Issue 48

M. Laredj et alii, Frattura ed Integrità Strutturale, 48 (2019) 193-207; DOI: 10.3221/IGF-ESIS.48.21 205 Response Predicted value numerical value Difference % CRSEH -110 -100 10% MCRS -248 -238 4.2% DMCRS 1.286611 1.21198 6.15% CRSZ 2.987402 2.78 7.46% MT 54.74 62 11.7 % PDZ 3.61 3.333 8.47 % Table 4 : Predicted and numerical values. Figure16 : predicted and numerical values of residual stress This small difference between the simulation and the predicted models confirms the excellent effectiveness of the mathematical models established by the experimental design method. C ONCLUSION n this work, the study focuses to establish and validate a predictive model of residual stresses generated from cold hardening. Thus the assessment allows industrialists to predicting the fatigue lifetime duration of riveted or bolted structures. An experimental setup was used to points out the effectiveness of degree of expansion, thickness of the part as well as the yield strength on the residual stresses profiles. The application of numerical with the experimental method allows determining the optimal values of the different factors to predict the residual stress profiles for any values of the factors. By proposing a mathematical model that takes into account these factors the latter allowed us to obtain a better description of the residual stress profile. The validation of the model confirms the excellent quality of the mathematical model established by the experimental design method. Based on our assessment, the effect of the degree of expansion on the improvement of the service lifetime is clearly significant than the yield strength and the thickness. Therefore, the classifications of impact parameters are in the following order: degree of expansion, the yield strength and the thickness of the structure. I