Deep learning techniques for predicting nonlinear multi-component seismic responses of structural buildings

•A new approach to nonlinear seismic response predictions of structures using multiple-component features.•Deep learning time-series predictions through hybrid ConvLSTM models.•Signal processing improved with DWT of acceleration time-history.•Hybrid model can predict nonlinear seismic response of in...

Full description

Saved in:

Bibliographic Details
Published in:	Computers & structures Vol. 252; p. 106570
Main Authors:	Torky, Ahmed A., Ohno, Susumu
Format:	Journal Article
Language:	English
Published:	New York Elsevier Ltd 01.08.2021 Elsevier BV
Subjects:	Artificial neural networks Butterworth filters Computer architecture Convolutional long short-term memory neural networks Data processing Deep learning Deep neural networks Discrete Wavelet Transform Discrete wavelet transforms Fast Fourier transformations Fourier transforms Machine learning Multi-component response Neural networks Nonlinear seismic response prediction Recurrent neural networks Seismic response Superstructures Wavelet transforms Deep neural networks Multi-component response Discrete wavelet transforms Convolutional long short-term memory neural networks Nonlinear seismic response prediction
ISSN:	0045-7949, 1879-2243
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	•A new approach to nonlinear seismic response predictions of structures using multiple-component features.•Deep learning time-series predictions through hybrid ConvLSTM models.•Signal processing improved with DWT of acceleration time-history.•Hybrid model can predict nonlinear seismic response of industrial-level building. This paper presents a new approach for nonlinear multi-component seismic response prediction of structures using hybrid deep learning techniques. Modern recurrent neural networks map the relationship between acceleration time-series of the base/ground of a building and the superstructure, as a form of nonlinear time-history analysis method. Seismic responses were measured in three components which enables multi-component seismic predictions with adequate deep learning architectures. While long short-term memory (LSTM) neural networks can obtain data from a single component per surrogate model, hybrid convolutional-LSTMs (ConvLSTM) neural networks are utilized for multi-component purposes. A guide for pre-processing data and structuring the architecture of deep neural networks are proposed. Also, two filtering methods are compared, Fast Fourier Transform (FFT) Butterworth filter and discrete wavelet transform (DWT) decomposition. Decimation is implemented to reduce the features to useful values, as a dimension reduction approach. With enhancements to the architecture of the network, training time can be reduced significantly, and accuracy could be further improved. A challenging case study is addressed that covers an industrial level practical building. Results show that proposed hybrid models can even predict the capacity curves of a structure indirectly, providing new prospects for engineers to evaluate the seismic performance of a building.
Bibliography:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ISSN:	0045-7949 1879-2243
DOI:	10.1016/j.compstruc.2021.106570