טכניון מכון טכנולוגי לישראל
הטכניון מכון טכנולוגי לישראל - בית הספר ללימודי מוסמכים  
M.Sc Thesis
M.Sc StudentLeonid Burshtein
SubjectOptimal Strength Distribution for Seismic Design of
Frames Satisfying Ductility Demands
DepartmentDepartment of Civil and Environmental Engineering
Supervisors Professor Levy Robert
Professor Emeritus Rutenberg Avigdor


Abstract

In this paper a new method for optimal design of frames for satisfying given ductility demands is presented. The optimal design method, developed in the present work, is based on non-linear time history analysis and obtains optimal strength distribution by an iterative procedure. At every optimization iteration, the strength of each plastic hinge changes according to the ductility obtained in the previous iteration of the optimal design. The strength allocation formula that is proposed in the present study allows one to obtain not only equal ductility distribution, but also to achieve a desired level of ductility for every plastic hinge. This fact is very important, because it allows one to design frames according to the “strong column - weak beam” philosophy, that is: beams are designed to target ductility and columns remain elastic. The proposed technique allows design for an envelope of several earthquakes.

Several parameters have been studied in order to examine their influence on the optimal strength distribution. The parameters are: a number of storeys in the structure, different ductility levels, beam to column stiffness ratio and different relationships betwen strength and stiffness.

In the optimal design of frames carried out in the present study a continuous variation of cross-section properties is assumed. The strength to stiffness relationship used in the design procedure is based on realistic cross-sections.

Sensitivity analysis was performed in order to define the sensitivity of the structural systems to deviations from the optimum.

A technique for the determination of equivalent static load based on optimum strength distribution is also proposed.