ABSTRACT
These directions, denoted principal directions, are used for the determination of their critical orientation, i.e. the orientation that yields the maximum value of each response quantity of interest. This determination can be achieved by application of the response spectrum method. Thus, (Smeby and Der Kiureghian (1985)), assuming that the input is a wide-band stationary process and using random vibration theory, calculated the critical incident angle for the case of two horizontal seismic components with identical spectral shapes, as well as the spectral moments of response for the case of different spectral shapes. Magliulo et al. (2014), evaluated the elastic and in-elastic response quantities under varying orientations of ground motion and proved the significant influence of incident angle of ground motion. López and Torres (1997), using the response spectrum method, calculated the critical incident angle and the associated maximum structural response for the general case of three ground motion components that may have different or identical spectral shapes. Menun and Der Kiureghian (1998), presented the CQC3 rule for the determination of the critical incident angle and the corresponding maximum response. (Lopez et al. (2000), López et al. (2001)) proved that the critical value for a single response quantity can be up to 20% larger than the usual response produced when the seismic components are applied along the structural axes. Finally in (Menun and Kiureghian
1 INTRODUCTION
Several researchers have investigated the influence of seismic incident angle on elastic as well as inelastic structural response. Analytical formulae for the determination of the critical angle of seismic incidence and the corresponding maximum structural response subjected to three correlated components have been developed Athanatopoulou (2005). Athanatopoulou concluded that the critical angle corresponding to peak response over all angles varies not only with the ground motion pair under consideration, but also with the response quantity of interest. These findings are confirmed in Kalkan and Kwong (2013) where the impacts of direction ground motion including those corresponding to the fault normal and fault parallel directions on several different engineering demand parameters are shown based on a linear 3D computer model of a six-story instrumented building. Kostinakis et al. (2008) examined the critical orientation of ground motion and the corresponding maximum response on the basis of the formulae developed by Athanatopoulou (2005) for special classes of buildings subjected to bidirectional ground motions. In general, these translational components of ground motion are correlated processes, but according to Penzien and Watabe (1974) there exists a set of orthogonal directions along which the components of ground motion may be considered uncorrelated.
