Flooding Overlay: Difference between revisions
| Line 15: | Line 15: | ||
==Numerical scheme== | ==Numerical scheme== | ||
The Tygron Engine Inundation module relies on an explicit finit volume method, taken from Kurganov and Petrova (2007). This scheme relies on a reconstruction of cell bottom, water level and velocity at the interfaces between computational cells as proposed by Lax and Wendroff (see Rezzolla, 2011). The reconstruction method, taken from Bolderman et all (2014) ensures numerical stability | The Tygron Engine Inundation module relies on an explicit finit volume method, taken from Kurganov and Petrova (2007). This scheme relies on a reconstruction of cell bottom, water level and velocity at the interfaces between computational cells as proposed by Lax and Wendroff (see Rezzolla, 2011). The reconstruction method, taken from Bolderman et all (2014) ensures numerical stability at the wetting and drying front of a flood wave. | ||
==Computational time step== | ==Computational time step== | ||
Revision as of 08:14, 29 June 2018
What is the Inundation overlay
The inundation is developed for the computation of surface (water) flow in a wide range of applications. These applications include large scale inundations due to dike breaches and inundations at the bottom of hill-slopes. Therefore the commonly applied 2D Saint Venant Equations are implemented on a highly parallel applicable explicit numerical scheme suitable for Tygron GPU clusters.
2D Saint Venant equations
The 2D Saint Venant equations describe the conservation of mass in a gridcell and the conservation of momentum in both x and y, direction:
File:Inundation overlay 01.PNG
The Saint Venant equations describe the following processes:
- friction
- bed slope
- water pressure
- convection (changes in bathemetry over space)
- inertia (increase or decrease of velocity over time)
Numerical scheme
The Tygron Engine Inundation module relies on an explicit finit volume method, taken from Kurganov and Petrova (2007). This scheme relies on a reconstruction of cell bottom, water level and velocity at the interfaces between computational cells as proposed by Lax and Wendroff (see Rezzolla, 2011). The reconstruction method, taken from Bolderman et all (2014) ensures numerical stability at the wetting and drying front of a flood wave.
Computational time step
References
- Rezzolla L (2011) ∙ Numerical Methods for the Solution of Partial Differential Equations
- Bollermann A, Chen G, Kurganov A and Noelle S (2014) ∙ A Well-Balanced Reconstruction For Wetting/Drying Fronts https://www.researchgate.net/publication/269417532_A_Well-balanced_Reconstruction_for_Wetting_Drying_Fronts
- Kurganov A, Petrova G (2007) ∙ A Second-Order Well-Balanced Positivy Preserving Central-Upwind Scheme for the Saint-Venant System ∙ http://www.math.tamu.edu/~gpetrova/KPSV.pdf