Back to search

University of Patras

Μία νέα μέθοδος επέκτασης των αναλύσεων στατικής οριακής ώθησης σε ασύμμετρα κτίρια και συγκρίσεις με υφιστάμενες μεθόδους

Abstract

dc:description

The behavior of a building under earthquake excitation can be viewed as a combination of its translational and torsional response. One of the most important factors that affects it is the geometry of its structural system, namely whether it is symmetric or not. Buildings with generally symmetric structural systems develop mostly translational vibrations under earthquake excitations, whereas torsional vibrations usually define the behavior of asymmetric buildings. Torsional response is considered undesirable, since it usually leads to undesirable increases in the earthquake induced stresses of the structural members located at the perimeter of a building. Hence, for more than sixty years the scientific community has shown a strong interest in the problem of accurately calculating and dealing with the effects of building inelastic torsion under earthquake excitations, and, although several interesting results have been derived as a product of this extensive research, torsion is still considered an “open issue”. One area that has attracted the interest of several researchers in recent years is the possibility of extending the well-known pushover analysis in problems of asymmetric buildings. Pushover analysis has been developed primarily for assessment of the seismic capacity of structures and for the identification of possible “weaknesses” in their design. It is an approximate procedure, whose main advantage lies in its simplicity, as it requires less data and time for the non-linear analysis of a given structure than the more complex, but more accurate, time history dynamic analysis. The method was developed for planar application and can provide results of acceptable accuracy when used for the analysis of symmetric buildings. However, several extensions of the procedure for spatial analyses of asymmetric buildings have been recently proposed. The proposed procedures are in many cases based on arbitrary assumptions, subsequently verified from analysis results using mostly simplified, “research-type” buildings. In other cases the proposed procedures are overly complicated, thus defeating the main advantage of the “conventional” pushover procedure, namely its simplicity. Considering the above, the development of a new method to extend the pushover procedure into problems of asymmetric buildings susceptible to torsional behavior was set as the main objective of the present thesis. Given the nature of the issue at hand and the shortcomings of existing methods, the main requirements and goal set for the development of the new procedure were for it to be based on scientifically consistent assumptions while remaining as simple as possible. Keeping that in mind, several static and dynamic approaches for the inelastic torsional behavior of asymmetric buildings using pushover analyses are examined. The main aim of this investigation is to evaluate the applicability and accuracy of each method under realistic “scenarios”. Thus, all the buildings used in the analyses were conventionally designed following current Regulations and using commercial design software. Our initial evaluations led to the proposal of a simple enough, yet “dynamic”, approach for the problem of inelastic torsion of asymmetric buildings, which provided the backbone of the new method for an extension of pushover analysis to non-symmetric buildings in 3-D. A properly formed, simple, 3 D.O.F. one-story model, with shear-beam type elements is used in order to quantify the effects of inelastic torsion on the overall seismic response of the building. Using results from non-linear time history analyses of the simple model, correction coefficients for the stiff and flexible edge translations are calculated and their values are then applied to modify the results of “conventional”, planar pushover analyses of a detailed model of the building. The simplified shear-beam type model is chosen as it is extremely easy to form and analyze, while the results obtained from its analyses are qualitatively similar to those of more complex, full scale models. For ease of reference the new method was called BA-3D pushover method (B, A: initials of Baros and Anagnostopoulos). In order to evaluate the proposed procedure the five and ten-story conventionally designed buildings were modeled as spatial frames using the well known plastic hinge model for the non-linear behavior of each member. Non-linear time history analyses were performed for each building using two sets of properly scaled, natural, far field and near field earthquake records. Both components of each record were considered to be acting simultaneously at each horizontal direction during the dynamic analyses in order to properly represent the effects of the earthquake excitation on each asymmetric building. The proposed procedure is compared with two “popular” existing methods addressing the same problem, namely the Modal Pushover Analysis (MPA) and the N2 method. All three methods are also evaluated by comparing their results for selected response quantities with those obtained from multiple non-linear time history analyses, which are considered representative of the target, “accurate” solution. The main conclusion from the investigations described above is that the new proposed procedure for the extension of pushover analysis into problems of asymmetric buildings susceptible to torsion leads to clearly improved results over the “conventional” pushover method. Compared to other alternative methods, namely the MPA and N2, the BA-3D pushover method gives similar and often better agreement with results obtained by non linear dynamic analyses. Moreover, the new method is considered more reliable, since the dispersion of the results obtained from the analysis of different buildings is generally lower than that of the two other procedures which were evaluated. Even in its current early version, the new method may be considered as a reliable and effective tool for the seismic assessment of any tosionally stiff asymmetric buildings, as well as for medium rise torsionally flexible ones (i.e. up to five or six stories). However, in the extreme cases of high-rise, torsionally flexible buildings all the existing examined extensions of the pushover method appear to fail, thus their use should be avoided. For such complex cases of buildings multiple, non-linear time history analyses appear to be the only reliable and accurate option.

Degree

thesis:*
Grantor dc:publisher
University of Patras
Year dc:date
2014

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Baros, Dimitrios
  • Μπάρος, Δημήτριος

Subjects

dc:subject × 16

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/51916

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Baros, Dimitrios; Μπάρος, Δημήτριος. Μία νέα μέθοδος επέκτασης των αναλύσεων στατικής οριακής ώθησης σε ασύμμετρα κτίρια και συγκρίσεις με υφιστάμενες μεθόδους. University of Patras, 2014. http://hdl.handle.net/10442/hedi/51916