Faster Post-Earthquake Damage Assessment Based on 1D Convolutional Neural Networks

Haibin Zhang,Donald Wunsch,Genda Chen,Liujun Li,Dustin Tanksley,Xinzhe Yuan

doi:10.3390/app11219844

Haibin Zhang, Donald Wunsch + Show 4 more

Open Access

https://doi.org/10.3390/app11219844

Copy DOI

Journal: Applied Sciences	Publication Date: Oct 21, 2021
Citations: 12	License type: CC BY 4.0

Affiliation: Missouri University of Science and Technology

Abstract

Contemporary deep learning approaches for post-earthquake damage assessments based on 2D convolutional neural networks (CNNs) require encoding of ground motion records to transform their inherent 1D time series to 2D images, thus requiring high computing time and resources. This study develops a 1D CNN model to avoid the costly 2D image encoding. The 1D CNN model is compared with a 2D CNN model with wavelet transform encoding and a feedforward neural network (FNN) model to evaluate prediction performance and computational efficiency. A case study of a benchmark reinforced concrete (r/c) building indicated that the 1D CNN model achieved a prediction accuracy of 81.0%, which was very close to the 81.6% prediction accuracy of the 2D CNN model and much higher than the 70.8% prediction accuracy of the FNN model. At the same time, the 1D CNN model reduced computing time by more than 90% and reduced resources used by more than 69%, as compared to the 2D CNN model. Therefore, the developed 1D CNN model is recommended for rapid and accurate resultant damage assessment after earthquakes.

Highlights

The convolutional neural networks (CNNs) training time with wavelet transform (WT) encoding reached more than 20 h on a general-purpose computer without graphics processing unit (GPU) farms, according to Lu et al [23]
The prediction performances and the computational three models, withneural the same training, validation, and testbased datasets from efficiencies oftrained these three network models are evaluated on obtained the case-study the nonlinear time history analyses (NLTHA)
This comparison further performance assessment using the same unseen test set, which was not involved in indicated that the CNN models outperformed the feedforward neural network (FNN) model in the ground motion records (GMRs)-dependent the training process

Summary

Introduction

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. The contemporary CNN-based seismic damage assessments avoid the computation and selection of hand-crafted IMs, the encoding of a 1D GMR time series to 2D GMR images increases the training time for a large set of GMRs. The CNN training time with WT encoding reached more than 20 h on a general-purpose computer without graphics processing unit (GPU) farms, according to Lu et al [23]. The prediction performances and the computational three models, withneural the same training, validation, and testbased datasets from efficiencies oftrained these three network models are evaluated on obtained the case-study the nonlinear time history analyses (NLTHA). The prediction performances and the computational efficiencies of presented These three neural network models are evaluated based on the case-study results. Conclusions from of our research and future research recommendations are presented

Methodology

Configuration of Three Neural Network Models

BenchmarkBuilding

The diagram of selecting balanced dataset of 3201

Figure

Findings

Discussion

Conclusions