Developing a Mathematical Model to Predict the Optimum Friction Phase Parameters for Friction Welding of High Speed Steel to Medium Carbon Steel

This work was carried out in order to optimize the friction phase parameters, of friction welding of M2 high speed steel, to AISI 1040 medium carbon steel, namely; rotational speed, friction pressure and friction time. The experiments were designed as per Taguchi method. The optimization of the experimentally obtained results was carried out by trying three mathematical models, namely; a multiple linear regression model without interaction effect, a multiple linear regression model with interactions effect, and a second-order polynomial regression model. The three models were evaluated using the experimental data, the coefficient of multiple determination R 2 , and Standard error of the regression (S), were used as the evaluation criteria of the models. The polynomial model was chosen, and optimized using a Genetic Algorithm. The optimal value of the joint strength of 411 MPa was obtained at the highest value of the time ( 44.9 sec.) and the pressure of 112 MPa and the speed of 1349 r.p.m.