Optimal design of reinforced concrete frames considering seismic design provisions using genetic algorithms with modified leader reproduction

• Main Author: Balili, Alden Paul D.
• Format: text
• Language: English
• Published: Animo Repository 2009
• Subjects: Structural frames; Earthquake resistant design; Genetic algorithms; Structural Engineering
• Online Access: https://animorepository.dlsu.edu.ph/etd_masteral/6935
• Institution: De La Salle University

This paper discusses the application of Genethic Algorithms (GA) and a concrete database in the design optimization of concrete frames that are resistant against seismic loads and is compliant to relevant seismic provisions of the latest version of the NSCP. The basics of GAs are introduced first and limitations of plain GA are discussed. Then, the application of GA to the optimization of concrete frames is discussed. After this, the Concrete Optimization Program (COP) and its 4 main components are discussed briefly. The performance of the COP using different GA constants and improved operators were tested on a 2-storey frame. From the initial results, it can be observed that using simple GA alone even with the enhancements is insufficient to provide a effective and efficient means of optimizing a concrete frame. In light of this situation, the modified leader reproduction (MLR) was conceived to improve the overall performance of the COP. The performance of the COP with MLR was tested on both a 2-storey and 4-storey frame. From the results of the COP run with MLR, it was confirmed that using MLR greatly enhanced the efficiency and the capability of the COP to yield economical and feasible results.