Nonlinear time history analysis (NTHA) is widely used to evaluate wind-induced responses of tall buildings. In ETABS, NTHA can be conducted using either the Direct Integration Method (DIM) or Fast Nonlinear Analysis (FNA). Through linear and nonlinear time history analyses of a 50-story building, this study demonstrates that FNA does not fully capture P-delta effects under full loading conditions and tends to underestimate nonlinear hinge behavior, with the latter identified as the primary source of discrepancy. Consequently, displacement responses predicted by DIM, where hysteretic energy dissipation is more accurately represented, were on average approximately 27% smaller than those obtained from FNA. The results indicate that the differences between the two approaches stem not from simple numerical variation, but from fundamental distinctions in how nonlinear hinge behavior and P-delta effects are modeled, which are quantitatively examined herein. To reduce this discrepancy, a calibration framework is proposed that maps the displacement response at a representative location of a tall building onto an equivalent reduced-order system. The method first estimates the linear and nonlinear parameters of a two-degree-of-freedom model and then incrementally incorporates nonlinear hinge behavior and P-delta effects to adjust FNA predictions toward DIM-level accuracy. The proposed procedure significantly improves displacement agreement, enhancing calibration efficiency by up to 30.55% for across-wind responses and 28.84% for along-wind responses. The most pronounced improvements are observed in across-wind and upper-story displacements.
Kim et al. (Sun,) studied this question.