The doctoral dissertation describes the methodology for the assessment of the seismic resistance of reinforced concrete (RC) structures at different levels of complexity. The methodolgy starts with a preliminary rapid visual screening method which is based on the American method. This is followed by procedures which operate at five different levels of complexity, including simple procedures which are based on methods that were originally developed in Japan (levels 1 and 2), the N2 method with two variants of the mathematical model (levels 3 and 4), and the non-linear dynamic analysis (level 5). A separate chapter deals with the assessment of the load-carrying capacity of structural members, which is assessed differently at the different level of complexity. Models for the assessment of flexural and shear behaviour are presented, and the results obtained when using the different models are compared with experimental results for column and wall specimens from the SERIES database. In the case of flexural behaviour, the assessment of flexural strength of the structural members is presented, together with the effective yield rotation and the ultimate rotation. Based on data given in the SERIES
database, assumed values for the ultimate stresses are selected in the case of columns at the 1st level. In the case of the three highest levels of complexity the assessments of the seisimc resistance of the structures are performed by using a mathematical model of the structure. A simplified model is used at the 3rd level, where the pushover analyses are performed using a slightly modified version of the original NEAVEK program. Here the mathematical model is based on an extension of the pseudo-threedimensional mathematical model into the non-linear range. In contrast to the 3rd level, at the 4th and 5th levels standard mathematical modelling is applied. In this case the non-linear analyses are performed using the OpenSees software, together with the PBEE-toolbox. Seismic resistance assessment was performed on twelve variants of several frame structures, four variants of (cantilever) wall structures, and five variants of dual structures. The results of the performed seismic resistance assessments for all the investigated buildings indicate a small difference between the N2 method and the non-linear dynamic analysis, whereas the results of the procedures at the first two levels were less reliable, and are much more conservative. On the other hand, the amount of input data and the scope of computational work increases with the increasing level of complexity. The reasons for the observed conservatism of the lower two levels, and also of the 3rd level, are explained by an evaluation of the individual assumptions made. Research into the definition of load-carrying capacity is needed, especially with respect to the shear capacity of structural members and the capacity of the whole structure. It was found that the seismic demand depends strongly on the initial effective stiffness of the structure. For this reason the choice of an adequate initial stiffness at the element level is very important. A uniformly reduced stiffness to one half of that corresponding to the uncracked gross-sections, as allowed by EC8, may grossly underestimate the seismic demand. The choice of initial stiffness may have a larger influence on the seismic resistance than the choice of the analysis procedure.
|
