Upload
donhan
View
228
Download
1
Embed Size (px)
Citation preview
Modelling LIDs using PCSWMM and EPA SWMM5
March 28, 2012
Presented by: Rob James (CHI)
Credit to: Lewis Rossman (US EPA)
1977SWMM ported to minicomputer
1981EPA SWMM 3released
1984PCSWMM released1st PC versionFORTRAN
1988EPA SWMM 4released
2004EPA SWMM 5released
1991PCSWMM4First GUI versionBASIC
1995PCSWMM ‘95First 32-bit Windowsversion - Visual Basic
1998PCSWMM ’98First GIS-based versionVisual Basic
2002PCSWMM 2002Genetic algorithmbased calibration
IBM PC/AT Intel 80-486 IBM PowerPC Intel Pentium III Intel Quad Core
1993SWMM-USERSlistserv
2007PCSWMM.NETFirst multi-threadedversion - C#, .NET 3.5
2010-2011PCSWMM 2011• Real-time flood
forecasting• Integrated 2D
modeling• French, Spanish,
Chinese language versions
Intel i7
Processing power increase over the history of PCSWMM
100,000,000
TIMES
PCSWMM`s computational grid at CHI
1,000,000,000,000
FLOPS
So how much is 1 trillion floating point operations?
1 x 60s x 60m x 2100h x 40y
302,400,000
1,000,000,000,000= 0.03%
PCSWMM: Spatial DSS for EPA SWMM5
PCSWMM: Spatial DSS for EPA SWMM5
PCSWMM/SWMM5 LID toolbox
• Physically-based processes
• Flexible components
• SWMM Version 5.0.022
• Simplifies the modeling of:– Lot level implementations
– Watershed scale or city master planning
– Single event
– Continuous
– Performance degradation
Examples of SWMM5 LID controls
Vegetative Swale Rain Garden
Street Planter
Rain Barrel
Porous Pavement Infiltration Trench
EPA SWMM5 LID toolkit: follows the methodology of PCSWMM for PP
Conceptual Model of an LID Process
Surface Zone
Soil Zone
Storage Zone
Runon
Infiltration
Percolation
Infiltration
Overflow
Underdrain
ET
Flow Balance Equations
11101 qfeqt
d
pfeft
D 212
3333 qfeft
dp
Surface d1
Soil
Storage d3
q0
f1
fp
f3
q1
q3
e1, e2, e3
D2
Infiltration Flux
Classical Green-Ampt Eqn: (depth d is normally ignored) F
dKf sat
))((1
0
2
4
6
8
10
12
14
0 5 10 15 20 25 30
Time (minutes)
Infi
ltra
tio
n R
ate
(in
/hr)
depth included depth ignored
0
1
2
3
4
5
6
7
0 50 100 150 200 250 300 350 400
Time (minutes)
Po
nd
ed
Dep
th (
inch
es)
depth included depth ignored
Effect of ponded depth (d) on infiltration rate (Ksat = 0.5 in/hr)
DKf p
)(1)(
Rate of percolation (fp) through the unsaturated soil zone as a function of moisture content ( is described by Darcy’s Law:
))(exp( HCOKK sat
))(exp(135 PCOFC
K = hydraulic conductivity, = capillary tension, = porosity, FC = field capacity, and HCO and PCO are coefficients.
Percolation Flux
Outflow Fluxeso Flux rates are functions of zone’s water depth (d)
o Surface zone
– Overland flow using Manning’s formula
Q = (1.49/n)AR2/3S1/2 where A and R depend on d
– Overflow using the weir equation
Q = CWL(d)1.5
o Storage zone underdrain flow
Q = CD(d)
n = Manning’s roughness, A = flow area, R = hyd. radius, S = surface slope, L = length of weir crest, and Cw, CD, and are coefficients.
Representing Different LID Alternatives
LID Alternative Zones Processes (besides ET)
Rain Barrel Surface,
Storage
Surface Overflow
Storage Underdrain Flow
Porous Pavement Surface,
Storage
Surface Overland Flow
Storage Infiltration*
Infiltration Trench Surface,
Storage
Surface Overflow
Storage Infiltration
Vegetative Swale Surface Surface Overland Flow
Surface Infiltration
Bioretention Cell Surface,
Soil,
Storage
Surface Overflow
Soil Infiltration
Soil Percolation
Storage Infiltration*
*May also include storage underdrain flow
Bio-retention cell – ‘Street Planter’
LID example: Valleyfield, PQ
• New residential development to include 22 boulevard planters
• Reduce minor system inflow
• Reduce pollutant loading
Boulevard Planters
Alternate designs
Valleyfield, Quebec – LID retrofit example
Bio-retention cell to be installed in typical residential neighbourhood
PCSWMM LID representation
PCSWMM LID representation: surface
PCSWMM LID representation: surface storage
PCSWMM LID representation: soil
PCSWMM LID representation: storage
PCSWMM LID representation: tile drain
Continuous LID analysis
Continuous LID analysis: small events
Continuous LID analysis: large events
Continuous LID analysis: clogging potential
Enabling continuous LID analysis: Rainfall Processing with PCSWMM
Runoff for 1-inch, 6-hr Event
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0:00 1:00 2:00 3:00 4:00 5:00 6:00 7:00
Ru
no
ff (
cfs
)
No Planters 3 Planters 5 Planters
Rossman (2009)
Effect of Number of Planters
0
10
20
30
40
50
60
70
80
90
0 1 2 3 4 5 6
Number of Planters
Perc
en
t R
un
off
Rossman (2009)
36
Loss of Stored Water Over Time
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
0 10 20 30 40 50
Time (hrs)
Losses (in/hr) Storage Depth (ft)
Rossman (2009)
Runoff With No Planters
Elapsed Time (days)
400350300250200150100500
Ru
no
ff (
CF
S)
0.8
0.6
0.4
0.2
0.0
Runoff With Five Planters
Elapsed Time (days)
400350300250200150100500
Ru
no
ff (
CF
S)
0.8
0.6
0.4
0.2
0.0
Long Term Performance
Rossman (2009)
Olds College Demonstration Project
Olds College: lot-level analysis
Olds College: subcatchments
Olds College : roof to cistern
Olds College: roof to landscaping
Olds College : absorbent landscaping
Olds College: bio-retention area
Olds College : parking lot 2
Olds College: oil & grit separator
Olds College: parking lot 2
Olds College: permeable pavement
Olds College: wet pond
Olds College: Irrigation
Continuous simulation of re-use
Irrigation
If the rainfall in the preceding two days is more than 10 mm, irrigation is delayed.
Irrigation
• Rule 1IF SIMULATION MONTH = 5AND SIMULATION DAY = 6AND NODE SU_RG DEPTH = 0AND NODE SU DEPTH > 2THEN PUMP P1 SETTING = 1
• Rule 2IF SIMULATION MONTH = 6AND SIMULATION DAY = 3AND NODE SU_RG DEPTH = 0AND NODE SU DEPTH > 2THEN PUMP P1 SETTING = 1
• Rule 3IF SIMULATION MONTH = 6AND SIMULATION DAY = 6AND NODE SU_RG DEPTH = 0AND NODE SU DEPTH > 2THEN PUMP P1 SETTING = 1.5ELSE PUMP P1 STATUS = OFF
Single event LID analysis: detention pond sizing
Design storm analysis: pond sizing no LIDs
Design storm analysis: pond sizing with LIDs
PCSWMM/SWMM5 LID modeling
• Physically-based processes
• Flexible components
• SWMM Version 5.0.022
• Simplifies the modeling of:– Lot level implementations
– Watershed scale or city master planning
– Single event
– Continuous
– Performance degradation