November 24, 2024
Abdoreza Fazeli

Abdoreza Fazeli

Academic Rank: Assistant professor
Address:
Degree: Ph.D in Civil Engineering
Phone: -
Faculty: Faculty of Engineering

Research

Title
Evaluation of the behavior factor of moment resisting RC frames using Gene Expression
Type Thesis
Keywords
ضريب رفتار- توصيف ژن- قاب بتن مسلح- تحليل استاتيكي غيرخطي
Researchers Shaker Hashemi (Primary advisor) , Abdoreza Fazeli (Advisor)

Abstract

Generally, structures have a nonlinear behavior during the occurrence of an earthquake. Thus, the exact design of a structure is accomplished through nonlinear analysis. However, designers mostly employ the linear analysis method and the reduced earthquake loads due to the complex and time-consumption nature of the nonlinear analysis. Overall, the linear earthquake response spectrum is utilized to obtain an earthquake lateral force for the design of structures in a linear analysis. Indeed, it is reduced using a coefficient called behavior factor (R) or response modification factor. In this thesis, gene expression programming (GEP) is employed to estimate the behavior factor of the reinforced concrete moment-resisting frames. The GEP technique is a genetic algorithm that applies a population of individuals and selects them based on a fitness function. Then, it introduces genetic changes using one or more genetic operators. In this regard, more than five hundred reinforced concrete frames were designed and analyzed in SAP2000 software. The nonlinear static pushover analysis has been utilized to evaluate these models. Also, various parameters have been considered in these models. These parameters included the number of stories (2, 4, 6, 8, 10, 12, and 15), the ratio of the span length to the story height (1, 1.5, 2, and 2.5), the design base acceleration (0.25, 0.30, and 0.35), site classification of the soil (type II and III), and the ratio of the concrete compressive strength to the longitudinal reinforcement yield stress (0.08 and 0.075). After calculating the behavior factor using valid methods, a database has been created through the behavior factors derived from the models and transferred to the GEP. Finally, GEP has been utilized to develop an empirical equation for estimating the behavior factor. This equation has resulted in a regression coefficient of 90%. Also, evaluations demonstrated that the proposed equation had acceptable accuracy. After extracting the