Feasibility study on the use of neural networks in predicting the dynamic collapse capacity of a structure using data obtained from overflow analysis
Mehdi Maghsoudlou1
1. Civil and Municipal Administrator
Abstract
The purpose of this study was to determine the feasibility of using neural networks in predicting the dynamic collapse capacity of a structure using the obtained data from incremental analysis. The present study was initially carried out using a library and research method in the research theory section. Then Matlab software was used to design structural models. Undoubtedly, the most important and main result obtained is the confirmation of the usability of neural networks in predicting the capacity of collapse of structures, and also the data that are the result of static nonlinear analysis can be converted to artificial neural networks into a submergence capacity Dynamic collapse. The next discussion was to examine the number of input parameters to the neural network, which results showed that the use of only two parameters of overlap analysis, ie, the coefficient of over strength and ductility, can not be considered as input to the neural network and it is necessary that other parameters The first mode period is considered as a network input.
Keywords: Neural Networks - Enhancement - Dynamic Dispersion - Increased Expansion
