Prediction of Heavy Rain Damage Using Deep Learning

被引:15
作者
Lee, Kanghyeok [1 ]
Choi, Changhyun [2 ]
Shin, Do Hyoung [1 ]
Kim, Hung Soo [1 ]
机构
[1] Inha Univ, Dept Civil Engn, Incheon 22212, South Korea
[2] KB Claims Survey & Adjusting, Risk Management Off, Seoul 06212, South Korea
基金
新加坡国家研究基金会;
关键词
deep learning; disaster management; heavy rain damage; prediction model; NEURAL-NETWORKS; CLASSIFICATION;
D O I
10.3390/w12071942
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Heavy rain damage prediction models were developed with a deep learning technique for predicting the damage to a region before heavy rain damage occurs. As a dependent variable, a damage scale comprising three categories (minor, significant, severe) was used, and meteorological data 7 days before the damage were used as independent variables. A deep neural network (DNN), convolutional neural network (CNN), and recurrent neural network (RNN), which are representative deep learning techniques, were employed for the model development. Each model was trained and tested 30 times to evaluate the predictive performance. As a result of evaluating the predicted performance, the DNN-based model and the CNN-based model showed good performance, and the RNN-based model was analyzed to have relatively low performance. For the DNN-based model, the convergence epoch of the training showed a relatively wide distribution, which may lead to difficulties in selecting an epoch suitable for practical use. Therefore, the CNN-based model would be acceptable for the heavy rain damage prediction in terms of the accuracy and robustness. These results demonstrated the applicability of deep learning in the development of the damage prediction model. The proposed prediction model can be used for disaster management as the basic data for decision making.
引用
收藏
页数:18
相关论文
共 56 条
[31]   Predicting hurricane wind damage by claim payout based on Hurricane Ike in Texas [J].
Kim, Ji-Myong ;
Woods, Paul K. ;
Park, Young Jun ;
Kim, Taehui ;
Son, Kiyoung .
GEOMATICS NATURAL HAZARDS & RISK, 2016, 7 (05) :1513-1525
[32]   Enhanced Self-Renewal and Accelerated Differentiation of Human Fetal Neural Stem Cells Using Graphene Oxide Nanoparticles [J].
Kim, Jin ;
Yang, Kisuk ;
Lee, Jong Seung ;
Hwang, Yong Hwa ;
Park, Hyun-Ji ;
Park, Kook In ;
Lee, Dong Yun ;
Cho, Seung-Woo .
MACROMOLECULAR BIOSCIENCE, 2017, 17 (08)
[33]  
Kons Z, 2013, INTERSPEECH, P1481
[34]   Preface: Damage of natural hazards: assessment and mitigation [J].
Kreibich, Heidi ;
Thaler, Thomas ;
Glade, Thomas ;
Molinari, Daniela .
NATURAL HAZARDS AND EARTH SYSTEM SCIENCES, 2019, 19 (03) :551-554
[35]   ImageNet Classification with Deep Convolutional Neural Networks [J].
Krizhevsky, Alex ;
Sutskever, Ilya ;
Hinton, Geoffrey E. .
COMMUNICATIONS OF THE ACM, 2017, 60 (06) :84-90
[36]   A Damage Localization Approach for Rahmen Bridge Based on Convolutional Neural Network [J].
Lee, Kanghyeok ;
Byun, Namju ;
Shin, Do Hyoung .
KSCE JOURNAL OF CIVIL ENGINEERING, 2020, 24 (01) :1-9
[37]   A Novelty Detection Approach for Tendons of Prestressed Concrete Bridges Based on a Convolutional Autoencoder and Acceleration Data [J].
Lee, Kanghyeok ;
Jeong, Seunghoo ;
Sim, Sung-Han ;
Shin, Do Hyoung .
SENSORS, 2019, 19 (07)
[38]   Landslide susceptibility assessment using SVM machine learning algorithm [J].
Marjanovic, Milos ;
Kovacevic, Milos ;
Bajat, Branislav ;
Vozenilek, Vit .
ENGINEERING GEOLOGY, 2011, 123 (03) :225-234
[39]   Flash Flood Risk Perception by the Population of Mindelo, S. Vicente (Cape Verde) [J].
Martins, Bruno ;
Nunes, Adelia ;
Lourenco, Luciano ;
Velez-Castro, Fatima .
WATER, 2019, 11 (09)
[40]   Multi-variate flood damage assessment: a tree-based data-mining approach [J].
Merz, B. ;
Kreibich, H. ;
Lall, U. .
NATURAL HAZARDS AND EARTH SYSTEM SCIENCES, 2013, 13 (01) :53-64