Abstract
Under strong earthquakes, steel structures are prone to undergoing ultra-low cycle fatigue (ULCF) fracture after sustaining cyclic large-strain loading, leading to severe earthquake-induced damage. Thus, establishing a prediction method for ULCF plays a significant role in the seismic design of steel structures. However, a simple and feasible model for predicting the ULCF life of steel structures has not been recognized yet. Among existing models, the ductile fracture model based on ductility capacity consumption has the advantage of strong adaptability, while the loading history effect in the damage process can also be considered. Nevertheless, such models have too many parameters and are inconvenient for calibration and application. To this end, focusing on the prediction methods for ULCF damage in steel structures, with the fragile parts being in moderate and high stress triaxiality, this paper proposes a simplified uncoupled prediction model that considers the effect of stress triaxiality on damage and introduces a new historical-effect related variable function reducing the calibration work of model parameters. Finally, cyclic loading test results of circular notched specimens verify that the proposed model has the advantages of a small dispersion of parameters for calibration, being handy for application, and possessing reliable results, providing a prediction method for ULCF damage of structural steels.
Highlights
Steel structures are widely used in engineering structures due to their high strength, light weight, and good ductility; under strong earthquakes, steel structures will encounter fracture damage due to the crack initiation of ultra-low cycle fatigue (ULCF) [1,2]
The ULCF of structural steel occurs under cyclic large plastic strain loading, and the corresponding fatigue life can generally be considered to be less than 20 cycles [3,4], which is largely different from low cycle fatigue (LCF) and high cycle fatigue
While comparing the proposed ductile fracture model with the linear damage model and the cyclic void growth model (CVGM), we find that the proposed simplified model in this paper has the advantages of possessing good prediction accuracy and being more convenient to calibrate model parameters
Summary
Steel structures are widely used in engineering structures due to their high strength, light weight, and good ductility; under strong earthquakes, steel structures will encounter fracture damage due to the crack initiation of ultra-low cycle fatigue (ULCF) [1,2]. Many researchers have conducted considerable theoretical studies and experiments on ULCF and proposed some ULCF prediction methods, such as the Gurson-TvergaardNeedleman (GTN) model [7,8], continuous damage mechanics (CDM) model [9,10], the Coffin–Masson formula [11,12], the cyclic void growth model (CVGM) [13,14], and the ductile fracture model considering the loading history effect [15,16] These methods can be divided into coupled and uncoupled models [17] according to whether the effect of damage on constitutive properties of materials during the loading process is considered or not. While comparing the proposed ductile fracture model with the linear damage model and the CVGM, we find that the proposed simplified model in this paper has the advantages of possessing good prediction accuracy and being more convenient to calibrate model parameters
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
