Classification of seismic vulnerability based on machine learning techniques for rc frames

S. H. Ghasemi, H. Bahrami, M. Akbari

Research output: Contribution to journalArticlepeer-review

6 Scopus citations

Abstract

Seismic vulnerability means the inability of historical and monumental buildings to withstand the effects of seismic forces. This article presents a classification model to specify the damage state of the Reinforced Concrete (RC) frames based on a collection of datasets from the damaged buildings in Bingol earthquake of Turkey for use in the learning process of the algorithm. The proposed model uses two classifiers including the redundancy and also the construction quality of the buildings to estimate the class of damage from four categories including none, light, moderate and severe. The available database of the considered earthquake includes the information of 27 damaged RC buildings which are published in the literature. The model provided a simple structure for engineers to predict the class without complex calculations in which it needs a few steps to determine the class of damage for RC frames. The results show that the presented model can estimate the class of each input vector with an acceptable error.

Original languageEnglish (US)
Pages (from-to)13-21
Number of pages9
JournalJournal of Soft Computing in Civil Engineering
Volume4
Issue number2
DOIs
StatePublished - 2020
Externally publishedYes

All Science Journal Classification (ASJC) codes

  • Civil and Structural Engineering
  • Building and Construction
  • Computer Science Applications
  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'Classification of seismic vulnerability based on machine learning techniques for rc frames'. Together they form a unique fingerprint.

Cite this