Optimization of performance of RC buildings against progressive collapse and comparison between loading Modes Using neuro-fuzzy Inference System

In recent years, the vulnerability of buildings to progressive collapse has attracted many researchers. Progressive collapse is a failure in the whole structure or relatively large part of it, caused by accidents damaging part of the structure and the inability of adjacent members to redistribute overload through a path that can maintain the overall stability of the structure. In general, reinforcement of structures in most of these cases is much more economical than construction rebuilding. For this reason, many scholars have addressed the issue of reinforcement. Strengthening the structure is further sought to increase strength and improve the ductility and behavior of existing members. These reinforcements can lead to an increase in the flexural or shear capacity of the structure, or both. In this research, the purpose of analyzing and evaluating the performance of reinforced concrete buildings against progressive collapse is by means of an optimal neuro-fuzzy inference system along with a genetic algorithm. For this purpose, the proposed model will first be introduced into Excel using initial values and then the results will be analyzed using the proposed neural network method and definition of its layers based on the input values. The data input in the genetic algorithm is presented as four designs, considering the two stages of training as a 3-day age and 7-day age. The results of this algorithm are used in the neural network.

Keywords:

Concrete buildings , progressive collapse , reinforced concrete slab , neuro fuzzy inference system