HYDRO 2013_Pradyumna

© Lahmeyer International (India) Pvt. Ltd.© Lahmeyer International (India) Pvt. Ltd.
One-dimensional sediment modelling for Chuzachen
and Devsari hydroelectric power projects to check the
feasibility of reservoirs’ usage as pseudo-desanders
Author and Presenter : Pradyumna Machhkhand
Senior Manager (Hydropower & Water Resources)| Lahmeyer International (India) Pvt. Ltd. | Gurgaon-122002
1

© Lahmeyer International (India) Pvt. Ltd.
Outline of the presentation
2
Introduction
Background
Information
Reservoir
Sedimentation
Sediment
Modelling
Theory
Case Studies
Modelling
Scheme
Model Set Up
Data
Organization
Results
Comparison
Conclusion
Model
Description

• Reservoir sedimentation in hydroelectric projects (HEP): a major problem
that interrupts the smooth functioning of hydropower plants.
A schematic and classical illustration of reservoir sedimentation
Introduction
3

• In general, the problem is seen as a big hindrance and therefore, as a
remedial measure – “n” number of desanders, based on studies, are usually
proposed to manage the sediment removal operation.
• The presented case studies foray into the primary investigation of reservoir
sedimentation prior to making decisions on sediment management.
• The investigation of sediments lies on two basic searches:
Where & How Much
Introduction
4
To examine the location and capacity
of the reservoir in discrete form
To understand the relation between sediment
concentration and discharge in the reservoir

• Two different hydroelectric projects have been studied:
a. Devsari HEP (Devsari Reservoir)
b. Chuzachen HEP (Rangpo Reservoir & Rongli Reservoir)
• Project features:
contd...
Background information
5
Project features Devsari HEP
Chuzachen HEP
(Rangpo)
Location Uttarakhand, India Sikkim, India

Background information
6
Project features Devsari HEP
Chuzachen HEP
(Rangpo)
Type of structure Dam Dam
Height of structure 35m 48m
Extent of reservoir 4.8Kms from dam axis 508m from dam axis
Storage 9.026 Mm
3
0.360 Mm
3
…contd

• Empirical methods
a. Advantage : explicit methods (such as, Hazen, Vetter, and Camp), computational
ease while computing the sediment trap efficiency.
b. Limitations :
i. cover only the settling phase of reservoir operation,
ii. and other factors which are NOT included in the empirical methods are:
 the sediment transporting capacity of flow in a settling reach,
 the change in conditions as reservoir fills with sediment,
 the effect of variation in flow depth down a basin or reservoir,
 and the additional turbulence caused by inlet condition to a basin or
reservoir. contd...
Reservoir sedimentation
7

• 1-dimensional numerical model
a. Advantages :
 models the effect of turbulence in sediment movement and mass
deposition,
 not only simulates deposition, but also sluicing,
 does not require grids to approximate cross-sections,
 and requires less field data to set up.
8
…contd

• 1-dimensional numerical model
b. Limitations:
 does not offer a detailed view of hydrodynamics of reservoir
system like 2-D or 3-D models.
Although 2-D or 3-D models have certain advantages, but at the cost of a
longer computational time and substantial amount of field data to
capture the complexities of 2-D or 3-D flow.
contd...
9

• SHARC , a software developed by HR Wallingford, UK, is a suite of integrated
programs designed to assist in the identification and solution of sediment
problems at intakes in rivers and canal systems.
• DOSSBAS stands for Design of Sluiced Settling Basins.
• DOSSBAS is tool built within SHARC to model sediment depositions in
basins/reservoirs and can model both regular and irregular basins.
• Two suites of DOSSBAS: 1) Deposition Model. 2) Sluicing Model.
Model description : DOSSBAS tool of SHARC
10
contd…

• Deposition model
 Assumptions:
o the flow is steady,
o the velocities and concentrations are constant across the width of the
channel,
o and the concentrations in one size fraction do not affect other size
fraction.
11
…contd

• Deposition model [Basic equation of turbulence (Dobbins ,1994)]
…Equation (1)
= sediment diffusion coefficient in y-direction (m2/sec) ,
= sediment diffusion coefficient in x-direction (m2/sec),
= and settling velocity (formula by Gibbs et al, 1971) of sediment for the sediment
size fraction (j) (m/sec).
12
…contd
2
2
2
2
x
C
y
C
y
V
y
C
x
C
u
j
x
jy
sj
j
y
j



















where, u = flow velocity at height y above the bed (m/sec),
y = height above bed (m),
Cj = sediment concentration at height (y) above bed for size fraction( j),
x = distance co-ordinate along channel (m),
y
x
sjV

• Deposition model
13
…contd

Deposition model : Turbulent diffussion within a sub-reach
14

•
Deposition model : Boundary conditions
15

Overall structure
of deposition model
16

• Sluicing model
 Sluicing model simulates only sand movements. It is assumed that any silt in the
exposed bed material is sluiced instantly and therefore, only sand transport controls
the sluicing rates.
 Diffusion is not a dominant process in sluicing. Sluicing is modelled using equation,
Note: The threshold value to differentiate between silt and sand in DOSSBAS is 63 micron (default).
17
…contd

Overall structure
of sluicing model
18

Modelling scheme of reservoir sediment management plan
Modelling scheme
19
contd…

Modelling scheme
20
…contd
• Set of parameters
 X1 is the set of input parameters, viz., reservoir geometry, and sediment data
 X2 is the set of parameters that includes the temperature, and the sand concentrations to
be applied in both deposition and sluicing model.
 The input parameters for deposition model can be expressed as, (X1 ∪ X2).
 The set X3 is composed of additional parameters such as, flushing/sluicing discharge, the
water level at the time of sluicing, and the duration of sluicing.
 Y1 is the set of longitudinal bed profiles, which are results of deposition models.
 The input parameters for sluicing model can be expressed as, (Y2 ∪ X2 ∪ X3).
 The set Y2 comprises results of sluicing models in terms of elapsed time for flushing out
the sediments and change in bed profiles at different time.

• Hydrological year mean flow duration curves
Data organization
21
contd…
0
20
40
60
80
100
120
140
160
180
200
220
240
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Discharge(m3/sec)
Exceedanceprobability (%)
Flow duration curve divisions: Devsari HEP
Flow duration curve
Qdesign (Intake) = 120.76 m3/sec
Division3 Qm = 44.8 m3/sec
Division2 Qm =160.1 m3/sec
Division1 Qm = 205.3 m3/sec
0
20
40
60
80
100
120
140
160
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Discharge(cumec)
Exceedance Probability(%)
Flow duration curve divisions : Rangpo
Flow duration curve
Qm =33.2 m3/sec (Division 3)
Qm=95.8 m3/sec (Division 1)
Qm=58.1 m3/sec (Division 2)
Qm=8.4 m3/sec (Division4)
Qdesign =21.5 m3/sec

• Pre-processed data: average discharge and mean concentration
Data organization
22
contd…
Divisions
Devsari Rangpo
Qm Mean
conc.
Qm Mean
conc.
m3/s ppm m3/s ppm
1 205.3 511 95.8 707
2 160.1 436 58.1 389
3 44.8 192 33.2 187
4 - - 8.4 47

• Optimization of input parameters
Data organization
23
contd…
y= 16.54x0.6445
R² = 0.6115
0
100
200
300
400
500
600
700
800
900
0 50 100 150 200 250
MeanConcentration[ppm]
MeanDischarge [m3/s]
Mean discharge vs mean concentration
Obs.Mean…
Power (Obs.Mean …
 Devsari Reservoir
 6445.0
54.16
1
xCm



• Optimization of input parameters
Data organization
24
…contd
 Rangpo Reservoir
o from the table of the pre-processed data, the annual sediment volume of
Rangpo is about 70% the capacity of the Reservoir, and therefore, it is quite
conservative to adopt the same values.

• Bed gradation curves
Model setup
25
contd…

• Input parameters
Model setup
26
…contd
Divisions Flow
Sediment
concentration
Deposition
cycle
time
Flushing
discharge
Silt Sand
m3/sec ppm ppm Days Hours m3/sec
1 205.3 465 199 37 876 84.5
2 160.1 397 170 44 1051 39.3
3 44.4 174 74 285 6833 NA
Divisions
Flow
Sediment
Concentration
Deposition
cycle
time
Flushing
discharge
Silt Sand
m3/sec ppm ppm Days Hours m3/sec
1 95.8 495 212 37 876 74.3
2 58.1 273 117 37 876 36.6
3 33.2 131 56 73 1752 11.7
4 8.4 33 14 219 5256 NA
Devsari Rangpo

• Input parameters
Model setup
27
…contd

• Deposition model: Devsari
Results
28
1265
1270
1275
1280
1285
1290
1295
1300
1305
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Distance downstream (m)
Minimum bed level (m)
Bed levels attime 0.00 (hours)
Final water level
Longitudinal Profile Down Basin
Elevation(m)
Divisions
Sediment
Volume
Outflow
Sediment
Trap
Efficiency
Reservoir
Storage
Silt Sand Silt Sand Silt Sand
Befor
e run
After
run
Mm3 Mm3 PPM PPM % % Mm3 Mm3
1 0.1 0.1 215 0 53.7 100.0 9.1 8.9
2 0.1 0.1 164 0 58.6 100.0 9.1 8.9
3 0.1 0.1 35 0 79.8 100.0 9.1 9.0
contd…

• Sluicing model: Devsari
1265
1270
1275
1280
1285
1290
1295
1300
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Minimum bed lev el (m)
Bed lev els at time 0.000 (hours )
Elevation(m)
Results
29
contd……contd
Divisions
Volume removed Elapsed
time
for
flushing
Silt Sand
Total
sediment
Mm3 Mm3 Mm3 Hours
1 0.1 0.1 0.2 14.44
2 0.1 0.1 0.2 28.59
3 NA

• Deposition model: Rangpo
Results
30
contd…
880
885
890
895
900
905
910
915
0 100 200 300 400 500 600
Minimum bed level (m)
Final water level
Elevation(m)
Divisions
Sediment
Volume
Outflow
Sediment
Trap
Efficiency
Reservoir
Storage
Silt Sand Silt Sand Silt Sand
Before
run
After
run
Mm3 Mm3 PPM PPM % % Mm3 Mm3
1 0.013 0.045 350.0 5.0 12.6 97.5 0.36 0.31
2 0.009 0.015 217.0 1.0 20.4 99.3 0.36 0.34
3 0.007 0.008 89.0 0.0 31.7 100.0 0.36 0.35
4 0.003 0.002 14.0 0.0 58.8 100.0 0.36 0.36

• Sluicing model: Rangpo
Results
31
contd……contd
880
885
890
895
900
905
910
0 100 200 300 400 500 600
Minimum bed lev el (m)
Elevation(m)
Divisions
Volume removed Elapsed
time
for
flushing
Silt Sand
Total
sediment
Mm3 Mm3 Mm3 Hours
1 0.014 0.043 0.06 0.41
2 0.009 0.015 0.02 0.35
3 0.005 0.008 0.01 0.79
4 NA

• Velocities and required length of reservoirs to settle 0.2mm particle:
Comparison with empirical method
32
Devsari Rangpo Units
Full reservoir level (FRL) = 1300 910 MASL
Minimum drawdown level(MDDL) = 1295 893 MASL
Division1 average discharge (Qa) = 205 90 m3/sec
The approximate average area of flow (A) = 4700 1978 m2
The corresponding flow through velocity (V) = Qa/A = 0.044 0.046 m/sec
Mean flow through velocity for 0.2mm particle to
settle down, v’ = 0.2 0.2 m/sec
Settling velocity of 0.2mm particle, w = 0.022 0.022 m/sec
Length of reservoir required to settle 0.2mm particle,
L*= v’/w x ( FRL-MDDL) = 45.5 154.5 m
Mid-Length from dam axis of reservoir, L = 2500 254 m
…contd

• It can be safely inferred from the model results that the sediment grain size
of 0.2mm and greater than 0.2mm shall be settling down in the reservoirs.
• If periodical flushing operations are realized during the whole life of the
reservoir, the entrance sill of the intake will not be affected by the deposited
sediments; as it is already being witnessed in the model studies.
• The model studies have potential scopes to enhance the analysis; the
sensitivity analysis of the models is one of them.
Conclusion
33
contd…

• It would be in the interest of HEPs, if the utility of the desanders decelerates
before placing them in the planning stages, thereby encouraging physically-
based hydraulic modelling study of the sediments in the interim to assess
whether the reservoirs are self-sufficient for managing the sediment
processes or not.
• The case studies of this paper offer supporting results to the usage of
reservoirs for sediment management, but since the rapidification of
desanders are commonly observed in HEPs; despite the reservoirs being
gigantic in size, the term “pseudo-desanders”, used in the title of the paper,
for reservoirs appears reasonable.
Conclusion
34
…contd

• This work would have been due without the moral support of my wife,
Imane Ibnoussina. The timely information and encouragement offered by
Dr. S.K.Mazumder is highly appreciated; my sincere regards to him and his
family. The intensity of the work is recognized and is supported for
presentation by Dr. A.K.Jha. The encouragement offered by Lahmeyer
International (India) Pvt Ltd. is highly appreciated.
• http://r4d.dfid.gov.uk/Output/5145/
Acknowledgements
35

HYDRO 2013_Pradyumna

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Viewers also liked

Viewers also liked (13)

Similar to HYDRO 2013_Pradyumna

Similar to HYDRO 2013_Pradyumna (20)

HYDRO 2013_Pradyumna