Artificial neural network guided prediction of surface modification and mechanical behavior in Al 6061-TiC reinforced TIG welds under controlled heat input

Abstract

Aluminium 6061 alloy is widely used in aerospace and automotive applications owing to its light weight, corrosion resistance, and machinability. Nevertheless, the poor hardness and limited surface integrity of Al 6061 alloys restrict its prolonged use under extreme conditions. In this work, the properties of Al 6061 alloy has been augmented by developing hard Titanium Carbide (TiC) coatings on its surface by tungsten inert gas (TIG) cladding technique. The developed TiC coatings enhanced the hardness of Al 6061 alloy up to 3 times, with average micro-hardness in the range of 431- 496 HV_0.05. From this motivation, on the later stage of the work, titanium carbide particles are reinforced in to the weld joints of Al 6061 alloys by TIG arc heat under variable TIG current conditions. The integration of TiC reinforcement into the weld joints enhanced the ultimate tensile strength from 20–29 MPa to 59–83 MPa. Additionally, the maximum hardness attained as 173.72 HV_0.05, depicting an improvement upto 1.5 times. Further, an Artificial Neural Network (ANN) model using backpropagation accurately predicted mechanical properties, showing strong agreement with experimental data and validating its forecasting capability. It achieved a high co-efficient of determination (R² ≈ 0.99) and depicted minimal prediction error, signifying a strong correlation between the experimental data and the predicted outcomes.