same time, the joining of such materials is quite complex because of its different welding characteristics. The main contribution of this work is to optimize and predict the microhardness and tensile failure properties of dissimilar metals by hybrid lightning search algorithm-simplex method (LSA-SM) and ANN algorithm.
The dissimilarly joined metals namely, duplex stainless steel 32760 and aluminum 7075-T6 are utilized in this work and the experimentation is performed for six different test runs. Besides, the predicted values from the proposed algorithm are compared with the existing support vector machine (SVM) method. The predicted values of 314 Hv microhardness and 4967N tensile shear failure load are optimized by the ANN-LSA-SM method at 1900 rpm. From the test results, it shows that the proposed model is efficient to predict the mechanical properties of the dissimilar welded materials than the experimental and SVM method.
