Seismic response and failure modes of steel silos with isotropic stepped walls: The effect of vertical component of ground motion and comparison of buckling resistances under seismic actions with those under wind or discharge loads.

• Dynamic buckling behaviors of three stepped wall steel silos were assessed by an IDA study. • Critical PGA, base shear and base moment at the buckling instant were calculated. • Normal pressures and frictional tractions formed on wall of the silos were presented. • The effects of vertical component of seismic records on the results were evaluated. • Seismic buckling capacities of silos were compared with those under wind or discharge loads. Steel silos are one of the main structures for bulk solids handling and storage in many industries and agricultural sectors. A growing body of research suggests the importance of including the vertical component of earthquake ground motion in seismic analysis and design of certain structures. In the case of steel silos, this may induce additional meridional compression to exacerbate buckling failures in such thin shells. Accordingly, this paper investigates the buckling behavior of three cylindrical steel silos (i.e., a squat, an intermediate slender and a slender silo) with stepped walls subjected to horizontal only (H) and horizontal and vertical (HV) ground accelerations to address this point. Seven pairs of seismic records, each scaled to multiple levels of the peak ground acceleration (PGA), were applied to the silos. Using Incremental Dynamic Analyses (IDA), the critical base shear, base moment and the PGA at the buckling instant were evaluated for each structure. Moreover, the normal pressures and frictional tractions induced from ensiled material on silo walls were extracted under seismic conditions to reveal the influence of including the vertical component of seismic records on contact pressures. In accordance to the analyses performed, two different buckling modes were observed under seismic actions. Squat and intermediate slender silos buckled in the elastic range characterized by local diagonal shear wrinkles in the upper parts of the shell walls, while slender silo showed an elastic-plastic elephant's foot buckling mode at the base. Considering the obtained critical values, the observed dynamic buckling modes and the extracted pressure distributions, it is concluded that including the vertical component of seismic records in calculations has quite marginal effects on seismic response of the silos. In order to provide a helpful insight into governing action in silos design, as a supplementary section to the paper, the buckling capacities of sample silos under seismic actions were compared with those evaluated under wind actions or discharge loads. Regarding to the different seismicity levels of Europe, governing action was determined for each silo in the context of Eurocode limit state design. The results obtained highlight the importance of considering the seismic action in buckling design of steel silos. [ABSTRACT FROM AUTHOR]