Civil Engineering Association

Capacity Design Optimization of Steel Building Frameworks Using Nonlinear Time-History Analysis

A PhD thesis in Civil Engineering presented to the University of Waterloo, Ontario, Canada

Author(s)/Editor(s): Yusong Xue | Size: 3,77 MB | Format: PDF | Quality: Unspecified | Year: 2012 | pages: 160

This study proposes a seismic design optimization method for steel building frameworks following the capacity design principle. Currently, when a structural design employs an elastic analysis to evaluate structural demands, the analysis results can be used only for the design of fuse members, and the inelastic demands on non-fuse members have to be obtained by hand calculations. Also, the elastic-analysis-based design method is unable to warrant a fully valid seismic design since the evaluation tool cannot always capture the true inelastic behaviour of a structure. The proposed method is to overcome these shortcomings by adopting the most sophisticated nonlinear dynamic procedure, i.e., Nonlinear Time- (or Response-) History Analysis as the evaluation tool for seismic demands.
The proposed optimal design formulation includes three objectives: the minimum weight or cost of the seismic force resisting system, the minimum seismic input energy or potential earthquake damage and the maximum hysteretic energy ratio of fuse members. The explicit design constraints include the plastic rotation limits on individual frame members and the inter-story drift limits on the overall performance of the structure. Strength designs of each member are treated as implicit constraints through considering both geometric and material nonlinearities of the structure in the nonlinear dynamic analysis procedure. A multi-objective Genetic Algorithm is employed to search for the Pareto-optimal solutions.
The study provides design examples for moment resisting frames and eccentrically braced frames. In the examples some numerical strategies, such as integrating load and resistance factors in analysis, grouping design variables of a link and the beams outside the link, roundingoff the objective function values, are introduced. The design examples confirm that the proposed optimization formulation is able to conduct automated capacity design of steel frames. In particular, the third objective, to maximize the hysteretic energy ratio of fuse members, drives the optimization algorithm to search for design solutions with favorable plastic mechanisms, which is the essence of the capacity design principle.  
For the proposed inelastic-analysis-based design method, the seismic performance factors (i.e., ductility- and overstrength-related force reduction factors) are no longer needed. Furthermore, problem-dependent capacity design requirements, such as strong-column-weak-beam for moment resisting frames, are not included in the design formulation. Thus, the proposed design method is general and applicable to various types of building frames.