Tsarpalis D., Vamvatsikos D., Vayas I. (2019). Simplified models for the nonlinear analysis of ARSW structures under seismic loading. Proceedings of the COMPDYN2019 Conference on Computational Methods in Structural Dynamics and Earthquake Engineering, Crete, Greece
Abstract | Automated Rack Supported Warehouses (ARSW) are the state of the art in storage technology, as they provide substantial savings in terms of cost, space and energy with respect to traditional solutions. Despite their lightness, ARSWs carry very high live loads, by far higher than their self-weight, in contrast to what happens in typical civil engineering structures. Thus, standard design approaches are not applicable, especially when one considers lateral loading, i.e. seismic and wind loading.
In the frame of the STEELWAR project, the behavior factor (q) as well as the seismic fragility shall be assessed for a number of archetype warehouses. FEM modelling for such structures is a tedious task; they consist of hundreds or thousands steel members and nodes connected to each other through simple and semirigid joints. Modern computers accompanied with efficient computational algorithms can handle linear systems with ease and thus, linear analysis can be performed by including all structural components in the analysis model. Problems arise when one considers nonlinear phenomena i.e. material and geometric nonlinearity. Simulations that take into account all ARSW members and their nonlinear response may lead to prohibitive computational costs, while introducing convergence and numerical stability problems. As a direct remedy, a reduced-order physical model is proposed that enables accurate assessment of nonlinear behavior without compromising convergence performance.