عنوان مقاله [English]
Autofrettage is an effective method for increasing the load capacity of thick-walled cylinders. Since the behavior of materials in loading and unloading steps is different, it is necessary to provide a suitable constitutive model to accurately predict the behavior of the material and estimate the induced residual stresses. In this study, the Chaboche material model was used to numerically investigate the autofrettage process. The six constants of this model were determined using the cyclic behavior of material, based on the von Mises yield criteria for Al7075-T6 aluminum. The results were compared with the results obtained from the elastic perfectly plastic model for the material. Experimental tests were done in order to determine the accuracy of the results of each model in prediction of the residual stress distributions. The central hole drilling (CHD) method was used for determination of the residual stresses along the thickness direction. Several specimens were made with different aspect ratios (ratio of outer to inner radii of cylinder), and subjected to various internal pressures to obtain the desired values for autofrettage ratios (ratio of the plastic portion to total thickness of cylinder). Two aspect ratios; 1.4 and 1.5, and three values; 30\%, 45\% and 60\%, were considered for the autofrettage ratio. Then, stresses were measured in the outer region of the cylinder wall. The results show that the elastic perfectly plastic model predicts the residual stresses greater than experimental values, while the Chaboche model predicts smaller amounts. However, the results predicted by the Chaboche model are closer to the experimental data, more than the elastic perfectly plastic model. For both material models, the plastic region increases with increasing autofrettage pressure. The residual stress values increase by increasing the aspect ratio, and also, the autofrettage ratio. So, maximum residual stress will be obtained when all of the wall thickness becomes plastic (100\% autofrettage ratio).