IRJET-Comparative Analysis of Unit Commitment Problem of Electric Power System using DynamicProgramming Technique

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 05 Issue: 06 | June 2018 www.irjet.net p-ISSN: 2395-0072
© 2018, IRJET | Impact Factor value: 6.171 | ISO 9001:2008 Certified Journal | Page 1199
Comparative Analysis of Unit Commitment Problem of Electric
Power System using Dynamic Programming Technique
Amandeep Singh1, Harkamal Deep Singh2
1M.Tech. Research Scholar, Department of EEE, IKGPTU University, Punjab
2Assistant professor, Department of EEE, IKGPTU University, Punjab
-------------------------------------------------------------------- ***----------------------------------------------------------------
Abstract: In this paper shows a Dynamic programming
based on algorithm to solve the (UCP) Unit commitment
problem bookkeeping voltage security consideration and
imbalance limitations. In the present electrical power
system, where electricity demands are in its pinnacle, it
has turned out to be extremely troublesome for
administrators to satisfy the demand. There are
numerous regular and transformative programming
methods utilized for the solution of the unit (UCP) issue.
Dynamic optimization is conventional algorithm used to
take care of the deterministic issue. The created
calculation has been executed on 4 and 10 unit’s power
system. The outcomes got from this strategy was
approved with the accessible procedures and result
discovered satisfactory. The responsibility such that
aggregate cost of generation is reduced to minimize.
Keywords: Dynamic optimization, Fuel cost, Voltage
stability, Unit commitment and Economic dispatch.
I. Introduction
Because of the idea of evolving innovation, (UC) unit
commitment is likewise experiencing an adjustment in
its answer strategy. This is on account of there must be a
proficient technique to confer the generators to meet the
load demand. Numerous strategies have been
acquainted with understand (UC) unit commitment.
Regardless of whether the techniques have favorable
circumstances, the greater part of the strategies
experiences the ill effects of nearby joining and revile of
dimensionality.[1] While booking the activity of the
generating units at least working expense or operating
cost in the meantime satisfying the equality and
inequality limits is the advancement emergency
associated with commitment of the units. The high
dimensionality and combinatorial nature of the unit
commitment issue abridges the endeavors to build up
any thorough scientific enhancement strategy equipped
for solve of the entire issue for any genuine size of power
system. For both deterministic and stochastic loads the
(UCP) is relevant.[6] The deterministic approach gives us
clear and interesting conclusions. Anyway the
dependable outcomes are not gotten for stochastic loads.
All things considered the imperatives are changed into
controlling requirements in stochastic models and after
that by any of the typical calculations the detailing can be
worked out. In the UC issue is settled by itemizing every
single plausible amalgamation of the producing units
and afterward the combination that gives the littlest
measure of the cost of activity is chosen as the most ideal
arrangement. While considering the need list technique
for the conferring the units, replication time and
memory are spared, and it can likewise be related in a
bona fide control power system. Conversely, the need list
strategy has weaknesses that result into problematic
arrangements since it won’t consider every last one of
the conceivable combinations of generation. Dynamic
optimization computer programs are the one of the
techniques which gives ideal arrangement. To give
greatness answers for the UC issue various arrangement
approaches are proposed. Despite the fact that the
dictatorial strategies are basic and quick, they
experience the suffer effects of numerical convergence
and way out greatness issues. This paper gives a definite
analysis’s of the unit commitment issue arrangement
utilizing Dynamic Programming technique, real
commitment is assurance of UC plan with consideration
towards what is known as power system voltage
security. The endeavor is first of its kind in UC
calculation.
II. PROBLEM FORMULATION OF (UCP)
The goal of the (UC) unit commitment is limiting the
aggregate working expense keeping in mind the
operating cost to meet the desire demand. [8] It is
expected that the fuel cost, for unit ‘i’ in a given time
interim is a quadratic function of the output power of the
generators.
2
1
( ) ( ) $ / .
NG
ih i ih i ih i
i
FC P a P bP c hrs

   (1)

Where ai, bi, ci are the comparing unit‟s cost coefficients.
For the booking time frame ‘T’ the total of the generation
costs acquired from the comparing submitted units gives
the aggregate working cost
( 1) ( 1)
1 1
[ ( )* *(1 )* *(1 )* ]
H NG
NH i ih ih ih i h ih ih ih i h
h i
Cost FC P U STUC U U SDC U U 
 
     (2)
Where,
NHCost
is the total operating cost over the scheduled
horizon
( )i ihFC P
is the fuel cost function of units
( 1)i hU 
is the ON/OFF status of ith unit at
( 1)
th
h 
hour.
ihU
is the ON/OFF status of ith unit at hth hour.
U is the decision matrix of the ihU
variable. for
i=1,2,3,........NG.
ihP
is the generation output of ith unit at hth hour.
ihSTUC
is the start-up cost of the ith generating unit at hth
hour.
ihSDC
is the shut-down cost of the ith generating unit at
the hth hour.
NG is the number of thermal generating units
{0,1}ihU  and ( 1) {0,1}i hU  
The accompanying imperatives are incorporated:
a. Power Balance Constraint
The aggregate produced power and load at comparing
hours must be equivalent.
(3)
b. Power generation limit
The produced power of the units should be within max.
and min. power limits.
(4)
III. DYNAMIC PROGRAMMING OPTIMIZATIO
The reason for Dynamic Programming (DP) is the
hypothesis of optimality illustrated by Bellman in 1957.
This strategy can be utilized to clarify emergencies in
which numerous sequential conclusions are to be taken
in characterizing the ideal activity of a power system,
which comprises of particular number of stages. The
seeking might be in forward or in reverse heading. Inside
a day and generation the combinations of units are
known as the states. In Forward dynamic programming a
superb monetary calendar is acquired by beginning at
the starter arrange gathering the aggregate costs, at that
point backtracking from the combination of minimum
amassed cost beginning at the last stage and completing
at the underlying stage. The phases of the DP issue are
the times of the investigation skyline. Each stage as a
rule compares to one hour of activity i.e., mixes of units
ventures forward one hour on end, and target plans of
the units that are to be booked are put away for every
hour. At long last, by retreating from the plan with
littlest measure of aggregate cost at the last hour all
through the finest way to the course of action at the
fundamental hour the most temperate timetable is
obtained. The estimation of every last mix isn't helpful
clearly. Furthermore, a few of the combinations are
restricted because of lacking existing limit.
The well ordered method for dynamic programming
approach is as per the following:
1) Begin haphazardly by considering any two units.
2) Assemble the aggregate output of the two units as
discrete load levels.
3) Determine the most temperate combination of the
two units for all the load levels. It is to be watched that at
each load level, the monetary activity might be to run
either a unit or the two units with a specific load sharing
between the two units.
4) Obtain the more practical cost curve for the two units
in discrete frame and it can be dealt with as cost curve of
single proportional unit.
5) Add the third unit and the cost curve for the
combination of three units is acquired by rehashing the
system.

6) Unless all the current units are viewed as the system
is rehashed.
The advantage of this technique is that having the most
ideal method for running N units, it is easy to discover
the most ideal route for running N + 1 units. The DP
approach based on the subsequent recurring equations.
(5)
Where FM(P) is the base cost in $/hr of generation of P
MW by M generating units. FM(Q) is the cost of
generation of Q MW by Mth unit. FM-1(P-Q) is the min.
cost of generation of (P-Q) MW by the rest of the
(M - 1) units. In its essential shape, the dynamic
programming calculation for (UCP) assesses each
conceivable state in each interim. The dimensionality of
the issue is essentially declined which is the main
preferred standpoint of this strategy. The hypotheses for
organizing the well ordered strategy for dynamic
programming technique are followed underneath.
1) A state comprises of a gathering of units with just
exact units in benefit at once and the remaining
disconnected.
2) While the unit is in off state the start-up cost of a unit
is autonomous of the time particularly it remain fixed.
3) For shutting the unit there will be no cost included.
4) The request of priority is firm and a little amount of
power must be in task in every interim.
IV. FLOW CHART FOR DYNAMIC PROGRAMMING
STRATEGY
Fig.1 Flow chart for Dynamic Programming strategy
The major skilled cost effective combination of units can
be all around decided utilizing the recursive connection.
Impressive computational cost minimize can be achieved
by utilizing this strategy. It isn't compulsory to tackle the
co-ordination equation. The aggregate figure of units
easy to get to, their individual cost attributes and load
cycle should be known. Just when the operations at the
prior stages are not influenced by the choices at the later
stages this strategy is suitable.

V. TEST POWER SYSTEM AND MATLAB RESULTS
The unit (UCP) arrangement strategy is actualized in
Matlab R2010a. A generation organization with 4 and 10
generating units to outline the proposed technique. In
our execution, energy balance and power reserve are
considered at the same time in the detailing
8 hours and 24 hours scheduling period is considered.
Fuel cost function of each unit is evaluated into quadratic
equation .Unit information, load demand, fuel cost
coefficient and market costs are given in Tables I and IV.
Table: I Generating Unit characteristics-4 Unit Model
UNI
TS
Pmi
n
Pma
x
M
Ui
M
Di
Hco
st
Ccost Cho
ur
Initi
al
Stat
e
Unit
1
25 80 4 2 15
0
350 4 -5
Unit
2
60 25
0
5 3 17
0
400 5 +8
Unit
3
75 30
0
5 4 50
0
110
0
5 +8
Unit
4
20 60 1 1 0 0.0
2
0 -6
Table: II Time varying load demand of 4 unit system
Load
Deman
d (MW)
45
0
53
0
60
0
54
0
40
0
28
0
29
0
50
0
Time
in Hour
1 2 3 4 5 6 7 8
Table: III Result of 04 units system using proposed
technique
Table: IV Generating unit characteristic-10 unit
system
UNITS Pmax Pmin A B C MUi MDi Hcost Ccost Chour IniState
Unit1 455 150 1000 16.19 0.00048 8 8 4500 9000 5 8
Unit2 455 150 970 17.26 0.00031 8 8 5000 10000 5 8
Unit3 130 20 700 16.6 0.002 5 5 550 1100 4 -5
Unit4 130 20 680 16.5 0.00211 5 5 560 1120 4 -5
Unit5 162 25 450 19.7 0.00398 6 6 900 1800 4 -6
Unit6 80 20 370 22.26 0.00712 3 3 170 340 2 -3
Unit7 85 25 480 27.74 0.00079 3 3 260 520 2 -3
Unit8 55 10 660 25.92 0.00413 1 1 30 60 0 -1
Unit9 55 10 665 27.27 0.00222 1 1 30 60 0 -1
Unit10 55 10 670 27.79 0.00173 1 1 30 60 0 -1
Hour Demand Tot.Gen Min MW Max MW ST-UP Cost
Prod.Cost F-Cost State Units ON/OFF
0 - - 135 550 0 0 0 13 0 1 1 0
1 450 450 135 550 0 9208 9208 13 0 1 1 0
2 530 530 135 550 0 10648 19857 13 0 1 1 0
3 600 600 155 610 0 12450 32307 14 0 1 1 1
4 540 540 135 550 0 10828 43135 13 0 1 1 0
5 400 400 135 550 0 8308 51444 13 0 1 1 0
6 280 280 135 550 0 6192 57635 13 0 1 1 0
7 290 290 135 550 0 6366 64002 13 0 1 1 0
8 500 500 135 550 0 10108 74110 13 0 1 1 0

Table: V Time varying load demand of 10
unit system
Hour Demand Tot.Gen Min MW Max MW ST-UP Cost Prod.Cost F-Cost State
0 - - 300 910 0 0 0 615
1 700 700 300 910 0 13683 13683 615
2 750 750 300 910 0 14554 28238 615
3 850 850 325 1072 900 16809 45947 764
4 950 950 345 1202 560 19146 65653 838
5 1000 1000 345 1202 0 20020 85673 838
6 1100 1100 365 1332 1100 22387 109160 924
7 1150 1150 365 1332 0 23262 132422 924
8 1200 1200 365 1332 0 24150 156572 924
9 1300 1300 410 1497 860 27251 184683 1006
10 1400 1400 420 1552 60 30058 214801 1018
11 1450 1450 430 1607 60 31916 246777 1023
12 1500 1500 440 1662 60 33890 280727 1024
13 1400 1400 420 1552 0 30058 310785 1018
14 1300 1300 410 1497 0 27251 338036 1006
15 1200 1200 365 1332 0 24150 362186 924
16 1050 1050 365 1332 0 21514 383700 924
17 1000 1000 365 1332 0 20642 404341 924
18 1100 1100 365 1332 0 22387 426728 924
19 1200 1200 365 1332 0 24150 450879 924
20 1400 1400 420 1552 920 30058 481856 1018
21 1300 1300 410 1497 0 27251 509107 1006
22 1100 1100 370 1237 0 22736 531843 868
23 900 900 320 990 0 17645 549488 701
24 800 800 300 910 0 15427 564916 615
Table: VI Result of 10 units system using proposed
dynamic optimization
Time in Hour Load Demand (MW)
1 700
2 750
3 850
4 950
5 1000
6 1100
7 1150
8 1200
9 1300
10 1400
11 1450
12 1500
13 1400
14 1300
15 1200
16 1050
17 1000
18 1100
19 1200
20 1400
21 1300
22 1100
23 900
24 800

Table: VII Turn on/off status of 10 units system using
proposed dynamic optimization
Load
Demand
(MW)
U1 U2 U3 U4 U5 U6 U7 U8 U9 U10
700 1 1 0 0 0 0 0 0 0 0
750 1 1 0 0 0 0 0 0 0 0
850 1 1 0 0 0 0 0 0 0 0
950 1 1 0 0 1 0 0 0 0 0
1000 1 1 0 1 1 0 0 0 0 0
1100 1 1 0 1 1 0 0 0 0 0
1150 1 1 1 1 1 0 0 0 0 0
1200 1 1 1 1 1 0 0 0 0 0
1300 1 1 1 1 1 0 0 0 0 0
1400 1 1 1 1 1 1 1 0 0 0
1450 1 1 1 1 1 1 1 1 0 0
1500 1 1 1 1 1 1 1 1 1 0
1400 1 1 1 1 1 1 1 1 1 1
1300 1 1 1 1 1 1 1 1 0 0
1200 1 1 1 1 1 1 1 0 0 0
1050 1 1 1 1 1 0 0 0 0 0
1000 1 1 1 1 1 0 0 0 0 0
1100 1 1 1 1 1 0 0 0 0 0
1200 1 1 1 1 1 0 0 0 0 0
1400 1 1 1 1 1 1 1 1 0 0
1300 1 1 1 1 1 1 1 0 0 0
1100 1 1 1 0 0 0 1 0 0 0
900 1 1 0 0 0 1 0 0 0 0
800 1 1 0 0 0 0 0 0 0 0
Total Cost ($) 564916
Table: VIII Comparison of result of UCP using
proposed algorithm
S.NO METHOD UNIT TOTAL
COST($)
1 EGA 4 77628.91
2 DP 4 74110.00
3 EGA 10 563937.57
4 DP 10 564916.00
VI. CONCLUSION
This mathematical optimization technique has been
displayed to take care of thermal unit (UCP) by utilizing
dynamic programming approach. For singular sub
problem dynamic programming without discrediting
power generation levels ended up being a proficient
approach. [11] This strategy gives the advantage of non-
discretization of generation levels and is turned out to be
effective for power system with a couple of incline rate
constrained units. The heuristic technique created to get
achievable arrangements is powerful and close ideal
arrangements are gotten.
REFERENCES
[1] Titti Saksornchai, Wei-Jen Lee, Kittipong
Methaprayoon, James R. Liao and Richard J. Ross (2005),
‘‘Improve the Unit Commitment Scheduling by using the
Neural-Network-Based Short-Term Load Forecasting’’
IEEE Transactions on Power Systems,Vol. 41, Year 2005
, pp. 169 – 179.
[2] Shantanu Chakraborty,Tomonobu Senjyu, Atsushi
Yona, Ahmed Yousuf Saber and Toshihisa Funabashi
(2009), “Generation Scheduling of Thermal Units
Integrated with Wind-Battery System Using a Fuzzy
Modified Differential Evolution”, Year 2009 , pp. 1-6

[3] Morteza Eslamian, Seyed Hossein Hosseinian, and
BehroozVahidi (2009), ‘‘Bacterial Foraging-Based
Solution to the Unit-Commitment Problem ’’ IEEE
transactions on power systems, vol.24, No.3 year Aug.
2009, pp. 1478-1488.
[4] Yare Y., Venaya gamoorthy G. K., and Saber A. Y.
(2009), “Economic Dispatch of A Differential Evolution
Based Generator Maintenance Scheduling of A Power
System”, IEEE Transactions on Power Systems, Vol. 12
July 2009, pp. 1-8.
[5] S.O. Orero and M.R. Irving (2002), ‘‘A Genetic
Algorithm Modeling Power system and Solution
Technique for Short Term Optimal Hydrothermal
Scheduling’’ IEEE Transactions on Power Systems, Vol.
13, May 1998, pp 501-518.
[6] Gary W. Chang, Mohamed Aganagic, James G. Waight,
José Medina, Tony Burton ,Steve Reeves, and M.
Christoforidis (2001), “Mixed Integer Linear
Programming Based Approaches on Short- Term Hydro
Scheduling” IEEE Transactions ON power systems, vol.
16, No. 4, November 2001 pp.743-749.
[7] G. K. Purushothama and Lawrence Jenkins (2003),
‘‘Simulated Annealing With Local Search A Hybrid
Algorithm for Unit Commitment ’’ IEEE Transactions on
Power Systems, Vol.18, No. 1, Year feb. 2003, pp.273-
278.
[8] Ebrahimi, J.Hosseinian, S.H.(2011),‘‘Unit
Commitment Problem Solution using Shuffled Frog
Leaping Algorithm’’IEEE Transactions on Power
Systems, Vol. 26, Year 2011, pp. 573 -581.
[9] Ioannis G. Damousis, Anastasios G. Bakirtzis and
Petros S. Dokopoulos (2004), “A Solution to Unit
Commitment Problem Using Integer Coded Genetic
Algorithm ’’ IEEE Transactions on Power
Systems,Vol.19, No. 2, May 2004, pp. 1165- 1172
[10] Navpreet Singh Tung, Ashutosh Bhadoria,
Kiranpreet Kaur, Simmi Bhadauria.” Dynamic
programming model based on cost minimization
algorithms for thermal generating units”, International
Journal of Enhanced Research in Science Technology &
Engineering, Vol 1, issue 3,Dec 2012 pp.2319-7463.
[11] R.H. Kerr, J.L. Scheidt, A.J. Fontana and J.K. Wiley,
“Unit Commitment”, IEEE Transactions on Power
Apparatus and Systems, vol.PAS-85, No.5,May 1966,
pp.417-421,

IRJET-Comparative Analysis of Unit Commitment Problem of Electric Power System using DynamicProgramming Technique

Recommended

More Related Content

What's hot (17)

Similar to IRJET-Comparative Analysis of Unit Commitment Problem of Electric Power System using DynamicProgramming Technique (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET-Comparative Analysis of Unit Commitment Problem of Electric Power System using DynamicProgramming Technique