Upload
others
View
6
Download
0
Embed Size (px)
Citation preview
25-05-2018
1
Training for Asian Young Professionals on
Performance Assessment of Irrigation Systems
( 9th to 13th April 2018 in Beijing, China)
Structural interventions in modernization of irrigation systems-
Canal lining, Long crested weir, Baffle distributor, Measuring devices etc.
By
M G Shivakumar Chief Engineer
VJNL, UBP Zone, ChitradurgaWater Resources Department , Karnataka, India
Contents
• What are structural interventions?
• Canal lining
• Long Crested Weir
• Baffle distributor
• Measuring device
• Other structures (offtakes, cross regulators,escapes etc)
LINING IRRIGATION
CANALS
CANAL LINING - TO LINE or NOT TO LINE
• Canal lining is a very expensive element in canal construction - Cost of lining typically represents about 40 percent of the total cost.
• Before making such a large investment, there must be a clear idea of the benefits to be obtained
• The reduction of seepage losses is often assumed to be constant for the expected life of the lining to have a chance of achieving a favorable economic return
Issues of Canal Lining : Cost/Benefit
25-05-2018
2
• Seepage losses typically represent 10 to 40 percent of diverted water
• Sedimentation reduces seepage
• Within 6 years of canal lining, losses return to around 60 % of the initial seepage (WB)
Issues of Canal Lining: Seepage losses
Benefits of Canal Lining
• reduces seepage losses
• improves canal hydraulics
• reduces water-logging and salinization
• reduces weed growth
• improves equity and reliability of water
distribution?
• reduces maintenance (?)
Benefits of Canal Lining: Reduction of
weed growth
this benefit is questionable
in some projects with old
lined canals and poor
construction joints
Quality of construction and concrete
Saturation of the soils after
commissioning of the canal
may result:
– cracks in the
slabs
– opportunity for
seepage losses
under most of the
slab area
25-05-2018
3
Estimates of Seepage Losses:
a source of doubt and concern
• Difficult to compare the results of tests and
estimates because of number of variables
• Checking seepage losses independently of
operation records are expensive
• Published records often fail to record key
variables
Seepage Variables
• characteristics of the soils:
– porosity, permeability,
– chemistry, granulometry,
– stratigraphy
• geometry of canals
• position of water table
• chemistry of water and soils
• sedimentation: self sealing effect
Seepage Variables (cont.)
• quality of construction
• age of the canals
• cycles of filling/draining
• maintenance
Seepage Values: Pakistan
• Normally assumed 8 cfs/million ft2 (0.23m/day)
• Estimates of average losses up to outlets: 20 to
25 % of inflows at head of main canals
• range of measurement values: 0.15 to 0.30 m/day
(0.06 to 0.57m/day)
• ISRIP and IWMI measured 15 to 25 % average
losses from the main and secondary system –
CRBC, Fordwah, Eastern Sadiqia (1990-1995)
25-05-2018
4
Measurements using ponding method
(Singh, 1987)
•unlined canal 0.3 m/day
•lined canal 0.03 m/day
These data come from controlled tests in
laboratories and experimental sections with
excellent maintenance
INDIA: PUNJAB
brick lined canals
•Sunam branch ( four-month old)
0.06 m/day
•Mudki distributary (15 km, 5-year old)
0.29 m/day
•Mukstar distributary (6-year old)
0.49 m/day
Watercourses (24): seepage losses from channel
older than 4 years are comparable to losses from
unlined channels
Canal Lining: A way to modernize the
irrigation system?
• it is generally advocated by Irrigation Agencies to
reduce seepage
• there are obstacles to canal lining : e.g it
interrupts irrigation services
Line or not to Line: The Debate
• Since seepage recharges the
groundwater, overall water
saving could be marginal
• Cost of pumping
• Deterioration of water
quality, Waterlogging and
salinization of adjacent
lands
• Farmers use groundwater
more efficiently
25-05-2018
5
THE TECHNICAL CASE FOR CANAL
LINING IS NOT STRAIGHTFORWARD
According to the WB-Experts
•Canal lining with 1% crack area has a seepage rate
of 70% of that for unlined conditions (depth to
water table: 8m)
•There is now strong evidence that hard surface
linings deteriorate within a few years until seepage
losses return to that for an unlined canal
THE TECHNICAL CASE FOR CANAL
LINING IS NOT STRAIGHTFORWARD
1% crack area = 70% seepage of unlined conditions (depth
to water table: 8m) HOW POSSIBLE?
Water tableWater table
Unlined Cracked Lining
Answer lies in thewetted perimeter
Head makes thin concrete cracks watering a much larger strip behind the concrete layer.
Canal Lining: Questions ?
• Do we really need lined canals to improve water delivery service to water users?
• Do we want to carry out this very expensive option and risk losing the valuable money that could be spent more effectively on other interventions?
Canal Lining: Options
• Avoid systematic canal lining as a norm !
• Serious documentation for lining to be done !
• Wherever canal lining is justified/needed use only
high standards techniques [geo-membrane,
reinforced concrete]
• Adjust operation and maintenance to the needs of
lined canals.
25-05-2018
6
Long Crested Weir - Design
25-05-2018
7
Variables? Length - Height
Provision for flushing !
h
Q= c. Lc. [h]3/2
(c = 1 to 1.9)
LCW= Overshot structure
25-05-2018
8
Exercise: Crest length of a Long crested weir ?
Q= c. Lc. [h]3/2
Objective of control
Q1 ➔ H1
Q2 ➔ H2
Variation H= H2-H1
2/3
21
3/22
3/21
Chh
cQ
cQ
L
−
−
=
Crest lenth variation for Q=2 to 5 m3/s
0
10
20
30
40
50
60
70
80
90
100
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45
Water level variation
Cre
st
len
gth
Crest length of a Long Crested weir ?
H1-H2= 0.15 m
Then Length crest = 17 m
Objective: Ensure that flow is not submerged for Q max!
Height of a Long Crested weir
Partially submerged Drop is insufficient
Crest should be slightly higher than normal flow to allow a non submerged flow on the crest
Height of a Long Crested weir
25-05-2018
9
H1
H crest = H normal + ∆hcrestH flow = H crest+ H1Head losses = H1+ ∆hcrest
Head losses
H normal
Head losses
∆hcrest
∆hcrest ?0.10 0.15 m
Backwater effect = Head losses/slope
Good Design for flushing
Gate closed (periodic flushing)Part of Q2 through the gate (permanent flushing)
Gate
Significant reduction of Crest Length
Part of ∆Q (Q1-Q2) is taken care by periodic gate adjustments thus H1-H2 is minimum and Length crest also.
Advantage of Mixed CRAdapting existing CR as Mixed Regulator with the construction of a Weir to control the water level. Advantage: no need to have a permanent operator
Mixed CR: a Very good solution for improvement !
25-05-2018
10
25-05-2018
11
Canal Flow Measurement
25-05-2018
12
Basic Options1. Weirs
2. Flumes
3. Electronic devices
4. Rated sections
Weirs and flumes are based on Critical Flow Devices
For one depth, there is one corresponding flow rate
Critical Flow Devices
For one depth, there is one corresponding flow rate
0
1 0 0
2 0 0
3 0 0
4 0 0
5 0 0
6 0 0
7 0 0
5 6 7 8 91
0
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
H e ig h t c m
Dis
ch
arg
e i
n l
/s
1. Weirs
25-05-2018
13
90°H
1
4 H
L
H
L
TRIANGULAR (90° V-NOTCH) TRAPEZOIDAL (CIPOLLETTI) RECTANGULAR
Types of Weirs
Head
Staff Gauge
Weir Crest
Free Flow
Note: Air pocket behind crest
Weir Crest Note: Downstream water
level above crest
Submerged
2. Flumes
25-05-2018
14
Weir
Flume
Y1 Y2 Headloss = Y2 - Y1
Head
Headloss
Diff. Weir FlumesParshall Flumes
Replogle Flumes (RBC)
TADLA MoroccoPK Albania
TADLA Morocco
25-05-2018
15
6:1 downstream ramp
6:1 downstream truncated ramp
FLOW
FLOW
Replogle Flume - Ramp Recommendation
4-in Diameter Pipe
Install flush to concrete on both ends and flush to the bottom of the canal.
Replogle Flume - Flush Pipe
25-05-2018
16
25-05-2018
17
WinFlumeUSBR-Denver www.usbr.gov/pmts/hydraulics_lab/winflume
Flow Measurements
Accuracy !!!
HH
QQ = H=Q
A key issue:
Accuracy of gauge reading ∆H
25-05-2018
18
Accuracy of gauge reading:turbulence and waves
Higher the Head higher the accuracy
HH
QQ =
25-05-2018
19
Accuracy of measurement
inversely proportional to flow width
TOO WIDEHead is low particularly at low discharge
Head max = 0.60 mAccuracy at Q max=250 l/s for 2cm
Turbulence
HH
QQ =
Accuracy of measurement
inversely proportional to flow width
OK
0
1 0 0
2 0 0
3 0 0
4 0 0
5 0 0
6 0 0
7 0 0
5 6 7 8 91
0
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
H e ig h t c m
Dis
ch
arg
e i
n l
/s
153
194
421
481
25-05-2018
20
Accuracy of weir measurements
increases with available head
High Head ==> low width
Low head ==> high width
Accuracy of weir measurements
decreases with width of flow
3. Velocity sensorDoppler
25-05-2018
21
25-05-2018
22
4. Rated sections
Using rated sections
No measurement without periodic
calibration
THANK YOU