Hazard Assessment Based on the Combination of DAN3D and Machine Learning Method for Planning Closed-Type Barriers against Debris-Flow

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 44
  • Download : 0
If a slope located near a densely populated region is susceptible to debris-flow hazards, barriers are used as a mitigation method by placing them in flow channels; i.e., flowpaths. Selecting the location and the design of a barrier requires hazard assessment to determine the width, volume, and impact pressure of debris-flow at the moment of collision. DAN3D (Three-Dimensional Dynamic Analysis), a 3D numerical model for simulating debris-flow, has been widely used to perform hazard assessment; however, solely using DAN3D would be both insufficient and inefficient in finding the optimal barrier location. Therefore, the present study developed a framework that interprets the results from DAN3D simulation without considering any barriers. Then, the framework generates hazard assessment maps showing the impact parameters of debris-flow along the flowpath by various algorithms and machine learning methods, such as the k-means clustering algorithm, and also computes the width of the debris-flow, which is not explicitly calculated in DAN3D. A case study of the debris-flow at Umyeon mountain, Korea, in 2011, was used to generate hazard assessment maps. The maps were demonstrated to be a tool to quickly compute the impact parameters for conceptual barrier design with the aim of finding potential barrier locations.
Publisher
MDPI AG
Issue Date
2020-01
Language
English
Article Type
Article
Citation

WATER, v.12, no.1, pp.170

ISSN
2073-4441
DOI
10.3390/w12010170
URI
http://hdl.handle.net/10203/273915
Appears in Collection
CE-Journal Papers(저널논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0