An Investigation on the Effects of Optimum Forming Parameters in Hydromechanical Deep Drawing Process Using the Genetic Algorithm

The present research work is concerned with the effects of optimum process variables in elevated temperature hydro-mechanical deep drawing of 5052 aluminum alloy. Punch-workpiece and die-workpiece friction coefficients together with the initial gap between the blank holder and matrix were considered as the process variables which, in optimization terminology, are called design parameters. Since both the maximum reduction in sheet thickness and the final product uniformity (thickness variation) are important in the hydro-mechanical deep drawing, they are selected as the objective functions for optimization. After conducting 27 finite-element simulations of the operation and validation of the numerical results, a neural network was trained and combined with the genetic algorithm to obtain the optimum design parameters. The outcomes of this investigation have shown that these optimized process variables simultaneously resulted in the best values for both the objective functions, in comparison with all the conducted finite-element analyses.

Keywords:

Hydromechanical Deep Drawing , Maximum Reduction in Sheet Thickness , Product Uniformity , Neural Network , Genetic algorithm