Probabilistic evaluation of effect of column stiffness on seismic performance of steel plate shear walls

The lateral force resistance and hysteretic energy dissipation capacity of steel plate shear walls (SPSWs) are primarily provided through yielding of the infill steel plates attached to the boundary frame of the system. As such, it is desirable to achieve uniform distribution of infill plate yielding at every story of an SPSW system during an earthquake event. This paper investigates the effect of column stiffness on infill plate yielding distributions in representative two-story and six-story SPSWs. Specially considered in the investigation are the uncertainties existing in infill plate strength and lateral force distributions along the vertical direction. Three probabilistic evaluation methods including the Monte Carlo method, the Latin Hypercube Sampling method, and the Rosenblueth’s 2K + 1 Point Estimate method, are considered. It is shown that all methods are stable and effective for this investigation; however, the Rosenblueth’s 2K + 1 Point Estimate method requires the least computational cost. The results from the probabilistic evaluation show that increasing column stiffness and story ductility capacity can be effective to achieve infill plate yielding at each story of an SPSW. However, the minimum column stiffness required in the codes in North America does not necessarily lead to infill plate yielding at every story of a multi-story SPSW.

► Increasing column stiffness helps achieve plate yielding at each story of an SPSW. ► Increasing story ductility helps achieve plate yielding at each story of an SPSW. ► Current design codes do not ensure plate yielding at each story of an SPSW. ► The Monte Carlo, Latin hypercube, and Rosenblueth’s 2K + 1 methods were compared. ► The Rosenblueth’s 2K + 1 method is the most efficient in SPSW performance evaluation.