Hydropower Engineering: Energy and Power Analysis Using Flow Duration Approach
Hydropower Engineering: Energy and Power Analysis Using Flow Duration Approach
Hydropower Engineering: Energy and Power Analysis Using Flow Duration Approach
Chapter 3
Energy and Power Analysis Using Flow Duration Approach
CONTENTS:
1. Power duration curve
2. Load terminologies
CENG 6803
3. Load duration curve
2
REMEMBER FROM SECTION 2.3
Flow Duration Curves: is a plot of flow (Q) versus the percent
of time a particular flow can be expected to be exceeded.
CENG 6803
140
120
100
Discharge in m3/s
80
60
40
20
0
0 20 40 60 80 100
Exceedence in % 3
CHARACTERISTICS OF FLOW DURATION CURVES
The flow duration curve (FDC) shows how flow is distributed over a period (usually
a year).
A steep flow duration curve implies a flashy catchment – one which is subject to
extreme floods and droughts.
CENG 6803
Factors which cause a catchment to be flashy are:
Rocky, shallow soil,
Lack of vegetation cover,
Steep, short streams,
Uneven rainfall (frequent storms, long dry periods).
Such type of FDC (i.e. steep) is bad for hydropower development (especially run-of-
river type).
A flat flow duration curve is good because it means that the total annual
flow will be spread more evenly over the year, giving a useful flow for
4
longer periods, and less severe floods.
The selection of the time interval for FDC depends on the purpose of the
study.
As the time interval increases the range of the curve decreases (see Fig.).
While daily flow rates of small storms are useful for the pondage studies in
a run-off river power development plant, monthly flow rates for a number
of years are useful in power development plants from a large storage
CENG 6803
reservoir
CENG 6803
6
Flow duration curve, very often, plotted using the average monthly
values of the flow.
The capacity estimate for firm power is then made by using the entire
recorded flow data and plotting in a single flow duration curve.
CENG 6803
In such a case two different methods are in use.
(i) the total period method, and
(ii) the calendar year method.
Both methods utilize the flow data available for the entire period for
which records are available.
7
Total period method: the entire available record is used for drawing the
FDC. Thus, ten years’ record would produce 120 values of monthly average
flows.
These are first tabulated in the ascending order starting from the driest
month in the entire period and ending with the wettest month of the ten-
year duration.
The FDC would then be drawn with the help of 120 values.
CENG 6803
Calendar year method: each year’s average monthly values are first
arranged in ascending order.
Then the average flow values corresponding to the driest month, second
driest month, and so on up to the wettest month are found out by taking
arithmetic mean of all values of the same rank. These average values are
then used for plotting flow duration curve.
Such a curve would have only twelve points.
The total period method gives more correct results than the calendar year
method which averages out extreme events.
8
DISCHARGE CAPACITY OF A PLANT
Discharge capacity (Qc) of a plant is the discharge the plant
can pass at its full gate opening of the runner(s) of the
turbine(s) under design head.
CENG 6803
140 • A flow duration curve is
120
used to explain discharge
capacity (Qc) as labeled in
100
the Figure. Even though to
Discharge in m3/s
Exceedence in %
WATER POWER POTENTIAL
Before any power plant is contemplated, it is essential to assess the
inherent power available from the discharge of the river and the head
available at the site.
The gross head of any proposed scheme can be assessed by simple
CENG 6803
surveying techniques, whereas
Hydrological data on rainfall and runoff are essential in order to
assess the quantity of water available.
The hydrological data necessary for potential assessment are:
The daily, weekly, or monthly flow over a period of several
10
3.1 POWER DURATION CURVE
Remember: From section 2.1 P w Qgh
The above equation is for theoretical conditions. The
CENG 6803
actual output is diminished by the fact that the turbine
has losses in transforming the potential and kinetic
energy into mechanical energy. Thus an efficiency term
(ɳ), usually called overall efficiency, must be introduced
to give the standard power equation:
P w Qgh
If hydraulic head and the expected losses in the penstock
are known, it is possible to generate a power duration
curve from the flow duration curve. How?
11
If the river course is divided into a number of stretches, the total
power can be described by
P w g Qh
• The actual use of the equation for estimating the potential (P);
however, is made difficult due to the fact that the discharge of any
river varies over a wide range.
CENG 6803
• High discharges are available only for short durations in a year. Thus
the corresponding available power would be of short duration.
• If the discharge rate and the percentage duration of time for which it is
available are plotted, a flow-duration curve results.
• Power duration curve can also be plotted since power is directly
proportional to the discharge and available head.
• Discharge/Power duration curve indicates discharge or power
available in the stream for the given percentage of time.
12
The available power from a run-of-river plant could be
represented by a power duration curve exactly on lines analogous
to a FDC;
Generally, the head variation in a run-of-river plant is
considerably less than the discharge variation.
CENG 6803
If the head is presumed to be constant at an average value, power
duration curve would exactly correspond to FDC.
This is very often the procedure in elementary rough calculations.
If, however, a precise power duration curve is desired, then the
head corresponding to any discharge is required to be known.
13
P50 Pm P95 P100
Power
CENG 6803
0 50 95 100
Percentage of time equaled or exceeded
Minimum potential power computed from the minimum flow available
for 100 % of the time (365 days or 8760 hours). This is represented as
P100;
Small potential power computed from the flow available for 95 % of
time (flow available for 8322 hours). This is represented as P95; 14
Average potential power computed from the flow available for 50%
of the time (flow available for 6 months or 4380 hours). This is
represented as P50;
Mean potential power computed from the average of mean yearly
flows for a period of 10 to 30 years, which is equal to the area of
the flow-duration curve corresponding to this mean year. This is
CENG 6803
known as ‘Gross river power potential’ and is represented as Pm.
It would be more significant to find out the technically available
power from the potential power; According to Mosonyi, the losses
subtracted from the P values present an upper limit of utilization;
Technically available power: With conveyance efficiency of 70%
and overall efficiency of the plant as 80%, a combined multiplying
factor of 0.56 should be used with the average potential power, P50;
15
Pa 0.56P50
The value of net water power capable of being developed
technically is also computed from the potential water power
by certain reduction factors to account for losses of head in
the conveyance and losses associated with energy
conversion. EEC puts this factor to be about 0.75 or 0.80, i.e
CENG 6803
7.4 discharge
Pm net mean
Where Qm = the arithmetic to 8.0Qm h
The maximum river energy potential is given as
16
EXAMPLE 1: WHAT IS WRONG WITH THE PDC?
FDC Energy production
140
120
for a year or a time
period is the product
Discharge in m3/s
100
80
CENG 6803
60 of the power ordinate
40
20 Qc and time and is thus
0
0 20 40 60 80 100 the area under the
120 Exceedence in %
power duration curve
100
multiplied by an
80
appropriate
Power in MW
60
conversion factor.
40
20 Pc
0
0 20 40 60 80 100
Exceedence in % 17
Example 2. The following is the record of average yearly flow in a river for
15 years. If the available head is 15 m, construct the FDC and power
duration curve for the river.
Year 1956 1957 1958 1959 1960 1961 1962 1963 1964
CENG 6803
Flow
(m3/s) 905 865 1050 1105 675 715 850 775 590
Solution
The yearly flow values are arranged in ascending order (see
table below). The power corresponding to each flow values
are calculated assuming the head (=15 m) to be constant.
Then, FDC and power duration curves are plotted on the 18
same graph.
Rank Flow in ascending Power (=9.81 QH) Percentage of time exceeded
order (m3/s) [kW] =(15 +1-n)*100/15
CENG 6803
4 715 105212 80.0
5 775 114041 73.4
6 810 119192 66.7
7 850 125078 60.0
8 865 127285 53.4
9 885 130228 46.7
10 905 133171 40.0
11 925 136114 33.3
12 1025 150829 26.7
13 1050 154508 20.0
14 1105 162601 13.3
15 1150 169223 6.7
19
1200 180000
Flow Power
170000
1100
160000
1000
CENG 6803
150000
140000
900
Flow, m3/s
130000
Power, kW
800
120000
700 110000
100000
600
90000
500 80000
0.0 20.0 40.0 60.0 80.0 100.0
20
% of time equalled or exceeded
EXERCISE-1: DRAW THE FLOW AND POWER
DURATION CURVE…
Month Flow (m3/s) Head(m) Efficiency
January 65 83.5 0.87
February 50 83.5 0.83
CENG 6803
March 42 83.5 0.75
April 40 83.5 0.70
May 40 83.5 0.60 Qc = 270 m3/s
June 115 83.5 0.50 ρ= 1000 kg/m3
July 400 80 0.88 g= 9.81 m/s2
August 340 81.6 0.89
September 270 83 0.90
October 155 83.5 0.90
November 115 83.5 0.88
21
December 85 83.5 0.87
EXAMPLE 3:
A turbine installation developing 7.5 MW under 27.5 m head
with an overall efficiency of 0.83 is to be supplied from a
reservoir. The estimated runoff for 12 consecutive months
CENG 6803
each 30 days (in Million Cusecs) were: 96.2, 101.8, 86.3, 74.9,
67.9, 80.6, 113.2, 90.5, 86.3, 93.4, 99.0, and 89.1.
Assume the reservoir is full at the beginning of the first
month.
Determine the minimum capacity of the reservoir to ensure
the required demand and
Find the discharge wasted
22
SOLUTION
Q Demand Surplus Inflow to Deficit Wastage Total
(106 m3/ (106 m3/ (106 m3/ resevoir (106 m3/ (106 m3/ Wastage
month) month) month) month) month)
CENG 6803
86.3 86.8 -0.5 -0.5 0.5 0 24.4
74.9 86.8 -11.9 -11.9 12.4 0 24.4
67.9 86.8 -18.9 -18.9 31.3 0 24.4
80.6 86.8 -6.2 -6.2 37.5 0 24.4
113.2 86.8 26.4 26.4 11.1 0 24.4
90.5 86.8 3.7 3.7 7.4 0 24.4
86.3 86.8 -0.5 -0.5 7.9 0 24.4
93.4 86.8 6.6 6.6 1.3 0 24.4
99 86.8 12.2 1.3 0 10.9 35.3 23
89.1 86.8 2.3 0 0 2.3 37.6
3.2 LOAD TERMINOLOGIES
Load is the amount of power delivered or received at a given point at any instant.
Average Load is the total load produced divided by the number of hours in the
time period of interest.
Peak Load is the maximum instantaneous load or a maximum average load over
a specified period of time.
CENG 6803
Base load is the total load continuously exceeded;
CENG 6803
The load factor measures variation only and does not give any
indication of the precise shape of the load-duration curve.
The area under the load curve represents the energy consumed
in kWh; Thus, a daily load factor may also be defined as the
ratio of the actual energy consumed during 24 hours to the
peak demand assumed to continue for 24 hours.
Load factor gives an idea of degree of utilization of capacity;
Load
L shape, indicating a peak load of very
short duration with very low or no load
CENG 6803
during the major portion of the time.
As the load factor approaches unity,
Time
the duration curve will be somewhat
rectangular in appearance, indicating
high sustained loads.
Load
26
Time
CAPACITY FACTOR
The capacity factor is the ratio of the energy actually produced by
the plant for any given period of time to the energy it would be
capable of producing at its full capacity for that period of time.
The extent of use of the generating plant is measured by the
CENG 6803
capacity factor, frequently also termed plant factor or use factor. If
during a given period a plant is kept fully loaded, it is evident that it
is used to the maximum extent, or operated at 100% capacity factor.
The factor is equal to the average load divided by the rated capacity
of the plant.
Capacity factor and load factor become identical when the peak
load is equal to the capacity of the plant. The relationship between
the two factors is evidently
CENG 6803
The capacity factor for hydroelectric plants generally varies
between 0.25 and 0.75.
The capacity factor would be identical with load factor if the
peak load were equal to the plant capacity.
Thus, in the above example, if the maximum load was 75 MW
instead of 100 MW, then
6000000
L.F . 0.8 againest C.F . 0.6
75000100
28
UTILIZATION FACTOR
The utilization factor measures the use made of the total installed
capacity of the plant. It is defined as the ratio of the peak load and the
rated capacity of the plant.
Utilization Factor: is the ratio of the quantity of water actually utilized
for power production to that available in the river. If the head is assumed
CENG 6803
to be constant, then the utilization factor would be equal to the ratio of
power utilized to that available.
The factor for a plant depends upon the type of system of which it is a
part of. A low utilization factor may mean that the plant is used only for
stand-by purposes on a system comprised of several stations or that
capacity has been installed well in advance of need.
In the case of a plant in a large system, high utilization factor indicates
that the plant is probably the most efficient in the system. In the case of
isolated plants a high value means the likelihood of good design with
some reserve-capacity allowance.
The value of utilization factor varies between 0.4 and 0.9 depending on 29
the plant capacity, load factor and storage.
EXAMPLE 4
Consider the yearly load duration curve for a certain load
center to be a straight line from 20 to 3 MW. To meet
this load, three hydropower units, two rated at 10 MW
each and one at 5 MW are installed. Determine:
CENG 6803
Load factor (LF) MW
Capacity factor (CF) Solution
Utilization factor (UF) 20
3 20
100 % t
LF 2 100 % 57 . 5 %
20 100 % t 3
3 20
100 % t
2 100
CF 100 % 46 % Time (%)
(10 10 5 ) 100 % t
20 100 % t 30
UF 100 % 80 %
25 100 % t
DIVERSITY FACTOR
Diversity factor (DF) is the summation of the different types of
load divided by the peak load.
If there be four different types of load L1, L2, L3 and L4 and the
peak load from the combination of these loads is LP, then the
CENG 6803
diversity factor is expressed as:
(L1 + L2 + L3 + L4)/LP
Note that the diversity factor has a value
i n which is greater than
unity. DF L i L p
For n load combination: i 1
CENG 6803
Domestic - 3 MW
Irrigation - 5 MW
Water supply – 2 MW
Industries - ? Lp
Offices - 2 MW
Public utilities - 4 MW
CENG 6803
Peak Load
Average Load
Load[MW]
Base Load
0 6 12 18 24 33
Time
The area under the curve of a daily chronological load curve
measures the total energy consumed by the load during the day. This
energy is evaluated by:
t 24
CENG 6803
E kwdt
t 1
Street light?
Industrial load with one
shift operation Residential load
Load
Load
Time 34
Time 6 12 18 24
6 12 18 24
It will be necessary for system planning and operating estimates to express
the variation in, and the integration of, the total energy requirements for a
period of time in some concise form; the load-duration curve does this.
Firm Power: The firm or primary power is the power which is always
ensured to a consumer at any hour of the day and is, thus, completely
dependable power. Firm power would correspond to the minimum stream
CENG 6803
flow and is available for all the times;
35
Fundamentally the load-duration curve is nothing more than a
rearrangement of all the load elements of a chronological curve in
the order of descending magnitude. The areas under the load-
duration and corresponding chronological curves are equal. Since it
is impractical to determine the equation of load curve, the area or
energy is determined graphically.
CENG 6803
It is usually prepared for a longer duration such as a year.
The area under load duration curve for a time duration would be the
same as that of a load curve for the same particular period of time.
The area under a load duration curve represents the total energy
production for the duration. Thus, annual load factor is given by the
ratio of the area under the curve to the area of the rectangle
corresponding to the maximum demand occurring during the course
of the year.
36
The firm power could be increased by the use of pondage (storage).
CENG 6803
Curve with storage
Discharge
Primary power
Primary power without storage with storage
0 100
Percentage of time
CENG 6803
Primary power
CENG 6803
Operating head = 15 m
Plant efficiency = 80%
Estimate the daily load factor of the plant if the stream flow is 15
m3/s.
39
EXERCISE 4
A common load is shared by two stations, one being a base load
plant with 25 MW installed capacity and other being a standby
station with 30 MW capacity. The yearly output of the standby
station is 10.5 x 106 kWh and that of the base load plant is 125 x
106 kWh. The peak load taken by the standby station is 15 MW
CENG 6803
and this station works for 2500 hours during the year. The base
load station takes a peak of 22.5 MW. Find out:
(i) Annual load factor for both stations;
(ii) Capacity factor for both stations.
40
EXAMPLE 5
The following data are obtained from the records of the mean monthly
flows of a river for 10 years. The head available at the site of the power
plant is 60 m and the plant efficiency is 80%.
Mean monthly No. of
flow range occurrences
1. Plot the FDC and PDC
(m3/s) (in 10-yr period) 2. Determine the mean
CENG 6803
3
monthly flow that can be
100-149
expected and the average
4
150-199 power that can be developed.
200-249 16 3. Indicate the effect of storage
21 on the FDC obtained.
250-299
4. What would be the trend of
300-349 24
the curve if the mean weekly
350-399 21 flow data are used instead of
400-449 20 monthly flows?
450-499 9
500-549 2 41
Solution
1. The mean monthly flow ranges are arranged in the ascending order as shown
in Table (Next slide).
The number of times that each mean monthly flow range (class interval, C.I.)
has been equaled or exceeded (m) is worked out as cumulative number of
occurrences starting from the bottom of the column of number of occurrences,
CENG 6803
since the C.I. of the monthly flows, are arranged in the ascending order of
magnitude.
It should be noted that the flow values are arranged in the ascending order of
magnitude in the flow duration analysis, since the minimum continuous flow
that can be expected almost throughout the year (i.e., for a major percent of
time) is required particularly in drought duration and power duration studies,
while in flood flow analysis the CI may be arranged in the descending order
of magnitude and m is worked out from the top as cumulative number of
occurrences since the high flows are of interest. 42
TABLE: FLOW DURATION ANALYSIS OF MEAN MONTHLY FLOW DATA OF A RIVER IN A 10 YR
PERIOD (EXAMPLE 5)
Mean No. of No. time % of time lower Monthly P =
monthly occurrences equaled or value of CI 9.81x60x0.8xQ
flow C.I. (in 10-yr exceeded equaled or (MW)
(m3/s) period) (m) exceeded Q is lower value of
= (m/n) x 100% CI
CENG 6803
100-149 3 120 100 47.2
150-199 4 117 97.5 70.8
200-249 16 113 94.2 94.4
250-299 21 97 80.8 118
300-349 24 76 63.3 142
350-399 21 52 43.3 165
400-449 20 31 25.8 189
450-499 9 11 9.2 212
500-549 2 2 1.7 236
Total n =120 43
(i) The flow duration curve is obtained by plotting Q vs. percent of time in the
Fig. (Q = lower value of the CI.).
(ii) The power duration curve is obtained by plotting P vs. percent of time, see
the Fig.
2. The mean monthly flow
that can be expected is the
CENG 6803
flow that is available for 50%
of the time i.e., 335 m3/s from
the FDC drawn.
The average power that can
be developed i.e., from the
flow available for 50%
of the time, is 157 MW, from
the PDC drawn.
44
3. The effect of storage is to raise the flow duration curve on the dry
weather portion and lower it on the high flow portion and thus
tends to equalize the flow at different times of the year, as indicated
in Fig. above.
CENG 6803
4. If the mean weekly flow data are used instead of the monthly flow
data, the flow duration curve lies below the curve obtained from
monthly flows for about 75% of the time towards the drier part of
the year and above it for the rest of the year as indicated in Fig.
above
In fact the flow duration curve obtained from daily flow data gives
the details more accurately (particularly near the ends) than the
curves obtained from weekly or monthly flow data but the latter
45
provide smooth curves because of their averaged out values.
Exercise 5. The available flow for 97% of the time (i.e., in a year) in a river is 30
m3/s. A run-of-river plant is proposed on this river to operate for 6 days in a
week round the clock. The plant supplies power to a variable load whose
variation is given below:
CENG 6803
47
CENG 6803
(ii) Pondage required to produce the necessary power at the
peak
(iii) The plant load factor
48
Solution
(i) The load curve is shown below.
Total sum of loads at 2-hr intervals = 428.6 x 1000 kW
Average load = (428.6 x 1000 kW x 2hr)/24hr = 35.72 MW
Flow, Q, required to develop the average load
→ Q = 35.72MW/(9.81x22x0.8) = 207 m 3/s
CENG 6803
49
(ii) Flow required to produce the required load/demand
Q = P in 1000 kW/(9.81x22x0.8) = 5.8 x Load in 1000 kW
To determine the pondage capacity the table in the next slide is
prepared. From the table
Total deficiency = total excess = 510 m3/s
Therefore, pondage capacity required = 510 m3/s for 2 hrs
CENG 6803
= 510 x (2 x 60 x 60) = 3.67 x 106 m3 or 3.67 Mm3
(iii) Plant load factor is the ratio of average load to peak load,
35.72/74.2 = 0.482
50
CENG 6803
51
EXERCISE 6
1. The average weekly discharge as measured at a given site is as follows:
CENG 6803
52
(a) Plot
CENG 6803
(iii) The power-duration curve if the head available is 50 m and the
efficiency of the plant set is 85%.
(b) Determine the power that can be developed per m3, the maximum power,
average power, and total energy produced during 26 weeks.
53
EXERCISE 7
Average daily flows in a river in a typical low water week are given
below. A run-of-river hydropower plant is proposed on this river to
operate 6 days in a week round the clock. The full pond effective
head on turbines is 20 m and the plant efficiency is 80%. Maximum
allowable fluctuation of pond level is 1 m.
CENG 6803
Determine:
(i) the capacity of the pond required to give a maximum uniform
output;
(ii) the weekly energy output (kWh) from the plant.