Accuracy of Evaluation Methodologies for Low-Cycle Fatigue Performance of Steel Beam-End Connections

In the 2011 Tohoku earthquake, long-duration ground motion had shaken numerous high-rise steel buildings in metropolitan areas of Japan. Such an earthquake, steel beam-end connections were subjected to a significant number of cyclic deformations which has been raising concerns about cumulative damage to the structures. To evaluate the deformation capacity of steel beam-end connections subjected to inelastic deformations, assessment of low-cycle fatigue performance which refers to the number of cycles to failure of beam-end connection has frequently been considered in seismic design. Kishiki et al. proposed an equation following the concept of Manson-Coffin law for predicting low-cycle fatigue life, which is indicated in terms of maximum rotation angle of steel beam-end connection and coefficient of moment transfer contribution of the beam web to beam member. On the other hand, Building Research Institute in Japan has similarly presented an evaluation method, in which maximum plastic ratio and coefficient of beam-end connection details are used to predict low-cycle fatigue life. However, the accuracy of evaluation methodologies presented by earlier research has not been adequately investigated. In this paper, an experimental database of steel beam-end connections subjected to constant cyclic loading amplitude was constructed to investigate the accuracy of evaluation methodologies for low-cycle fatigue performance. As a result, the method based on maximum rotation angle was able to evaluate low-cycle fatigue life of beam-end connections with error ranges between 0.5 to 2.0. On the other hand, the methods based on maximum plastic ratio were broadly evaluated for safety side of the experimental results.