Computational intelligence-based models for estimating the fundamental period of infilled reinforced concrete frames

One of the most important parameters used in the frame design process is the fundamental period. Numerous relationships are provided in the regulations and articles to determine the above parameter. However, due to the lack of optimal accuracy of existing equations, researchers are trying to improve them and increase the computational accuracy of the period. In this article, two computational intelligence-based is provided to determine the period of the infilled reinforced concrete frames. In the first method, the period is determined by a relationship that was extracted from the structure of a trained neural network and statistical regression analyses. The second method is a computational framework based on the neuro-fuzzy approach, which can be used to determine the period. In both models presented in this article, five variables including the number of stories and spans, span length, opening ratio, masonry wall stiffness were used. Furthermore, a large set of analytical data has been considered to learn the models. In order to evaluate the accuracy of the proposed model, a comparative study has been performed and its results were also presented. It was shown that both formulations proposed in this article have good performance and accuracy in estimating the fundamental period of infilled reinforced concrete frames. Furthermore, the sensitivity analyses indicated that the opening ratio is the most effective parameter on the period of the considered frames.

Automated summary

This article introduces two computational intelligence-based methods for determining the fundamental period of infilled reinforced concrete frames. By using statistical regression analysis, trained neural networks, and a neuro-fuzzy approach, the period can be accurately estimated. The results show that the opening ratio is the most influential parameter on the period of the frames considered.