A modified Iwan model for bolt loosening monitoring in engineering structures: Contact evolution and preload relaxation behavior

Bolted joints structures undergo stiffness degradation and preload relaxation under complex loads. This affects structural reliability and safety. Accurately predicting the operational performance of bolts and characterizing their mechanical properties have become key research issues for ensuring the safety of complex equipment. Thus, predicting bolt performance and status through theoretical models has become an important research focus. The Iwan model is widely used to predict stiffness degradation and energy dissipation in bolted joints structures. However, the unreasonable assumptions of its core density function and the ambiguity in the physical meaning of its parameters have constrained its application. This study takes the high-strength nickel steel plate connection structure joined by M8 bolts as the subject of analysis. Based on the modified linear variable pressure distribution function, a modified Iwan model accounting for contact mechanisms is proposed. Furthermore, the preload relaxation function for contact is improved, with its prediction performance enhanced by 8.9 % compared with existing functions. Subsequently, the improved preload relaxation function is integrated into the modified Iwan model. Thereby establishing, for the first time, an improved Iwan model capable of dynamically depicting preload relaxation and contact mechanisms. Comparative verification between theory and simulation was conducted using finite element software. The results show that the Pearson correlation coefficient of the backbone curve of the modified model considering contact evolution reaches 0.961, which improves the model's prediction accuracy. Hysteresis loops of the Iwan model under relaxation conditions can characterize energy dissipation, and the inward contraction of these loops indicates bolt preload relaxation. The error of the dynamic parameters calculated by the finite element method and the improved Iwan model does not exceed 6.1 %, with a correlation coefficient reaching 0.989. Moreover, the prediction effect is improved by 27.27 % compared with the classical model, verifying the accuracy of the model's predictions. This work has solved the representation defects of existing theoretical models by examining the evolution of contact states and the relaxation of preloads. It has laid a theoretical foundation for solving the safety problems of bolted connections in steel structures and promoting structural health monitoring.