Structures for 1D FLOW in DelftShell

Introduction

In this document an inventarisation is made of the hydraulic structures used by 3 1d modeling software packages:
Sobek - Deltares
Mike11 - DHI Danish Hydraulics Institute
HEC-RAS - US Army Corps of Engineers
And how this functionality should be implemented in DelftShell

Analysis

Sobek will be integrated as 1D ModelFlow into DelftShell and thus it is the logical point of departure for DelftShell. The current implementation of Sobek has some aspects that need to be improved:

• The implementation in Sobek of structures seems unnecessary complex. There are for example 4 sorts of weirs
  -Compound structures are not implemented uniformly. Only river structures are supported.
  -Controllers and triggers are not implemented uniformly. River based structures can have up to 4 controllers and triggers. Non river based structures can have only 1 controller.
  HECRAS has less different types of structures but they are more complex. At the highest level they only support 4 types but internally they are more like the Sobek Structures.
  Mike11 has an approach like Sobek, but the number is more resticted and they have some additional structures like Control Stucture, Regulating which probably can be compared to Lateral Inflow, Extra Resistance, etc in Sobek.
  
  

  Sobek	HECRAS	Mike11
  River weir (0 - R)	Bridge/Culvert

• Deck/Roadway
  
• Pier
  
• Sloping Abutment
  
• Culvert
  
• Bridge|Weir|

  River advanced weir (1 - R)

  Inline Structure

• Weir/embankment
  
• Gate| |

  Weir (6)
  Universal weir (11- CS)
  Culvert, Siphon and Inverse siphon (10 - CS)		Culvert
  Orifice (7)
  General structure (2 - R)	Lateral Structure

• Weir/embankment
  
• Gate
  
• Culvert| |

  River pump (3 - R)	Pump Station	Pump
  Pump (9)
  Bridge (12 - CS)		Bridge
  Breach growth 1D Dam break node (13)		Dambreak Structure
  Breach growth 2D Dam break node (112)
  Database structure (4 - R)		User Defined
  		Tabulated Structure
  Compound Structure (#R)
  		Hydraulic Control
  [Controllers and triggers]		Control Structure
  		Regulating
  		Energy Loss

  Table 1.: structures in 1D modeling packages. For Sobek the number between brackets refer to the internal Sobek MDB type number (5 and 8 do not exist), R relates to River structures. Energy Loss is not available in Sobek 2.13 but will be covered by Extra Resistante in a next release. CS relates to structures in Sobek that have an internal Cross Section.
  The implementations of Sobek and Mike11 have much in common. HECRAS has a different approach in that it represents the physical reality more closely. This is also visible in the presentation. Mike11 and Sobek both use fixed symbols for displaying structures. Hecras draws structures as polygon on the map and in the structure editor displays a more realistic view of the structure. In fact the HECRAS structures are compound structures.
  
  
  figure1. Bridge in HECRAS is drawn as a polygon that represents the physical entity.
  
  figure 2. The structure dialog in HECRAS provides a more realistic view at the data.
  
  One of the requisitions of DelftShell is that it should be able to support the import of existing Sobek models to Model1DFlow. This makes a purely HECRAS approach less preferable. The uncertainty in translating the current Sobek structures to a HECRAS like setup is simply too big. The implicit use of compund structure is very nice but too restrictive for the current possibilities in Sobek. It should be more generic.
  What also could be nice is a presentation in the map view of structures as (multiple) polygon(s). A possibilty is a solution like cross sections in the network editor of DelftShell where they can be represented as a line or a symbol on the map.
  As for Mike11, their approach is like the Sobek implemtation and the User Interface is not very intuitive. The main difference is the implementation of a number of extra structure types.
  

  Dialog structure

  Another more cosmetic difference between Sobek on the one side and HECRAS and Mike11 on the other is the complexity of the structure dialogs.
  Mike11 has too complex and badly organised dialogs
  HECRAS dialogs are better structured but the possiblity to open multiple popups dialogs (7 for a bridge) can make things confusing
  Sobek also is better structured than Mike11, the usage of tabbed windows keeps items simple and reusable. On the other hand tabbed windows in Sobek do not allow the user to see multiple options simultaneaously. This is especially annoying when defining controllers and triggers for River based structures.
  

  Conclusions

• Make the use of compound structures implicit. If more than 1 structure (component) is defined at a location This solution can also be found in the 1D model Duflow, a stucture is placed at a stucture point, when a second strcuture is added to that point both are handled as paralel structures. this will be treated by the model engine as compound structure If restrictions apply they could be implemented like the topology rules in the network editor. The multiple structures component that make up the structure should be visisble on the map.
• Decrease the number of structure types
• implement controllers and triggers in a uniform and not restrictive way. This is an implementation like the River Structures but restricted to 4 controllers/triggers.

Discussion/improvements/wishes

• Use polygon(s) for rendering of structures on the map. Ideally make this zoom level dependent. Render a (bitmap) symbol if polygons become tool small.
• The usage of tabbed dialog/popup dialog/property grids all have their pros and cons.

A simplified definition of a 1D network:
Network has 1 or more branches
Branch has 0 or more branch features
Branch feature is Cross Section or Structure
Structure has 1 or more Structure Components
Structure Component is Weir, Culvert, Bridge, Pump or General Structure
Structure Component has 0 or more Controllers
Controller has 0 or more Triggers

Appendix Sobek

Sobek
For all structures

Controllers and triggers are implemented differently for several structures

If relevant the River Structures support up to 4 controllers and triggers, the rural structures only 1 controller.
The parameters that can be controlled are:
Crest level of a weir
Crest level and crest width of River weir
Crest level of Advanced weir
Crest level, crest width and gate height of General structure
Crest level of Database structure
Opening height of a orifice
Capacity of a pump (Rural Flow module only)
Valve opening of a culvert
Valve opening of a siphon
Valve opening of an inverted siphon


controller usage

controller definitions, only if usage is for at least 1 controllers

triggers, only if usage is for at least 1 controllers


controller implementation for Weir, Orifice, Pump, Culvert

Cross Section for
Culvert:Round, Egg Shape, Tabulated, Rectangle. Elliptical, Arch, Cunette, Steel Cunette
Bridge: Rectangle, Tabulated

Friction for Cross Section Culvert

Cross Section (YZ and Asymmetrical Trapezium and no separate friction possible) for Universal Weir

River weir (0 - R)

id = id of the structure definition
nm = name of the structure definition
ty = type structure (0=Sobek weir)
cl = crest level
cs = crest shape
0 = broad
1 = triangular
2 = round
3 = sharp
cw = crest width
po = positive correction coefficient (positive = in positive flow direction)
ps = positive submergence limit
pt pr = positive reduction table
first column = (h2 - z) / (h1 - z)
second column = reduction factor for positive flow direction
no = negative correction (negative =in negative flow direction)
first column = (h2 - z) / (h1 - z)
second column = reduction factor in negative flow direction
ns = negative submergence limit
nt nr = negative reduction table

River advanced weir (1 - R)

id = id of the structure definition
nm = name of the structure definition
ty = type structure (1=Sobek advanced weir)
cl = crest level
sw = sill width
ni = number of piers
ph = positive upstream face height
nh = negative upstream height
(=height of the weir relative to the bed level at the upstream side)
pw = positive weir design head
nw = negative weir design head
pp = positive pier contraction coefficient
np = negative pier contraction
pa = positive abutment contraction coefficient
na = negative abutment

Weir(6)

id = id of the structure definition
nm = name of the structure definition
ty = type of structure (6= SOBEK Urban/Rural weir)
cl = crest level
cw = crest width (-1 : look at profile)
ce = discharge coefficient (depends on crest shape)
sc = lateral contraction coefficient
rt = possible flow direction (relative to the branchdirection):
0 : flow in both directions
1 : flow from begin node to end node (positive)
2 : flow from end node to begin node (negative)
3 : no flow

Universal weir (11)

No Controllers
ty = type of structure
11 = Universal Weir
cl = crest level
si = id of cross section definition (profile.def), only YZ Table Profile and Asymetrical Trapezoidal Profile
ce = coefficient in discharge formulation
sv = water level based modular limit for rectangular sections, default 0.667 not to be edited
rt = possible flow direction (relative to the branch direction):
0 : flow in both directions
1 : flow from begin node to end node (positive)
2 : flow from end node to begin node (negative)
3 : no flow

Culvert, Siphon and Inverse siphon (10)

ty = type of structure
10 = culvert or siphon or inverse siphon
tc = type of culvert
1 = culvert
2 = siphon
3 = inverse siphon
rl = bed level (right)
ll = bed level (left)
si = id of cross section definition (profile.def), only closed profiles
li = inlet loss coefficient
lo = outlet loss coefficient
lb = bend loss coefficient
ov = initial opening level of valve
tv = table of loss coefficient
0 no table, no valve
1 valve present, reference to table in file valve.tab. See detailed decription of this file below.
rt = possible flow direction (relative to the branch direction):
0 : flow in both directions
1 : flow from begin node to end node (positive)
2 : flow from end node to begin node (negative)
3 : no flow
dl = length of culvert, siphon or inverse siphon
hs = start level of operation of siphon
he = end level of operation of siphon

Orifice(7)

id = id of the structure definition
nm = name of the structure definition
ty = type of structure (7 = SOBEK Urban/Rural orifice)
cl = crest level
cw = crest width
gh = gate height
mu = contraction coefficient
sc = lateral contraction coefficient
rt = possible flow direction (relative to the branch direction):
0 : flow in both directions
1 : flow from begin node to end node (positive)
2 : flow from end node to begin node (negative)
3 : no flow
mp = maximum flow in positive direction
mp <switch> <value>, where switch
0 : do not use
1 : use max. flow
mn = maximum flow in negative direction
mp <switch> <value>, where switch
0 : do not use
1 : use max. flow

Pump (9)

id = id of the pomp definition
nm = name of the pomp definition
ty = type of structure (9 = pump)
dn = flow direction/ control
1 = upward control
2 = downward control
3 = downward + upward control
-1, -2, -3 : same as positive, but flow direction opposite to branch direction. Or, in the case of a lateral structure, for an extraction of water.
rt cr 1 = reduction function of the water level difference, in a table
column 1 = water level difference (suction side - pressure side)
column 2 = reduction factor for capacity
rt cr 0 1 0 = constant reduction factor (in this case, the reduction factor is 1).
ct lt 1 = Table with 5 columns.
column 1 = extra capacity;
column 2 and 3 =start- en stop level for suction side
column 4 and 5 =start- en stop level for pressure side

River pump (3 - R)

controllers and triggers are not supported.
id = id of the structure definition
nm = name of the structure definition
ty = type of structure (3=Sobek River Pump)
dn = control direction
1 = upward control, controlled side faces the beginning of the branch, controlled side is suction, pumps in positive branch direction.
-1 = downward control, controlled side faces the end of the branch, controlled side is suction side, pumps in reverse branch direction.
2 = downward control, controlled side faces the end of the branch, controlled side is delivery side, pumps in positive branch direction.
-2 = upward control, controlled side faces the beginning of the branch, controlled side is delivery side, pumps in reverse branch direction.
rt cr = reduction table for the pump capacity, as a function of the water level difference.
first column = level difference,
second column = reduction factor
rt cr 0 = constant
rt cr 1 = table
ct lt = (one row) table with pump capacity and start and stop levels.
The first item is the pump capacity, the second and the third item are the start and stop levels at the suction side of the pump and the
fourth and the fifth item are the start and stop levels at the delivery side of the pump.

Bridge (12)

No Controllers
ty = type of structure
12 = bridge
tb = type of bridge
2 = pillar bridge
3 = abutment bridge
4 = fixed bed bridge
5 = soil bed bridge
si = id of cross section definition (profile.def), only open profiles (if tb =3,4, or 5)
pw = total width of pillars in direction of flow (if tb=2)
vf = form factor (if tb=2)
li = inlet loss coefficient
lo = outlet loss coefficient
dl = length of bridge in flow direction.
rl = bottom level

General structure (2 - R)

id = id of the structure definition
nm = name of the structure definition
ty = type of structure (2= Sobek general structure)
w1 = w1: width left (upstream) side of structure
wl = wSdl: width structure left side
ws = wS: width structure centre
wr = wSdr: width structure right side
w2 = w2: width right (downstream) side of structure
z1 = zb1: bed level left side of structure
zl = zbSl: bed level left side structure
zs = zbS: bed level at centre of structure
zr = zbSr: bed level right side structure
z2 = zb2: bed level right side of structure
gh = gate height
pg = positive free gate flow (positive=in positive flow direction)
pd = positive drowned gate flow
pi = positive free weir flow
pr = positive drowned weir flow
pc = positive contraction coefficient
ng = negative free gate flow (negative=in negative flow direction)
nd = negative drowned gate flow
nf = negative free weir flow
nr = negative drowned weir flow
nc = negative contraction coefficient
er = extra resistance
if not present a default (setting) value will be used

Database structure (4 - R)

id = id of the structure definition
nm = name of the structure definition
ty = type of structure (4= databse structure)
cl = crest level or reference level for values h1, h2 in data base
di = interpolation type
0 = linear
1 = spline
dm = third dimension of data base i.e. number of gate values. In this release 1.
d2 = value at second axis
0 : h2
1 : dh = h1 - h2


and where in the MATR record:
id = identification of structure
db qt = table containing the data base
db fi = table that defines the part of the data base that has been filled in by the user.

Breach growth 1D Dam break node (13)

id = id of the structure definition
nm = name of the structure definition
ty = structure type (13=Sobek dambreak)
cl = crest level = Initial top level w.r.t. reference level [m]
cs= crest shape
0 = broad crest (default)
cw = crest width - initial width/gap [m]
ml = minimum level w.r.t. reference level [m]
td = type of dambreak-formula
1 = vdKnaap (2000) [optional]
2 = Verheij-vdKnaap(2002) (to be implemented in SOBEK version 2.09)
3 = ... not defined yet
f1 = alpha constant for Verheij-vdKnaap(2002) (note: it is f1= F + one)
f2 = beta constant for Verheij-vdKnaap(2002)
uc =critical flow velocity sediment/soil [m/sec] for Verheij-vdKnaap(2002)
ce = coefficient of discharge (not used)
rt = possible flow direction (relative to the branch direction):
0 : flow in both directions (default and only possible value)
The following parameters are only used to generate the controller table
eq 0 = sand for vdKnaap (2000) formula (if type of dambreak formula = vdKnaap; thus td = 1)
eq 1 = clay for vdKnaap (2000) i.e td = 1 (if type of dambreak formula = vdKnaap; thus td = 1)
ts = time start in 'yyyy/mm/dd;hh:mm:ss'
dt = elasped time after the time start to reach lowest level ml in 'dd:hh:mm:ss'
ec = maximum breach width in m - for vdKnaap (2000)
0 200 = constant width of 200 m
Note: td 1 is by default and can be optional. For backward compatibility.
Initial top level w.r.t. reference level [m] - maximum initial opening depth [m] = minimum level w.r.t. reference level [m]

Breach growth 2D Dam break node (112)

id = id of the structure definition
nm = name of the structure definition
ty = structure type (112 = 2D Dam break node)
t0 = start time in hours from start of simulation
t1 = end time (not if if table) in hours
dh lt 1= decrease in height with time - table with time in hours and height decrease in m
dh lt 0 10.0 = constant value - linear decrease in height m start at start time
wg 0 0 = width growth at rate of 5% of grid size *** not yet implemented
wq 1 0.6 = width growth at rate specified in m/sec *** not yet implemented
wd 0 0 = no increase in width growth with time (assumed total grid size at beginning of break) *** not yet implemented
wd 10.5 = increase in width with time in m till grid size *** not yet implemented

Compound Structure (# - R)

Appendix Mike11

Mike 11 structures
For each structure the location is defined by:

Location

• Branch Name: Name of the river branch in which the pump is located.
• Chainage: Chainage at which the pump is located.
• ID: String identification of the pump. Used for identification of the pump in case of multiple structures at the same location. Specification of pump ID is recommended.
• Type: The lcation type may be Regular, Side Structure or Side Structure + Reservoir. See section 2.3 Tabular view: Structures for details

Weir
Attributes
Type:

• Broad Crested Weir: The calculation of Q/h relations assumes critical flow at the crest.
• Special Weir: The Q/h relationship table must be specified.
• Weir Formula 1: A standard weir expression is applied. See the Reference Manual.
• Weir Formula 2 (Honma): The Honma weir expression is applied. See the Reference Manual.
  Valve:
• None: No valve regulation applies.
• Only Positive Flow: Only positive flow is allowed, i.e. whenever the water level downstream is higher than upstream the flow through the structure will be zero.
• Only Negative Flow: Only negative flow is allowed, i.e. whenever the water level upstream is higher than downstream the flow through the structure will be zero.
  Head Loss Factors
  Negative
• Inflow
• OutFlow
• Free Overflow
  Positive
• Inflow
• OutFlow
• Free Overflow
  Geometry
  Level-Width: The weir geometry is specified as a level/width table relative to the datum.
  Cross Section DB: The weir geometry is specified in the cross section editor. A cross section with a matching branch name, Topo ID and chainage must exist in the applied cross section file. The Topo ID is assumed to be the same as that specified in the Branches Property page, see Topo ID .
  Datum: Offset which is added to the level column in the level/width table.
  Level/Width: Weir shape defined as levels and corresponding flow widths. Values in the levels column must be increasing.
  Weir formula Parameters (only weir formula 1)
  Width: Width of the flow.
  Height: Weir height. See Figure 2.9
  Weir Coeff.: Multiplication coefficient in the weir formula.
  Weir Exp.: Exponential coefficient in the weir formula.
  Invert Level: Bottom datum level. See Figure 2.9
  Weir formula 2 Parameters (only weir formula 2 (Honma))
  Weir coefficient (C1): Multiplication coefficient in the Honma weir formula.
  Weir width: Width of the flow.
  Weir crest level: Weir level.
  Weir formula 3 Parameters (only weir formula 2 (Honma))
  QH Relation
  Culvert
  
  Attributes
  Upstream Invert: Invert level upstream of the culvert.
  Downstr. Invert: Invert level downstream of the culvert.
  Length: Length of the culvert.
  Manning's n: Manning's bed resistance number along the culvert.
  No. of Culverts: Number of culvert cells.
  Valve Regulation:
  None: No valve regulation applies.
  Only Positive Flow: Only positive flow is allowed, i.e. whenever the water level downstream is higher than upstream the flow through the structure will be zero.
  Only Negative Flow: Only negative flow is allowed, i.e. whenever the water level upstream is higher than downstream the flow through the structure will be zero.
  Section Type: Closed or Open.
  Head Loss Factors
  Negative
• Inflow
• OutFlow
• Free Overflow
• Bends
  Positive
• Inflow
• OutFlow
• Free Overflow
• Bends
  Geometry
  The cross sectional geometry of a culvert can be specified as:
  Rectangular: The width and height specify the geometry.
  Circular: The geometry is specified by the diameter.
  Irregular Level-Width Table: The geometry is specified using a level/width table. Values in the level column must be increasing.
  Irregular Depth-Width Table: The geometry is specified using a depth/width table. Values in the width column must be increasing.
  Section DB: The geometry is specified by a cross section
  Pump
  
  Control Parameters
  Start Level: Water level at the inflow that activates the pump.Note that for pumps with internal outlet the inflow is situated at the previous h-point (previous with regard to chainage) in case of positive discharge and at the next h-point with regard to chainage) in case of negative discharge. The sign of the discharge follows from the specifications made under Pump Data.
  Stop Level: Water level at which the pump starts closing down.
  Start-up Period: Period for changing pump discharge from zero to full. The pump discharge is changed linearly in time.
  Close Down Period: Period for changing pump discharge from full to zero. The pump discharge is changed linearly in time.
  Pump Data
  Outlet:
  Internal: Water is pumped internally in the river branch.
  External: Water is pumped out of the river branch.
  Specification Type:
  Fixed Discharge: Pump rate independent of the local water head expect for the start/stop control.
  Tabulated Characteristic: Pump rate controlled by specified characteristic (Q-dH-curve) and the water level difference between upstream water level and outlet level/downstream water level.
  Discharge: Pump rate when applying "Fixed Discharge".
  Outlet Level: Level of pump outlet. The outlet may be submerged or free. Relevant only in case of "Tabulated Characteristic".
  Q-dH-curve: Q-dH-characteristic of the pump.
  Bridge
  
  The following types are supported:
  FHWA WSPRO
  USBPR
  Submerged Bridge
  Arch Bridges (Biery and Delleur)
  Arch Bridges (Hydraulic Research (HR))
  Bridge Piers (D'Aubuisson)
  Bridge Piers (Nagler)
  Bridge Piers (Yarnell)
  Energy Equation
  YFHWA WSPRO
  USBPR
  Submerged Bridge
  Arch Bridges (Biery and Delleur)
  Arch Bridges (Hydraulic Research (HR))
  Bridge Piers (D'Aubuisson)
  Bridge Piers (Nagler)
  Bridge Piers (Yarnell)
  Energy Equation
  TODO: fill out all options
  Dambreak Structure
  TODO: fill out all options
  User Defined
  Used to write plugins (dll's) with their own structure
  Tabulated Structure
  The Tabulated Structure property page is used for defining a structure regulated by a user defined relation between the discharge through the structure and the up- and downstream water level. The relation is defined in a table.
  Calculation Mode:
• Q = f(h U/S, h D/S)
• H U/S= f(h D/S, Q)
• H D/S= f(h U/S, Q)
  Number of Columns
  Number of Rows
  Water level datum
  Discharge factor
  Hydraulic Control
  Control Structure
  Regulating
  TODO: fill out all options
  Energy Loss
  
  Apply energy loss
  Alignment change
  Roughness coefficient
  Positive energy flow loss coefficient
  Negative energy flow loss coefficient
  

  Appendix HECRAS

  
  
  Bridge/Culvert
  
  
• Deck/Roadway
  
• Pier
  
• Sloping Abutment
  
• Bridge Modeling approach
  
• Culvert
  
• Multiple Opening Analysis
  
  
• HTab Param
  
• Bridge Design
  
  Inline Structure
• Weir/embankment
  This is the geometry of the weir cross section
  
  HECRAS inline structure
  
  HECRAS inline structure - Weir/ Embankment
  
• Gate
  
  
  HECRAS inline structure - Gate
  
  Lateral Structure
  
  
• Weir/embankment
  
  
• Gate
  
• Culvert
  
• Diversion RC
  
  Pump Station