ASFPM 2016: Applications of 2D Surface flow Modeling in the New HEC-RAS Version 5.0
Download as PPTX, PDF2 likes1,842 views
This document presents an overview of the applications of 2D surface flow modeling in HEC-RAS version 5.0, highlighting its advancements from prior 1D capabilities. It discusses various technical advantages, project benefits, and specific applications such as dam breach modeling, while also noting some challenges and limitations of the software. The integration of digital terrain maps and improved computational methods are emphasized as key improvements in this version.
ASFPM 2016: Applications of 2D Surface flow Modeling in the New HEC-RAS Version 5.0
- 1. June 23, 2016 Applications of 2D Surface Flow Modeling in the new HEC-RAS Version 5.0 ASFPM Annual National Conference, Grand Rapids, MI Concurrent Session G Derek Etkin, P.E.
- 2. Contents 1. Overview of HEC-RAS prior to 2D capabilities 2. Introduction to 2D Surface Flow Modeling in version 5.0 3. Applications of HEC-RAS 2D capability 4. Advantages and Opportunities 2
- 3. HEC-RAS Overview 3 Open channel flow freeware published by USACE HEC-2 developed in 1966 (FORTRAN), HEC-RAS released in 1995 Generates flood profiles from 1D Open Channel Flow energy and continuity equations
- 4. HEC-RAS Overview Cross sections developed from survey and LiDAR using HEC-GeoRAS, an extension in ArcMap 4 flowto
- 5. HEC-RAS Overview Inundation mapping generated from profile interpolated between cross sections and intersected with terrain data 5 Source: FEMA (2013) Federal Guidelines for Inundation Mapping of Flood Risks Associated with Dam Incidents and Failures (FEMA P-946)
- 6. HEC-RAS Overview – Steady State 1-D Steady Flow – constant discharge 6 2 o f ( /A)Q h+ gA( - S + S )=0 x x α Q=VA Momentum Continuity Simulate single, peak discharge from hydrologic analysis
- 7. HEC-RAS Overview – Unsteady Flow 1-D Unsteady Flow – variable discharge and storage 7 Simulate routed hydrograph, storage between cross sections 2 o f ( /A)Q Q hgA( ) 0S St x x α Q A+ =0 x t Momentum Continuity
- 8. Representing with 2D Flow with 1D HEC-RAS 8 Split Flow
- 9. Representing with 2D Flow with 1D HEC-RAS 9 Offline Storage Areas
- 10. Introducing HEC-RAS version 5.0! 10
- 11. 2D Flow Areas in HEC-RAS 5.0 11 2D Flow Areas can be added to geometry file like Storage Areas Digital Terrain Map now integrated into HEC-RAS input files 2D Flow Area properties imported from Digital Terrain Map
- 12. 2D Flow Areas in HEC-RAS 5.0 12
- 13. 2D Flow Area Calculations 13 Unstructured Mesh with Implicit Finite Volume solver Hydraulic properties of computational cells (and faces) pre-processed from Digital Terrain Map Diffusion Wave or Full Momentum 2D Equations Source: Brunner, G.W. (2014) Combined 1D and 2D Modeling with HEC-RAS Source: USACE (2016) HEC-RAS 2D Modeling User’s Manual (CPD-68A)
- 14. 2D Flow Area Calculations 14 Pre-processed stage-storage for each cell Cells do not have a “flat bottom” or single depth. Cell Volume
- 15. 2D Flow Area Calculations 15 Pre-processed cross section for each cell face from Terrain Map Conveyance between cells defined by rating curves Cell Face
- 16. 2D Flow Area Calculations 16 High resolution Digital Terrain Map Lower resolution computational mesh
- 17. 2D Flow Area Linkages 17 2D Flow Areas can be linked to 1D Flow Reaches 1D Flow Reaches Upstream Downstream Lateral Connections Storage Areas or other 2D Flow Areas External Boundary Condition Normal Depth, Rating Curve, Stage or Flow time series
- 21. HEC-RAS 2D Riverine Applications 21 Useful for riverine flow outside of well-defined, single channels: Inter-connected or braided channels Dam breach flood waves with unpredictable paths
- 22. HEC-RAS 2D Example Application – Dam Breach 22
- 23. HEC-RAS 2D Example Application – Dam Breach 23
- 24. HEC-RAS 2D Applications – Dam Breach 24
- 25. HEC-RAS 2D Applications – Dam Breach 25
- 26. HEC-RAS 2D Technical Advantages 26 Implicit Finite Volume approach Improved stability Cells can start completely dry More robust than finite element or finite difference Allows for larger time steps than explicit methods Unstructured Mesh Flexibility Cells do not have flat bottom Allows larger computational cells without loss of terrain details Cells can be sized according to terrain features
- 27. HEC-RAS 2D Project Advantages 27 Public Domain No license fees Large community of practitioners Widely accepted as HEC product Pre- and Post-processing in MAPPER (and GeoRAS) Can easily integrate 2D Flow Areas into existing HEC-RAS models Update old 1D HEC-RAS models No need to decide 1D or 2D when selecting modeling software Can use for screening approach and where to focus detail
- 28. CHALLENGES Mesh editing tools limited 28
- 29. CHALLENGES Some limited capacity for modeling structures Can represent weirs, levees, and culverts Cannot use full 1D bridge modeling capabilities 29
- 30. CHALLENGES Processing of map inundation, removing “islands” from high resolution grid; cleaning up “Leaking” Can use hydraulic connectivity plots 30
- 31. CHALLENGES Limited by quality of LiDAR/Raster… Adding bathymetry to LiDAR for “pure” 2D Flow Area runs 31
- 32. Thank you. 32
Editor's Notes
- #9: Careful selection of 1D reaches and orientation of cross sections that avoid overtopping. Assumes that bifurcation is understood Splits to well-defined channels
- #10: Areas outside of the main channel which may experience lateral inundation and provide peak attenuating storage of the floodwave. Assume a single water surface elevation. No momentum simulation.
- #11: Beta released in May 2014 First Final version released in 2016
- #12: RAS Mapper is more integrated in version 5.0 Digital Terrain Map is easily generated from more widely available LiDAR sets
- #13: 2D Flow Areas can be added and modified to the model geometry in the same GUI as always. 2D Flow Areas
- #14: Full Momentum (“Saint-Venant”) Gravity, friction, hydrostatic pressure, acceleration: waves and detailed flow transitions and collision with objects Good for rapid changes in elevation vs. Diffusion Wave Just gravity, friction, pressure
- #22: Interconnected braided channels = like the previous example No need for defining each single 1D reach before running simulation.
- #23: Two small reservoirs in series and on the steep valley slopes above a large New England river. We wanted to know how laterally the wave would spread based on the topographic data we had.
- #24: Two small reservoirs in series and on the steep valley slopes above a large New England river with development along the banks. A large culvert can handle 50-yr flow (~100 cfs), but the dam break we simulated in a 1D simulation had peak flow of ~3000 cfs. We wanted to know how laterally the wave would spread based on the topographic data we had. The quality of the LiDAR doesn’t show the structures in the flat area
- #25: Dam breach for a water supply reservoir outside of urban area. Another consultant made a 1D model with detailed structure information but also with cross sections not wide enough in some places. The peak WSELs appeared to cause flow splits. Made this screening run using their breach hydrograph to identify places where there might be overtopping into adjacent areas; how much,
- #26: The screening simulation took us about 4 hours in total to pull together and helped us figure out an approach for updating the existing 1D model with a lateral connection to a 2D flow area that follows the overflow path
- #29: We’re still learning how to make the best of the mesh editing tools, such as the fixed Break Line Cell Spacing. The orientation and location of the cell faces is important. Defines conveyance. Still looking into developing tools in GIS or maybe SMS
- #31: Also, helpfulThe default inundation mapping uses a sloping water surface where the surface is interpolated between cell centers,
- #32: Unlike 1D cross sections, it is difficult to manually edit the geometry data in 2D Flow Areas. You are really bound to the LiDAR you have. LiDAR of course doesn’t include bathymetry. There is a great tool that can use a 1D reach to carve out a channel in the Terrain (in MAPPER), but this is only for areas where you have already built a 1D model. For “pure” 2D runs, there is a challenge representing structures and large/deep rivers, where the LiDAR doesn’t capture the conveyance of the channel.