Approaches to the rapid seismic damage prediction of r/c buildings using artificial neural networks. (15th June 2018)

• Record Type: Journal Article
• Authors: Morfidis, Konstantinos; Kostinakis, Konstantinos
• Is Part Of: Engineering structures. Volume 165(2018)
• Page Start: 120
• Page End: 141
• Publication Date: 2018-06-15
• Subjects: Seismic damage prediction -- Artificial neural networks -- Pattern recognition -- R/C buildings -- Seismic vulnerability assessment -- Seismic response; Structural engineering -- Periodicals; Structural analysis (Engineering) -- Periodicals; Construction, Technique de la -- Périodiques; Génie parasismique -- Périodiques; Pression du vent -- Périodiques; Earthquake engineering; Structural engineering; Wind-pressure; Periodicals; 624.105
• DOI: 10.1016/j.engstruct.2018.03.028
• Languages: English
• ISSNs: 0141-0296

Abstract

Highlights: Investigation of MFP ANNs' ability to predict the r/c buildings' seismic damage. Formulation as a function approximation and as a pattern recognition problem. The impact of ANNs' configuration on the predictions' reliability is examined. The best configured ANNs' prediction ability is checked using seismic scenarios. The ANNs are able to give reliable damage predictions in real time after the shock. Abstract: The present paper deals with the investigation of the ability of Artificial Neural Networks (ANN) to reliably predict the r/c buildings' seismic damage state. In this investigation, the problem was formulated as a problem of approximation of an unknown function as well as a pattern recognition problem. In both cases, Multilayer Feedforward Perceptron networks were used. For the creation of the ANNs' training data set, 30 r/c buildings with different structural characteristics, which were subjected to 65 actual ground motions, were selected. These buildings were subjected to Nonlinear Time History Analyses. These analyses led to the calculation of the buildings' damage indices expressed in terms of the Maximum Interstorey Drift Ratio. The influence of several configuration parameters of ANNs to the level of the predictions' reliability was also investigated. In order to investigate the generalization ability of the trained networks, three scenarios were considered. In the framework of these scenarios, the ANNs' seismic damage state predictions were evaluated for buildings subjected to earthquakes, neither of which are included to the training data set. The most significant conclusion of the investigation is that the ANNs can reliably approach the seismic damage state of r/c buildings in real time after an earthquake.