Investigation of Parameters Affects on the Phenomenon of Progressive Collapse in Steel Structures Using Artificial Neural Network (ANN)

According to the American Building Loading Code, progressive failure is defined as the spread of failure in a structure from one element to another, which ultimately leads to the failure of the entire structure or a large part of it. Lead to this type of failure are: car impact, gas explosion, aircraft collision, construction error, fire, accidental overloading of members, explosion, etc. Most of these accidents have a short impact time that in The result leads to dynamic responses. In this research, 100 types of models will be considered and the progressive collapse reference codes will be used for analysis and the application software will be used to achieve the answer under nonlinear static analysis by removing the column and performing nonlinear static analysis. The behavior of structural members against this phenomenon will be evaluated and finally the results will be controlled by the rules in the regulations and the final behavior of the structure against this phenomenon will be examined. Artificial neural networks have been used as a powerful tool in various fields of engineering and their use is increasing. Civil engineering and the study of the behavior of structures are no exception to this rule and neural methods have been used in it. In this study, three types of neural networks have been used to investigate the phenomenon of progressive collapse. Input and output parameters are selected and modeled with the help of evaluation indicators of this phenomenon. Finally, by comparing the performance of artificial neural networks, the most successful type of neural network has been proposed.