See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/235356323

Improved Algorithms for Reservoir Capacity Calculation Incorporating

Storage-Dependent Losses and Reliability Norm

Article in Water Resources Research · October 1987

DOI: 10.1029/WR023i010p01819

CITATIONS READS

38 438

1 author:

Sharachchandra Lele
Ashoka Trust for Research in Ecology and the Environment
211 PUBLICATIONS 6,467 CITATIONS

SEE PROFILE

All content following this page was uploaded by Sharachchandra Lele on 21 May 2014.

The user has requested enhancement of the downloaded file.

 WATER RESOURCES RESEARCH, VOL. 23, NO. 10, PAGES 1819-1823, OCTOBER 1987

Improved Algorithms for Reservoir Capacity Calculation

Incorporating Storage-Dependent Losses and Reliability Norm
SHARAD M. LELE x

Departmentof ComputerScienceand Automation,Indian Institute of Science,Ban•Talore,India

Two algorithms that improve upon the sequent-peakprocedurefor reservoircapacity calculation are
presented.The first incorporatesstorage-dependent losses(like evaporation losses)exactly as the stan-
dard linear programmingformulation does.The secondextendsthe first so as to enable designingwith
lessthan maximum reliability even when allowable shortfallin any failure year is also specified.Together,
the algorithmsprovide a more accurate,flexible and yet fast method of calculatingthe storagecapacity
requirementin preliminary screeningand optimization models.

1. INTRODUCTION In this paper we present two simple algorithms that over-

The sequent-peakprocedure is a well-known simplistic al- come limitations (1) and (3) completely. In part I we describe
gorithm for determining the storage capacity requirement for an algorithm that is based on the sequent-peakprocedure but
a reservoir [Thomas and Fiering, 1963]. It has also been ex- takes into considerationstorage-dependentlossesas "exactly"
tended to cascadedmultireservoir systemsin analyses based as they are included in the standard LP formulation. (Al-
on a dynamic programming approach [Murray and Weiss- though they are called "evaporation losses" hereinafter, it
beck, 1980]. But the procedure suffers from serious limi- should be noted that they could be any kind of storage-
tations: dependent lossesor withdrawals.) In part II we show how this
1. Storage-dependentlossesor releasescannot be included improved algorithm may be used in a procedure that would
in the calculations. enable one to calculate the active storage capacity require-
2. It cannot be usedfor complex multireservoir systems. ment when the reliability norm specifiesnot only the maxi-
3. It cannot be used when less than maximum reliability mum number of failure years but also the allowable shortfall
(percentageof periods in which target demand is met) is de- in any failure year. Finally, we indicate how the procedure
sired; specificationof the shortfall that may be allowed in any may be able to overcome limitations (2) and (4) to some
extent.
failure year is, therefore,out of the question.
4. It cannot be usedwhen the reservoiroperating policy is
not the standard one. (A standard operating policy (SOP) is 2. PART I: AN ALGORITHM FOR INCORPORATING
one under which the seasonal release is equal to the target STORAGE-DEPENDENT LOSSES INTO
releaseor all the available water, whicheveris less.) THE SEQUENT-PEAK PROCEDURE
On the other hand, the procedure has the significantadvan-
We begin by defining the notation used. The original
tage of not being limited by the number of years and seasons
sequent-peak procedure is outlined in section 2.1, and the al-
of inflows used in the analysis,a factor that is crucial in deter-
gorithm is then derivedfrom it in the subsequentsections.
mining practical solvability in analysesbased on the common-
ly used linear programming (LP) formulation [Loucks et al., a slope of the area-capacity curve for the reservoir,
1981, pp. 236-238]. Further (at least in the case of single or L-1.
cascaded multireservoir systems),it enables the analyst to b reservoirwater spreadat deadstoragelevel,/2;
tackle problems with significantly nonlinear objective func- At reservoirwater spreadat the beginningof time
tions with the help of direct-searchoptimization methods. periodt, Le;
Some attempts have been made to overcome the short- CU t fixed lossfrom the reservoirin time period t
comings mentioned above. Modifications have been easily (suchas localconsumptive use),/2;
made to enable calculation of capacity for different levels of ½t averagerate of evaporationfrom reservoirsurface
reliability [Loucks, 1976]. Tejada-Guibert [1978] has present- in period t, L;
ed an iterative algorithm for calculating storage capacity at E Vt volume of water lost owing to evaporationin
maximum reliability in which evaporative losses are accu- periodt, L3'
rately included and also a half-interval search procedure for ICRIT last period in critical sequenceof inflows;
determiningthe storagerequired for a desiredreliability. None IOVF last period in which the reservoir spilled over
of these algorithms, however, is able to provide for specifi- before encounteringthe critical sequenceof inflows'
cation of the extent of shortfall that may be allowed in any K a approximateactivestoragecapacityof reservoir
failure or "short" year. calculated from sequent-peak procedure ignoring
storage-dependent
losses,/2'
•Now at Energy and ResourcesGroup, University of California, K,•* exactactivestoragecapacity,L3'
Berkeley.
K t sequentialcumulativedeficitat the end of period
Copyright 1987 by the American GeophysicalUnion. t, L3;
Paper number 7W0617. Qt inflowintoreservoir
duringperiodt,/•'
0043-1397/87/007W-0617505.00 R• releasefromreservoir
duringperiodt,/•;
1819
1820 LELE' IMPROVED ALGORITHMS FOR RESERVOIR CAPACITY CALCULATION

St active storagein the reservoir at the beginning Now, the original storage-continuity relationship in the
of period t, L3' standard LP model [Loucks et al., 1981, pp. 343-348] is
Y total number of years in the inflow sequence'
S number of within-year seasonsthat each year is
St + Qt- Rt- CUt- EVt > St+l t = 1,..., T
divided into' which, using (3), can now be rewritten as
T total number of periods in the inflow sequence
(= YS). St(1-- eta/2) + Qt - Rt - CUt - etb> St+•(1 + eta/2)
or
Note that it is assumed as usual that if t = T then t + 1 = 1,
i.e., the given set of inflows repeats itself. In other words, the St+• < [St(1-eta/2)+Qt-Rt-CUt-etb]/(1 +eta/2 ) (4)
sequence of inflows may be looked upon as not a "straight
line" but a "closedcircle." Therefore a phrase such as "back- The other set of constraints in the LP model that goes
ward from t = t• to t = t2" means(when t2 > t•) "from t = t• along with (4) is the set of constraints specifyingthe minimum
down to t= 1 and then t= T down to t=t2" Similarly, active storage capacity:
"forward from t = t• to t = t2" means (when t2 < t•) "up to
t = T and then from t = 1 up to t = t2." K•* > St t= 1,'", T
2.1. Sequent-Peak Procedure (The symbol K•* is used instead of K, to indicate that this
Given the values of the inflows Qt, releasesRt, and fixed would be the exact value of capacity required and not the
lossesCU t (and assumingEVt = 0) for t = 1, ---, T, the mini- approximate one given by the sequent-peakprocedure.) This
mum active storagerequirementK a can be calculatedusing constraint, when combined with (4), gives a rule for determin-
the sequent-peakprocedureas follows. ing St+ •, given St and Ka*, which is
Stepl. SetK 0=0.
Step2. Fort= 1 to T, calculateKt= max {O,(Kt_• + Rt St+• = min {K•*,
+ CUt- Qt)}. - [(St(1--eta/2)+Qt-Rt--CUt--etb)/(1 +eta/2)]} (5)
Step 3. If K r - K o, then go to step 4' elseif this is the first
iteration, then set K 0 -K r and go to step 2' else STOP' Note that if St+• is equal to K•*, it indicatesthat the differ-
sequent-peak analysis failed becausegross utilization is great- ence between the secondterm on the right-hand side and K•*
er than the average inflow. is the volume of water that has spilled over.
Step4. Ka = maximum{K,} overt = 1,.--, T. The questionof courseis "how can the valuesof St and K,*
Note that Rt and CU t values are actually given/specified be known (when Ka* itself is the basic unknown to be deter-
only for t = 1 to S; since they repeat from year to year, R t = mined)?". We shall now outline the algorithm that showshow
Rt_ s for t > S, etc. this can be done systematically.
Evaporation loss in any period t is actually proportional to The starting point of the algorithm is the fact that if one is
the average reservoir water spread area during that period, sure that during a certain period (or sequenceof periods) there
which in turn dependson the storage levels in the reservoir at has been no spillover, then for only that period (or sequenceof
the beginning and the end of that period, i.e., St and St+•. periods)
Since it is not possibleto determine St without knowing K•,
the above procedure obviously cannot include any such St+•=[St(1--eta/2)+Qt-Rt--CUt--etb]/(1 +eta/2 ) (6)
storage-dependent losses.

2.2. Inclusion of Evaporation Losses (which temporarily obviates the need to know K•*). Now,
there does exist such a sequenceof periods within which one
The manner in which evaporation lossesare incorporated in
can be sure that there have been no spillovers, namely, the
the standard LP formulation provides a pointer to the modifi-
critical sequenceof inflows. Further, one knows that (1) the
cations required in the sequent-peak procedure. In the LP
active storage at the end of the sequencewill be zero (the
formulation, the area-capacityrelationship for the reservoiris
reservoir will be at its maximum draw-down level) and (2) it is
approximated as
possible to identify the beginning and end of this sequence
At = aSt + b (1) from the calculations in the sequent-peak procedure.
The identification of the critical sequenceis a simple matter.
(Note that this linearization of the area-storagerelationship is In step 4 of the sequent-peakprocedure,all the values of the
an approximation that may lead to distortions in some cases, sequentialcumulativedeficit K t have to be examined.Clearly,
but it is neverthelessused very commonly. Our algorithm uses the value of t correspondingto the largestdeficit K t (which is
the same approximation and so is "exact" only to the extent then taken as K•) is nothing but the last critical period, say
that the LP formulation is.) ICRIT. Hence from remark 1 above,
The evaporation loss in any period t is taken to be the
product of the average reservoir water spread during period t SICRIT
+ 1 --- 0
and the averageevaporation rate for that period. Hence
Moreover, the beginning of the critical sequenceis the last
EV, = et[(At + At+ ,)/2] (2)
time the reservoir spilled over or "overflowed" before this low-
Using (1) and (2), one obtains an expressionfor the evapora- inflow sequencewas encountered, or going backward from
tion loss as a function of the active storage volumes' t = ICRIT, it would be the first period for which K t = 0. Let
this period be IOVF. Note that the storagein the reservoirat
EV, = et[a(St + St+•)/2 + b] = St(aet/2)+ St+•(aet/2) + etb
the end of this period as indicated by the sequent-peakpro-
(3) cedure(say,SiOVF+
•') would be equal to K•. One could then
 LELE:IMPROVED
ALGORITHMS
FORRESERVOIR
CAPACITYCALCULATION 1821

rewrite (6) as reliability when the allowable shortfall in any failure year is
also specified.
St=[St+ 1(1+eta/2)+Rt+CUt+etb-Qt]/(1-eta/2) (7)
and calculate backward from t - ICRIT to t - IOVF q- 1. 3.1. Further Notation and Definitions
Note that SlovF+l is the storagevolume that should have In addition to the symbols defined in section 2, we define
been there in the reservoir at the beginning of the critical the following:
sequenceif the reservoir were to provide for all the releases,
f number of failure years allowed;
consumptive losses and exact evaporation lossesin that se-
d reliability (expressedas a fraction);
quenceof periods.Hence in general,SiovF+ • would be greater •z fraction of normal yield desired in a failure
than SiOVF+ 1' (= Ka), and in fact, it is the desiredexact active year (= 1 - maximum allowable fractional shortfall);
storage,i.e.,
Rt* targetreleasein periodt l-L3].
Ka* = SIOVF+1 (8) Given the desired degree of reliability d, the allowable number
This claim, however, needsto be qualified. Since the critical of failure years can be calculated using the definition of reli-
sequencecould be different for different levelsof grossutiliza- ability:
tion, the inclusion of the evaporation lossesmay lead to a shift d = (Y --f)/(Y + 1)
in the critical sequenceitself. Such a shift would manifest itself
Therefore
in a negativevalue of St (i.e., a deficit) in some period subse-
quent to ICRIT. To take care of this possibility,the remaining
f=rnd {Y-(Y+ 1)d}
valuesof St need to be calculatedusingrule (5) with Ka* from
(8) above and checkedfor negativity.If any St is found to be where rnd means rounded off. So, the actual set of releases
negative,the sequent-peakprocedurewould have to be repeat- should be Rt = •zRt*for all the within-year periodsin a failure
ed with the new (nonzero) values of evaporative lossesEVtt yearand Rt = Rt* for all otherperiods.
(obtained using the current values of Sts in (3)); ICRIT and In the LP formulation that allows for less than maximum
IOVF would have to be recalculated and the procedure re- reliability, the identification of the actual failure years has to
peated all over again. be done by trial and error, since the critical years are not
known beforehand [Loucks et al., 1981, pp. 348-349]. Now,
2.3. Al•lorithm I however, using the algorithm from part I, we can identify the
Thus the algorithm may be statedsuccinctlyas follows. critical yearsdynamically,i.e., during the calculationitself, and
modify or "reset"the releasesfor those years to lower values.
Step 1: Initialization.
1. Initialization of Qt, R, e, CU t, a, b. 3.2. Procedure II
2. E V•= 0 for all t = 1,..., T.
Step l. Set Rt = Rt* for all t = 1,---,T.
Step 2: Sequent-PeakProcedure.
Step 2. Execute "modified" algorithm (of part I of this
1. K o -0.
paper) to determineKa* and ICRIT for current set of releases
2. CalculateK t = max {0, (Kt_ • + Rt + CUt + EVt -- Qt)}
for all t - 1,---, T.
Step .3. If this is the (f + 1)th iteration (i.e., if releaseshave
3. If K r = K o, then go to number4; elseif this is the first
been "reset" forf years), then STOP; else go on to step 4.
iteration in step 2, then put K o = Kr and go to number 2; else
Step 4. From ICRIT, determine the current critical year,
STOP: sequent-peakanalysisfailed becausegrossutilization
say ICY. If ICY is the same as for the last iteration, then reset
is greater than the averageinflow.
Rt = •zRt* for t correspondingto periodsin year previousto
4. Find the period for which K t is the maximum of all Kt,
ICY; elseresetRt = art* for t correspondingto periodsin the
t - 1, .-., T; call this ICRIT.
ICYth year. Go back to step 2.
5. Search backward from t- ICRIT until K t -0 is en-
Sometimes the critical period happens to be the first season
counteredfor the first time. Assignto IOVF this value of t.
of the "wet" year immediately following a drought sequence.
Step 3: Recalculation.
In such a case it is customary to choose the last year of the
1. SlCRIT+ I = 0.
2. For t = ICRIT + 1 backward to t = IOVF + 1, calcu-
drought sequenceas the critical year rather than the wet year
to which ICRIT belongs.
late St = [St+i(1 + eta/2) + Rt + CU t + etb-Qt]/(1 -eta/2 ).
Thus it is possible to generate a complete and accurate
3. Ka* = SIOVF+1.
"family" of storage-yield curves for a set of inflows of any
Step 4: Checking.
length. Clearly, the above procedure could be modified easily
1. For t = ICRIT + 1 forward to t = IOVF- 1, calculate
so as to reset the releases for one season at a time rather than
St+1=min{Ka*,[St(1-eta/2)+Qt-Rt-CUt-etb-]/(1 +eta/2)}.
for all the seasonsin a year at once. It can also be easily
2. If any St is negative,then go to number 3; elseSTOP;
altered to suit the concept of providing secondary yields of
Ka* is the requiredexactactivestorage.
lessthan maximum reliability (in addition to a firm yield with
3. Calculate EVt- et•a(St + St+•)/2 + b] for all t = 1,
maximum reliability) as outlined by Loucks [1976, pp. 168-
ß--, T, and go back to step 2.
169] or Loucks et al. [1981, pp. 350-351].
3. PART II: PROCEDURE TO CALCULATE THE ACTIVE
STORAGECAPACITY REQUIREMENTFOR DIFFERENT 4. CONCLUDING REMARKS
LEVELS OF RELIABILITY WITH A SPECIFIC
The algorithm for incorporating storage-dependentlosses
DEGREE OF SHORTFALL IN ANY FAILURE YEAR
was tested on a real 50-years and 12-months inflow sequence
In this part a procedureis outlined which enablesone to and was found to give the same resultsas those obtained by
calculate the active storage capacity for less than maximum using the LP formulation to the last significantdigit. The
1822 LELE.' IMPROVED ALGORITHMS FOR RESERVOIR CAPACITY CALCULATION

second procedure was also implemented as a computer pro- with the storagecapacity being calculatedat each point in the
gram; it provided a very quick and flexible way of accurately searchusing one of our algorithms. For example,for a single-
determining the effect of changesin the reliability norm, in- reservoir 100-year and 12-month model, if the objectivefunc-
cluding in the percentageshortfall allowed, on the storage tion is nonlinear (say, becauseof a nonlinearly varying dam
capacity requirement. cost), the LP approach requires the use of a piecewiselinear
Some remarks may be pertinent here as to the usefulness approximation and integer programming; the problem would
and scope of the algorithms. First, the inclusion of evapora- then be of more than 1200 variables and more than 2400
tion losses is itself of significance, because although the constraints.Instead, one could use algorithm I and pose the
averageannual evaporationlossmay be only about 6-10% of problem as one in which only the monthly releasesare the
the storagecapacity of the reservoir,the differencein the stor- decisionvariables.This would reducethe problemto a 12-
age capacity requirement as calculated before and after the variable (and maybe 12 bounds) problem; existing com-
inclusion of evaporation lossesis usually much higher, with puterized algorithms can solve nonlinear constrained opti-
the magnitude of the increasedependingon the length of the mization problems of this size quite rapidly. (See Lele [1986,
critical sequence.(For example, this increasewas 30% in the chap. 3] for such an application.) Since the algorithm calcu-
case of the same inflow data mentioned above.) Consequently, latesthe valuesof S, for all t, it could be usedevenin the case
a reasonably accurate estimation of the evaporative lossesis of a hydropowergenerationmodel wherein the objectivefunc-
necessaryfor the accurate estimation of the active storage tion is nonlinear becausethe generation head is a function of
capacity requirement, especially at sites with high rates of the storage level.
evaporation. When specifyingthe minimum reliability norm for water
Second,although we have developedthe algorithms for the resourcesprojects,decision-makingbodies often also specify
case when the reservoir operating policy is the SOP, they the extent of shortfall that may be allowed in any failure year.
could be used (with minor modifications) under other oper- For instance,the Planning Commissionin India specifiesthat
ating policies,too. For instance,a very general S-type linear the reliability of hydropower and irrigation projects may not
decision rule (LDR) of the kind be less than 90% and 75%, respectively,and the shortfall in
any failure year may not be more than 15-20% (Y. D. Pendse,
R, = csS
, + ds Central Water Commission, New Delhi, personal communi-
cation, 1985). The LP formulation uses chance-constrained
(subscripts denotesthat the decisionparametersare different modelsfor suchproblems,which require the estimationof the
only from seasonto season,not year to year) couldbe incor- cumulative distribution functions for the seasonal inflows;
porated easily by substituting(c + e,a/2) for (e,a/2) and they also usually do not specifythe extent of shortfall. Rather
(d + be,)for (be,)in the statementof the algorithmin section than go through these complex calculations,planners often
2.3. Similarly, it should be possibleto incorporate an SQ-type
prefer to designapproximately.In doing so, they err on the
LDR. Since LDRs have not been found to be very appropriate
side of overdesigning,i.e., for excessivelyhigh reliability and
for reservoircapacityscreeningmodels[Stedinger,1984], this low shortfall. Calculations show that the difference between
may not be a very useful avenuefor further exploration; the storagecapacity requirementsat, say, 99% and 90% reli-
nevertheless,it servesto demonstratethe inherentflexibility of
ability could be very significant; for the inflow sequencemen-
the algorithms. A more useful extension would be one tioned above, it was 30%. Similarly, exact consideration of
whereby a "less than 100% minimum storage reliability" shortfall could also affect the storage capacity calculation sig-
couldbe incorporated;resettingR, to (R,* -- Smin)rather than
nificantly. Overdesigningcould turn out to be quite costly,
0•R,*in step4 of procedureII for ff numberof periods(where especiallywhen one considersthe environmental impacts of
Smin is the minimumactivestoragerequirementthat may be land submergence.Procedure II provides a simple, fast, and
violated /• fraction of the time; if= rnd {YS/•}) should accurate method for recalculating the capacity and thus for
achieve that.
examining the tradeoffs between reliability and economic
Any algorithm that calculatesK,* directly obviatesthe and/or environmental costs.
need to include the 2T constraints (T constraints for storage
continuityand T for minimum storage)and the T variables Acknowledgments.The author wishes to thank S. Vedula of the
(correspondingto the storagein eachtimeperiod)in the reser- Indian Institute of Science,Bangalore, and Daniel P. Loucks of Cor-
voir capacityoptimizationproblem.It is this large numberof nell University, Ithaca, for their encouragement. The prompt and
constraintsand variablesthat imposessignificantlimits on the valuable comments of the anonymous referee are also gratefully ac-
knowledged.
size of the LP problem that can be solved practically (and
forces analysts to look for alternativesto the "complete" REFERENCES
model like the "critical sequenceof years"model or the "ap- Lele, S. M., An optimization model for macro-hyddprojects incor-
proximate within-year and over-year separatedsequences" porating energy/economiccosts of submergence,M.S. thesis, 160
model [Loucks et al., 1981, pp. 343-348]). Conversely,our pp., Indian Inst. of Sci.,Bangalore, 1986.
algorithms,with their very smallcomputationalrequirements Loucks, D. P., Surface water quantity management models, in Sys-
tems Approach to Water Management,edited by A. K. Biswas,pp.
and their "self-contained" form, could be used for solving
167-170, McGraw-Hill, New York, 1976.
modelswith a very largenumberof periodsand evenin multi- Loucks, D. P., J. R. Stedinger, and D. A. Haith, Water Resources
reservoir models. SystemsPlanningand Analysis,chaps. 5 and 7, Prentice-Hall, En-
Further, since the remaining constraints and variables are glewood Cliffs, N.J., 1981.
Murray, M., and K. Weissbeck,Computer program for river hydro
much fewer in number (typically involving only the within-
development,Intl. Water Power Dam Constr.,30(3), 30-40, 1980.
year releases,the active storagecapacity,and related con- Stedinger,J. R., The performanceof LDR models for preliminary
straints), it is possible to solve nonlinear problems using design and reservoir operation, Water Resour.Res., 20(2), 215-224,
direct-search methods for constrained nonlinear optimization, 1984.
 LELE: IMPROVED ALGORITHMS FOR RESERVOIR CAPACITY CALCULATION 1823

Tejada-Guibert,J. A., Algoritmospara la determinacionde capacida-

des de embalse(Algorithms for the determination of reservoir ca- S. M. Lele,EnergyandResources Group,Building
T-4, University
pacities),paper presentedat the Eighth Latin-American Congress of California,Berkeley,CA 94720.
on Hydraulics, Quito, Ecuador, i978.
Thomas, H. A., Jr., and M. B. Fiering, OperationsResearchin Water (ReceivedMay 7, 1987;
Quality Management,pp. 1-17, Harvard Water ResourcesGroup, revisedJuly 13, 1987;
Cambridge, Mass,, 1963. acceptedJuly 16, 1987.)

View publication stats