EFFECTS OF DIFFERENT OPTIMIZATION METHODS ON THE PREDICTIONS OF YLD2000 YIELD CRITERION COEFFICIENTS

The improved yield criteria are generally used in the finite element simulations of plastic deformation processes. Calculation accuracies of these criteria coefficients result successful simulation outcomes. In this study, the coefficients of the YLD2000 yield criterion are calculated by three most widely used optimization methods in literature, namely the least squares, nonlinear conditional optimization, and genetic algorithm methods. Two different aluminum alloys, AA7003-T6 and AA6063-T6 are selected to verify the prediction results. Results reveal that the nonlinear conditional optimization and genetic algorithm methods are very dependent on the initial values. Therefore, different result is determined for each different case. For this reason, it has been concluded that the least squares method should be preferred to calculate the coefficients of the yield criterion by using optimizing method