Assessment of nonlinear static (pushover) analysis procedures using field experience

Dilsiz, Abdullah
Recently, many nonlinear analysis procedures have been proposed for earthquake response determination of the structures. Although, the nonlinear response history analysis (NRHA) is accepted as the most accurate source of information for nonlinear seismic response, nonlinear static procedures (NSP) may also provide reasonable estimates of seismic demand and inelastic behavior. However, all proposed NSPs have limitations, due to the certain approximations and simplifications, such as invariable load pattern and single mode consideration. This study is concentrated on the “NSPs” which are generally compared with the “exact results” of NRHA. The current widely used NSPs’ results were compared with the results of both NRHA and the “real” results (real building performance records or experimental results). The results of observations of real structures which are subjected to strong ground motions were used for the assessment. In addition, the buildings were evaluated using nonlinear static and nonlinear dynamic detailed assessment procedures of the current codes. Considering “If I had known that this Earthquake would happen 1 day before the occurrence of the earthquake, could I estimate the damage states, using the widely used NSPs?” moderately and heavily damaged building samples have been collected from Adapazarı and analytical models formed. According to the results of NSPs and NRHA of studied buildings, there is no clear result that any of the procedures used can identify the performance point suitably for each condition. Most of the analyses results could not predict the level of damage accurately. Using these results it is not possible to determine the seismic response and the damage of the buildings before the occurrence of earthquake. The expectations obtained from the NSPs also do not comply with the results of NRHA. Thus; there is no safety for the compatibility of pushover procedures as well as the code specifications with field observations, yet. Considering the high effort given for the computation and post-process of the analyses results, global seismic performance of the buildings were assessed by preliminary assessment procedures. In contrary with the detailed assessment results, the vulnerable buildings studied could be evaluated successfully and qualified according to moderate or severe damage experienced during the earthquake by some preliminary assessment procedures. The valuable information about the seismic behavior of RC buildings obtained from the tests should be supported with more data obtained from the field. This strikes a pessimistic tone because if the inconsistencies between field data and assessment procedures described in guidelines on account of fluctuations of material properties, geometries, ground motion variations and many other parameters considered then a clear need exists to be sanguine about the predictive powers of these methods.