Seismic response spectrum rule for non-structural components in buildings

1. Introduction: The 2011 Great East Japan earthquake caused significant damages to non-structural components like suspended ceilings and other contents leading to substantial economic losses. The response of non-structural component can be evaluated through time history response analysis by modeling the components as secondary systems attached to the building. However, the time history analysis approach is time consuming and does not lend itself readily to design applications. For design purposes, a simplified prediction method that can give accurate estimate on the vibration response of non-structural components subject to the support input excitations from the building floors is very helpful. Previously developed methods sometimes have inconsistent accuracy because of limitations on the parameters used. 2. Dynamic Characteristics of a Single-Degree-of-Freedom Model: A single-degree-of-freedom (SDOF) model is used to represent the multi-degree-of-freedom (MDOF) building subjected to ground acceleration. The absolute acceleration response from the frame model is used as an input to another SDOF model representing non-structural component. Several building and non-structural components properties are selected to represent various possible range. These models are shown in Fig. 1. Results of time history analyses of the models are shown in Figs. 2, 3. The relationship between building acceleration and nonstructural component acceleration are investigated. 3. Estimation of Spectral Pseudo-Acceleration and Displacement of Non-Structural Components: The study in Chap. 2 has shown two different trends (Fig. 3): When the component vibration period is short relative to that of the building, the component response is similar to harmonic response. When it is long relative to that of building, the component response is like that directly subjected the ground motion. Thus, two different transfer functions are developed to relate the building acceleration to component acceleration (Fig. 4), and ground acceleration, via building acceleration, to non-structural component acceleration (Fig. 6). In order to consider ground motion's characteristics such as impulse (Fig. 7) and duration (Fig. 8), a few adjustments are made to improve the accuracy of the response prediction. The accuracies are shown for the pseudo-spectral accelerations (Figs. 10, 12) of non-structural components and spectral displacement (Fig. 13). 4. Multi-Mode Time-History Responses of Non-Structural Component in Multi-Story Building: Time-history responses of the non-structural component subjected to building accelerations are investigated by examining the effects from various modes of the building vibration. The 30-story building is used, and when the component vibration period is 0.3 sec, up to the 8th mode building accelerations are necessary to closely simulate the building acceleration (Fig. 14) as well as non-structural component acceleration history (Fig. 15). The modes higher than 9th makes the building modal responses similar and almost in-phase, but their total contribution is only 4% of the nonstructural component response, which is understood based on the participation vectors and their phase relationships. 5. Response Prediction of Non-Structural Component in Multi-Story Building: Instead of the time-history analysis discussed above, spectral combination rules are studied. The complete-quadratic-combination (CQC) rule to combine the modal response of the building is now applied to estimate the peak response of the non-structural component. The CQC rule is able to reflect the above-mentioned effects of the phase in contrast to the square-root-of sum-of-squares (SRSS) rule results (Fig. 17). Using the 30-story building, the component response spectra at almost the top and mid-story of the building are produced, respectively, by using the proposed method. The accuracies of the method in predicting the peak accelerations and displacements are recognized (Figs. 18, 20). 6. Conclusions: An improved simplified method to predict the pseudo-spectral acceleration as well as spectral displacement of non-structural components in buildings is proposed. It is verified using single-mass systems, as well as the multi-story building. © 2019 Architectural Institute of Japan. All rights reserved.