Study on A 2D depth-averaged hyper-concentrated flow model

Haij Rong Chung*, Te Yung Hsieh, Jinn Chuang Yang

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

A two dimensional hyper-concentrated shallow water flow model based on the orthogonal curvilinear coordinate system embedded with the quadratic rheology relation to demonstrate the non-Newtonian flow characteristics has been established. The implicit two step split-operator approach is employed to split the governing equations into the advection and propagation steps. The control volume concept is used for numerical discretization. Two dam-break surge cases, including straight flume in laboratory and 2D plane in field, are taken as the examples to verify the model capability for simulating the surge propagation and the stagnation position. Furthermore, an experimental case with supercritical flow condition in straight channel is simulated to validate the model's capability through the comparison of water surface elevation variation along the channel. The nonuniformity of flow behavior in horizontal plane induced by the irregular side-wall boundary was also analyzed by an experimental case under the condition of hyper-concentrated supercritical flow in bend. For all of the cases studied, the simulated results show well agreement with the experimental data. To demonstrate the applicability of the numerical model, the hyper-concentrated flow event of Tung-pu-ray creek during typhoon Toraji (2001) is simulated, and the super-elevation of water surface and the bed shear stress distribution show reasonable results according to field observation.

Original languageEnglish
Pages (from-to)1-13
Number of pages13
JournalTaiwan Water Conservancy
Volume60
Issue number2
StatePublished - Jun 2012

Keywords

  • 2D model
  • Bend flow
  • Depth-averaged
  • Hyper-concentrated flow
  • Non-Newtonian fluid
  • Quadratic rheology
  • Supercritical flow

Fingerprint

Dive into the research topics of 'Study on A 2D depth-averaged hyper-concentrated flow model'. Together they form a unique fingerprint.

Cite this