THE EVALUATION OF DURABILITY OF STRUCTURAL ELEMENTS USING NEURAL NETWORKS

While in operation structural elements may have cracks, which usually grow up to critical size due to further cyclic loading, so that the component is likely to be destroyed. Therefore, it is important to study the process of fatigue failure of structural materials. The aim of this work was to evaluate the durability of structural elements during their design phase, taking into account the achieved level of material damage, and to predict the fatigue crack growth (FCG) rate in an aluminium D16chT alloy under regular loading using neural network (NN). Based on experimental data of FCG at stress ratio R = 0, FCG rate was predicted in this alloy. The results obtained by the authors are in good agreement with the experimental data.