skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Waterlogging risk assessment based on self-organizing map (SOM) artificial neural networks: A case study of an urban storm in Beijing

Abstract

Due to rapid urbanization, waterlogging induced by torrential rainfall has become a global concern and a potential risk affecting urban habitant's safety. Widespread waterlogging disasters have occurred almost annually in the urban area of Beijing, the capital of China. Based on a self-organizing map (SOM) artificial neural network (ANN), a graded waterlogging risk assessment was conducted on 56 low-lying points in Beijing, China. Social risk factors, such as Gross domestic product (GDP), population density, and traffic congestion, were utilized as input datasets in this study. The results indicate that SOM-ANN is suitable for automatically and quantitatively assessing risks associated with waterlogging. The greatest advantage of SOM-ANN in the assessment of waterlogging risk is that a priori knowledge about classification categories and assessment indicator weights is not needed. As a result, SOM-ANN can effectively overcome interference from subjective factors, producing classification results that are more objective and accurate. In this paper, the risk level of waterlogging in Beijing was divided into five grades. As a result, the points that were assigned risk grades of IV or V were located mainly in the districts of Chaoyang, Haidian, Xicheng, and Dongcheng.

Authors:
 [1];  [1];  [2];  [1];  [3]
  1. Beijing Normal Univ., Beijing (China); Beijing Key Lab. of Urban Hydrological Cycle and Sponge City Technology, Beijing (China)
  2. Argonne National Lab. (ANL), Lemont, IL (United States)
  3. China Institute of Water Resources and Hydropower Research, Beijing (China)
Publication Date:
Research Org.:
Argonne National Lab. (ANL), Argonne, IL (United States)
Sponsoring Org.:
National Natural Science Foundation of China (NNSFC); USDOE
OSTI Identifier:
1372090
Grant/Contract Number:
AC02-06CH11357
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
Journal of Mountain Science
Additional Journal Information:
Journal Volume: 14; Journal Issue: 5; Journal ID: ISSN 1672-6316
Publisher:
Science Press - Springer
Country of Publication:
United States
Language:
English
Subject:
54 ENVIRONMENTAL SCIENCES; waterlogging risk assessment; self-organizing map (SOM) neural network; urban storm

Citation Formats

Lai, Wen-li, Wang, Hong-rui, Wang, Cheng, Zhang, Jie, and Zhao, Yong. Waterlogging risk assessment based on self-organizing map (SOM) artificial neural networks: A case study of an urban storm in Beijing. United States: N. p., 2017. Web. doi:10.1007/s11629-016-4035-y.
Lai, Wen-li, Wang, Hong-rui, Wang, Cheng, Zhang, Jie, & Zhao, Yong. Waterlogging risk assessment based on self-organizing map (SOM) artificial neural networks: A case study of an urban storm in Beijing. United States. doi:10.1007/s11629-016-4035-y.
Lai, Wen-li, Wang, Hong-rui, Wang, Cheng, Zhang, Jie, and Zhao, Yong. Fri . "Waterlogging risk assessment based on self-organizing map (SOM) artificial neural networks: A case study of an urban storm in Beijing". United States. doi:10.1007/s11629-016-4035-y. https://www.osti.gov/servlets/purl/1372090.
@article{osti_1372090,
title = {Waterlogging risk assessment based on self-organizing map (SOM) artificial neural networks: A case study of an urban storm in Beijing},
author = {Lai, Wen-li and Wang, Hong-rui and Wang, Cheng and Zhang, Jie and Zhao, Yong},
abstractNote = {Due to rapid urbanization, waterlogging induced by torrential rainfall has become a global concern and a potential risk affecting urban habitant's safety. Widespread waterlogging disasters have occurred almost annually in the urban area of Beijing, the capital of China. Based on a self-organizing map (SOM) artificial neural network (ANN), a graded waterlogging risk assessment was conducted on 56 low-lying points in Beijing, China. Social risk factors, such as Gross domestic product (GDP), population density, and traffic congestion, were utilized as input datasets in this study. The results indicate that SOM-ANN is suitable for automatically and quantitatively assessing risks associated with waterlogging. The greatest advantage of SOM-ANN in the assessment of waterlogging risk is that a priori knowledge about classification categories and assessment indicator weights is not needed. As a result, SOM-ANN can effectively overcome interference from subjective factors, producing classification results that are more objective and accurate. In this paper, the risk level of waterlogging in Beijing was divided into five grades. As a result, the points that were assigned risk grades of IV or V were located mainly in the districts of Chaoyang, Haidian, Xicheng, and Dongcheng.},
doi = {10.1007/s11629-016-4035-y},
journal = {Journal of Mountain Science},
number = 5,
volume = 14,
place = {United States},
year = {Fri May 05 00:00:00 EDT 2017},
month = {Fri May 05 00:00:00 EDT 2017}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Save / Share:
  • This paper presents a new neural network architecture called the parallel, self-organizing, hierarchical neural network (PSHNN). The new architecture involves a number of stages in which each stage can be a particular neural network (SNN). At the end of each stage, error detection is carried out, and a number of input vectors are rejected. Between two stages there is a nonlinear transformation of those input vectors rejected by the previous stage. The new architecture has many desirable properties such as optimized system complexity in the sense of minimized self-organizing number of stages, high classification accuracy, minimized learning and recall times,more » and truly parallel architectures in which all stages are operating simultaneously without waiting for data from each other during testing. The experiments performed in comparison to multilayered networks with back-propagation training indicated the superiority of the new architecture.« less
  • Convection-resolving ensemble simulations using the WRF-Chem model coupled with a single-layer Urban Canopy Model (UCM) are conducted to investigate the individual and combined impacts of land use and anthropogenic pollutant emissions from urbanization on a heavy rainfall event in the Greater Beijing Metropolitan Area (GBMA) in China. The simulation with the urbanization effect included generally captures the spatial pattern and temporal variation of the rainfall event. An improvement of precipitation is found in the experiment including aerosol effect on both clouds and radiation. The expanded urban land cover and increased aerosols have an opposite effect on precipitation processes, with themore » latter playing a more dominant role, leading to suppressed convection and rainfall over the upstream (northwest) area, and enhanced convection and more precipitation in the downstream (southeast) region of the GBMA. In addition, the influence of aerosol indirect effect is found to overwhelm that of direct effect on precipitation in this rainfall event. Increased aerosols induce more cloud droplets with smaller size, which favors evaporative cooling and reduce updrafts and suppress convection over the upstream (northwest) region in the early stage of the rainfall event. As the rainfall system propagates southeastward, more latent heat is released due to the freezing of larger number of smaller cloud drops that are lofted above the freezing level, which is responsible for the increased updraft strength and convective invigoration over the downstream (southeast) area.« less
  • China is undergoing rapid urbanization because of unprecedented economic growth. As a result, many cities suffer from air pollution. Two-thirds of China's cities have not attained the ambient air quality standards applicable to urban residential areas (Grade II). Particulate matter (PM), rather than sulfur dioxide (SO{sub 2}), is the major pollutant reflecting the shift from coal burning to mixed source pollution. In 2002, 63.2 and 22.4% of the monitored cities have PM and SO{sub 2} concentrations exceeding the Grade II standard, respectively. Nitrogen oxides (NOx) concentration kept a relatively stable level near the Grade II standard in the last decademore » and had an increasing potential in recent years because of the rapid motorization. In general, the air pollutants emission did not increase as quickly as the economic growth and energy consumption, and air quality in Chinese cities has improved to some extent. Beijing, a typical representative of rapidly developing cities, is an example to illustrate the possible options for urban air pollution control. Beijing's case provides hope that the challenges associated with improving air quality can be met during a period of explosive development and motorization. 21 refs., 19 figs., 6 tabs.« less
  • A frequency-based model of an artificial neural network is being explored for active learning in a simulated environment and for its response to multiple modalities of input data. Physical sensors couple naturally to such a network, providing an easy migration path from simulation to application. The combination of an artificial neural network processing frequency-coded sensor information and implemented on advanced computer architectures is seen as an answer to the problems arising in robotics and the fusion of large quantities of multisensory data.
  • Force curves recorded with the atomic force microscope on structured samples often show an irregular force versus indentation behavior. An analysis of such curves using standard contact models (e.g., the Sneddon model) would generate inaccurate Young's moduli. A critical inspection of the force curve shape is therefore necessary for estimating the reliability of the generated Young's modulus. We used a trained artificial neural network to automatically recognize curves of “good” and of “bad” quality. This is especially useful for improving the analysis of force maps that consist of a large number of force curves.