Optimal Placement of D-STATCOM Using Hybrid Genetic and Ant Colony Algorithm to Losses Reduction

International Journal of Applied Power Engineering (IJAPE)
Vol. 2, No. 2, August 2013, pp. 53~60
ISSN: 2252-8792  53
Journal homepage: http://iaesjournal.com/online/index.php/IJAPE
Optimal Placement of D-STATCOM Using Hybrid Genetic and
Ant Colony Algorithm to Losses Reduction
Askar Bagherinasab1
, Mahmoud zadehbagheri1
, Saifulnizam Abdul Khalid1
, Majid Gandomkar2
,
Naziha Ahmad Azli1
1
Faculty of Electrical Engineering, Universiti Teknologi Malaysia, 81310 Skudai, Johor, Malaysia
2
DEP.Electrical Engineering. Islamic Azad University, Saveh branch,saveh, Iran
Article Info ABSTRACT
Article history:
Received Dec 5, 2012
Revised Feb 21, 2013
Accepted Mar 7, 2013
In this work, a modern algorithm by hybrid genetic algorithm and ant colony
algorithm is designed to placement and then simulated to determine the
amount of reactive power by D-STATCOM. Also this method will be able to
minimize the power system losses that contain power loss in transmission
lines. Furthermore, in this design a IEEE 30-bus model depicted and three D-
STATCOM are located in this system according to Economic
Considerations. The optimal placement of each D-STATCOM is computed
by the ant colony algorithm. In order to optimize placement for each D-
STATCOM, two groups of ant are selected, which respectively located in
near nest and far from the nest. Moreover, for every output simulation of D-
STATCOM that is used to produce or absorb of reactive power, a genetic
algorithm to minimizing the total network losses is applied. Finally, the result
of this simulation shows net losses reduction about 150% that it verifies the
new algorithm performance.
Keyword:
Ant colony algorithm
D-STATCOM
Genetic algorithm
Optimization
Reactive power
Copyright © 2013 Institute of Advanced Engineering and Science.
All rights reserved.
Corresponding Author:
Saifulnizam Abdul Khalid,
Faculty of Electrical Engineering,
Universiti Teknologi Malaysia,
81310 Skudai, Johor, Malaysia.
Email: nizam@fke.utm.my
1. INTRODUCTION
Recently, Improving of power quality has been considered for power distribution companies and
both low and medium voltage costumers [1]-[4]. There are many reasons for more attention to power quality
of power distribution companies such as connect networks together to form larger networks because of fault
element in the network, increasing harmonics in power systems, customers' increasing awareness of power
quality issues, increased sensitivity of electrical devices against disturbances of distribution networks [5]-[7].
Because of the rise of unbalanced loading on each phase, biased faults always take place in the distribution
system which beginnings unstable voltage and current with negative component [8]. The distribution sector
as the main link between the people and the power industry role more evaluation and judgment than other
power parts and that's why the increasing quality of the electricity distribution is essential. In addition,
determine the optimum capacitor placement in the distribution system is used to minimize the energy losses
with improving the voltage profile of the system and then enhancement of the power factors of a distribution
system [9]-[13]. Many potential applications such as heuristics and linear non-linear optimization techniques
have been explored to solve the power quality problem [14]-[17].

 ISSN: 2252-8792
IJAPE Vol. 2, No. 2, August 2013 : 53 – 60
54
2. PRINCIPLE OF D-STATCOM
The D-STATCOM has been applied as a favourable device to provide an important role in the
distribution system such as voltage sag mitigation, voltage stabilization, flicker suppression, power factor
correction, and harmonic control [18]-[20]. It is notable that D-STATCOM is one of the important devices
that are able to solve the power quality problems at the distribution network [21]-[22]. D-STATCOM has
been used to solve the unbalanced faults in the system as a certain controller [22]. D-STATCOM is an
unbiased three-phase voltage or current through an ability shunt device so that can control the magnitude and
the phase angle [21]. Distribution Static synchronous compensator (D-STATCOM) includes a voltage source
inverter like a controller, a DC energy storage and Gate Turn off (GTO) thristor which causes a balanced set
of current or three phases sinusoidal voltage at the basic frequency. There is an efficient control of both
active and reactive power which use as connecting device between the D-STATCOM and the AC system.
Absorb or produce controllable active and reactive power can be applied by D-STATCOM construction [2].
Generally, the core of a D-STATCOM made of a three-phase inverter which on one side is
connected to the network through the transformer and from the other side end to a capacitor which is its DC
power. In addition, the input signals include voltage of bus (V), output current of convertor (I) and a
reference voltage (dc). The real power is determined by reference voltage which is absorbed by the AC
system to provide its internal losses.
3. BASIC CONCEPTS OF GENETIC ALGORITHM
In this paper Genetic Algorithm (GA) is used to Placement and determine the D-STATCOM as an
optimal system [23]. Also, GA is an effective method in bulky and extended places that has coded variables
so that led to the optimal solution [24]. The advantage of coded variables is that the code is the ability to
transform a continuous space to a discrete space [25]. We use the GA method optimization in population or a
set of points in a certain moment while the old method optimized has been applied for only point. This means
that the large number of projects can be processed at a same time by GA and also it is notable that, GA
method is based on directed randomness. In order to use of GA many concepts such as defined the objective
function or cost function, definition and implementation of genetic space and definition and implementation
of GA operators.
4. BASIC CONCEPTS OF ANT-COLONY OPTIMIZATION (ACA)
In recent years, extensive research on optimization methods has been used for solving dynamic
problems in the field of engineering and business as well as numerous optimization methods have been
considered [26]. The evolutionary methods that are well- known because their unique properties are more
considered [26]-[27]. The ant colony optimization (ACA) is one of the optimization methods that have been
investigated in evolutionary classification [26]-[28]. This optimization technique has been inspired by the
behaviour of real ants for finding the meals by optimal performance as has the exceptional ability to solve the
well-known optimization of engineering and business problems[26]. When an ant removed from its nest to
reach food, leaves a trace in his path that is a chemical substance called pheromone [26],[29]. Consequently,
the other ants guide to locate food with scent and follow the path marked out, so far the ants leaving
pheromone in their path to add to its concentration [26].
Fig.1. the test paths of ants
If exist several paths with dissimilar distances between the nest and the food of the ants, shortest
path will be optimized because more pheromone of different ants remained behind the shortest path as shown
in figure (1). In this schematic, at first, two paths that consist of ACB and ADB select by ants, After a
number of ants in the shorter path ACB increased while the number of ants decreases in the prolonged path
(ADB), accordingly, all the ants move in the shortest way.

IJAPE ISSN: 2252-8792 
Optimal Placement of D-STATCOM Using Hybrid Genetic and Ant Colony Algorithm (Askar Bagherinasab)
55
5. POWER SYSTEM SIMULATION
Simulation results on a 30 bus of the IEEE are investigated that is one of the famous systems of
power quality. After simulation of 30 bus of IEEE, Placement Optimizer of D-STATCOM is designed and
then determine of D-STATCOM reactive power is calculated by using a combination of GA and ACA.
The used power system
One IEEE 30-bus model used so that 5 number of its bus have a generator as shown in figure (3).
Fig. 3. IEEE 30-Bus Model
Data of transmission lines is one required information in system simulation; therefore, it should be reminded
that this system has 41 transmission lines.
6. PLACEMENT OF D-STATCOM IN NETWORK
Determine the required number of D-STATCOM, placement and also the amount of reactive power
generated or absorbed by the D-STATCOM is very significant stage to D-STATCOM designing. Due to
economic considerations especially in power systems three similar D-STATCOM is installed. Because, the
system has 30 buses, so, 30 choices are for the location of each D-STATCOM. Generally, one of bus uses to
reference bus, and therefore, different modes for 3 D-STATCOM installed at 29 buses is as follows.
)!1(!
)!1(
rnr
n
s



(1)
Where (n) is total of bus and (r) is the number of D-STATCOM. Also, in order to obtain possible states(s) for
three number of installed D-STATCOM in the system, we can write.
3654
)!3130(!3
)!130(



s
(2)
Thus, 3654 non-repetitive mode is available for the installation of D-STATCOM in the power system and
also, finds an optimal point in all cases is focused by the combination of AG and GA.
7. GENETIC ALGORITHMS TO PLACEMENT AND DETERMINATION OF D-STATCOM
Due to important designing of active loads in each bus of D-STATCOM, the active and reactive
power of each system is determined as follows.
p= [40.0000 -2.4000 -7.6000 -94.2000 0 -22.8000 -30.0000 0 -5.8000 0 -11.2000
40.0000 -6.2000 -8.2000 -3.5000 -9.0000 -3.2000]
-9.5000 -2.2000 -17.5000 40.0000 40.0000 -8.7000 0 -3.5000 40.0000 0 -2.4000 -
10.6000] (MW)
Q= [5.3000 -1.2000 -1.6000 -19.0000 0 -10.9000 -30.0000 0 -2.0000 0 -7.5000
25.0000 -1.6000 -2.5000 -1.8000 -5.8000 -0.9000 -3.4000 -0.7000 -11.2000 15.0000
8.4000 -6.7000 0 -2.3000 30.0000 0 -0.9000 -1.9000] (MVAR)

 ISSN: 2252-8792
56
As can be seen, the values of active and reactive power at the reference bus (bus 1) is unidentified
so, these unidentified powers can be solved by the Newton –Raphson method.
There is not compensation by first bus in the range of 2 to 30, so, cost function can be defined as.
( 1000[ ( (1) (2))]) ( 1000[ ( (1) (3)])) ( 1000[ ( (2) (3)]
1
1000 1000 1000
round x x round x x round x x
e e eF
     
  
(3)
Where x (1), x (2) and x (3) are location of D-STATCOM respectively and also, If the selected location will
be repeated, F1=0.
The following equation is used to placement x (4) and x (6) in the range of 2 to 30.
1000( (1) 31) 1000( (2) 31) 1000( (3) 31) 1000(1 (1)) 1000(1 (2)) 1000(1 (3))
2
1000 1000 1000 1000 1000 1000
X X X x x x
e e e e e eF
     
      (4)
Where x (4), x (5) and x (6), respectively, represent the reactive power generated or absorbed by each of the
D-STATCOM. Also F3 is applied in the range -50MW to +50MW to calculate the amount of reactive power
generated or absorbed by each of the D-STATCOM as follows.
1000( (1) 50) 1000( (2) 50) 1000( (3) 50) 1000( 50 (2)) 1000( 50 (3))1000( 50 (1))
3
X x x x xx
ee e e e eF
       
      )5(
After determining the amount and placement of each D-STATCOM in its bus by GA and then loss of the
entire network is calculated by Newton- Rawson Method will be calculated, subsequently, cost function to
minimize is defined by the genetic algorithm, so, we can write.
1 2 3tF F F F  
The following values are selected to minimize Ft by the GA.
Population size = 40, Mutation function = Gaussian, Mutation scale = 1and Mutation shrink = 1
The simulation results in is shown as below.
Table 1. Simulation results of the GA
Reactive power
output
Number of bus
3138.25.4D-STATCOM.1
23.53485D-STATCOM.2
32.450110D-STATCOM.3
Total system losses with and without the D-STATCOM that is included the power losses in total
transmission lines are shown in Table (2).
Table 2. system losses
(MW)5.1685
Losses Without
STATCOMD-
3.6441 (MW)
Losses with
D-STATCOM
The voltage range of each bus in the system without D-STATCOM and with D-STATCOM is shown in
figure (2) as:

IJAPE ISSN: 2252-8792 
57
Fig.2. Profile voltage in 30-bus system with and without D-STATCOM
As shown in figure (2) existing of three D-STATCOM in the system not only reduces losses, but also
improves the voltage profile and increasing the voltage of all bus in power system.
8. A HYBRID ALGORITHM WITH ACAF AND GA TO DETERMINE THE BOTH VALUE AND
PLACEMENT OF D-STATCOM
So as to determine the Placement of three D-STATCOM in 30bus network, ACA is used and for
determining the amount of reactive power generated or absorbed by each D-STATCOM, GA is utilized at all
stages.
Fig.3. Flowchart of proposed method
As shown in the flowchart in Figure (3), first, a point is chosen randomly between 1 and 3654. This
point is obtained of the equation (1) where this point is the non-recurring location of D-STATCOM at 3
installation bus between 2 to 30 of bus. The either optimum values of reactive power generated or absorbed
of three D-STATCOM is calculated by GA By to reduce the total system losses. Optimization by GA is
applied to the three variables as x (1), x (2) and x (3) which these variables represent each of the reactive
power of D-STATCOM respectively. The GA will be stopped after 200 repetitions. Subsequent to this cycle
(200 repetition), the lowest level of network losses will be stored as best answer.

 ISSN: 2252-8792
58
In addition, a loop is formed that the number (N) of iterations is 1000. Subsequently, to reduce
losses, ACA is employed by smart searching in the best point of the whole network to install three of D-
STATCOM. Two random points near the nest and away the nest is selected to obtain Cost function as below.
3FLossPowerTotalFT  (3)
In all stages the algorithm will be stopped after 1000cycle and also exist 200 iterations of the genetic
algorithm both near the nest and away the nest. As, 400,000 iterations are total reps to reach the optimal
solution.
Simulation results of the proposed algorithm
Total losses of distribution lines are as two modes, one with three D-STATCOM with table (3) conditions,
Table. 3. The results of the proposed algorithm
Reactive power production (MVA)Number of bus
35.98284D-STATCOM 1
45.799810D-STATCOM 2
-16.803815D-STATCOM 3
And the second, without D- STATCOM is given in table (4).
Table.4. Losses of system
6.1685 (MW)Losses Without D-STATCOM
2.7934 (MW)Losses Wit D-STATCOM According Table (3)
As shown in Table (4) the loss rate of a system with three D-STATCOM is dropped to 45.2849.
Since the useful life of a power system is equal to 30 years, significantly reduce the cost will be prepared in
production and transmission of power. System voltage profiles in two different modes, with and without the
three D-STATCOM system is shown in figure (4).
Fig.4. System voltage profiles, with and without the three D-STATCOM
As shown in figure (4), using D-STATCOM in the network not only reduces ohmic losses in the
transmission system, but also significantly improves the voltage profiles.
Comparison of the results of the two methods provided (genetic algorithm and the proposed
algorithm)
The comparison of the two methods is shown in table (5) briefly as:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
0.85
0.9
0.95
1
1.05
1.1
Bus Number
Voltage(pu)
With D-STATCOM
Without D-STATCOM

IJAPE ISSN: 2252-8792 
59
Table 5. Comparison of the results of genetic algorithm and the proposed algorithm
Fig.5.Voltage profiles at three different methods
As shown in figure (5)Voltage profiles in three different states is investigated as, without D-
STATCOM, with D-STATCOM which its properties is calculated by the genetic algorithm and with D-
STATCOM by proposed method.
9. CONCLUSION
Recently, improving of power quality has been considered for compensation of reactive power and
harmonics because to solve the problem of optimum reconfiguration in distribution systems, an optimal
manner has been needed. This paper presents a new approach for optimal manner of distribution systems
which ACO is used to determine the placement of three of D-STATCOM in 30bus network and GA is
utilized for determining the amount of reactive power generated or absorbed by each D-STATCOM.
Installation and utilization of the D-STATCOM in distribution networks leads to especially significant for
network qualities such as reducing of ohmic losses in transmission lines, improve voltage profiles and system
efficiency. Finally, maintenance costs of the D-STATCOM in distribution networks and power systems are
negligible so that the energy savings and economizing will be significant.
REFERENCES
[1] Adya, A., et al. “Application of D-STATCOM for isolated systems, in Tencon 2004 - 2004 Ieee Region 10
Conference, Vols a-D”, Proceedings: Analog and Digital Techniques in Electrical Engineering. Pp. C351-C354,
2004.

 ISSN: 2252-8792
60
[2] Barnes, M., et al. “Power quality improvement for wave energy converters using a D-STATCOM with real energy
storage”, 2004 1st International Conference on Power Electronics Systems and Applications Proceedings, ed.
K.W.E. Cheng. Pp. 72-77, 2004.
[3] Cetin, A., et al. “Reactive power compensation of coal conveyor belt drives by using d-statcoms”, in Conference
Record of the 2007 Ieee Industry Applications Conference Forty-Second Ias Annual Meeting, Vol. 1-5. Pp. 1731-
1740, 2007.
[4] Cai, R., et al. “Control of D-STATCOM for voltage dip mitigation”, 2005 International Conference on Future
Power Systems, Pp. 126-131, 2005.
[5] Blazic, B. and I. Papic. “A new mathematical model and control of D-StatCom for operation under unbalanced
conditions”, Electric Power Systems Research, Vol/Issue: 72(3). Pp. 279-287, 2004.
[6] Somsai, K., T. Kulworawanichpong, and Ieee. “Modeling, Simulation and Control of D-STATCOM using
ATP/EMTP”, in 2008 13th International Conference on Harmonics and Quality of Power, Vol. 1 and 2. Pp. 377-
380. 2008.
[7] Niknam, T., H.Z. Meymand, and M. Nayeripour. “A practical algorithm for optimal operation management of
distribution network including fuel cell power plants”, Renewable Energy, Vol/Issue: 35(8). Pp. 1696-1714, 2010.
[8] Noroozian, R. “A Performance Comparison of D-STATCOM and DC Distribution System for Unbalanced Load
Compensation”, International Review of Electrical Engineering-Iree, Vol/Issue: 7(2). Pp. 4194-4207, 2012.
[9] Barukcic, M., S. Nikolovski, and F. Jovic. “Hybrid evolutionary-heuristic algorithm for capacitor banks allocation”,
Journal of Electrical Engineering-Elektrotechnicky Casopis, Vol/Issue: 61(6). Pp. 332-340, 2010.
[10] Secui, D.C., et al. “An ACA Algorithm for Optimal Capacitor Banks Placement in Power Distribution Networks”,
Studies in Informatics and Control, Vol/Issue: 18(4). Pp. 305-314, 2009.
[11] Padmanaban, K.P. and G. Prabhaharan. “Dynamic analysis on optimal placement of fixturing elements using
evolutionary techniques”, International Journal of Production Research, Vol/Issue: 46(15). Pp. 4177-4214, 2008.
[12] Su, C.T., C.F. Chang, and J.P. Chiou. “Optimal capacitor placement in distribution systems employing ant colony
search algorithm”, Electric Power Components and Systems, Vol/Issue: 33(8). Pp. 931-946, 2005.
[13] Chang, C.-F. “Reconfiguration and Capacitor Placement for Loss Reduction of Distribution Systems by Ant Colony
Search Algorithm”, Ieee Transactions on Power Systems, Vol/Issue: 23(4). Pp. 1747-1755, 2008.
[14] Atighehchian, A., M. Bijari, and H. Tarkesh. “A novel hybrid algorithm for scheduling steel-making continuous
casting production”, Computers & Operations Research, Vol/Issue: 36(8). Pp. 2450-2461, 2009.
[15] Sedki, A. and D. Ouazar. “Hybrid particle swarm optimization and differential evolution for optimal design of water
distribution systems”, Advanced Engineering Informatics, Vol/Issue: 26(3). Pp. 582-591, 2012.
[16] Saffar, A., R. Hooshmand, and A. Khodabakhshian. “A new fuzzy optimal reconfiguration of distribution systems
for loss reduction and load balancing using ant colony search-based algorithm”, Applied Soft Computing, Vol/Issue:
11(5). Pp. 4021-4028, 2011.
[17] Valenzuela, C., et al. “A 2-level Metaheuristic for the Set Covering Problem”, International Journal of Computers
Communications & Control, Vol/Issue: 7(2). Pp. 377-387, 2012.
[18] Mariun, N., et al. “Design of a prototype D-statcom using DSP controller for voltage sag mitigation’, IEEE Power
India Conference, Vol. 1 and 2. Pp. 727-732, 2006.
[19] Xi, Z., et al. “Improving Distribution System Performance with Integrated STATCOM and Supercapacitor Energy
Storage System”, in 2008 Ieee Power Electronics Specialists Conference, Vol. 1-10. Pp. 1390-1395, 2008.
[20] Parkhideh, B., et al. “Integration of Supercapacitor with STATCOM for Electric Arc Furnace Flicker Mitigation”, in
2008 Ieee Power Electronics Specialists Conference, Vol. 1-10. Pp. 2242-2247, 2008.
[21] Coteli, R., et al. “Phase Angle Control of Three Level Inverter Based D-STATCOM Using Neuro-Fuzzy
Controller”, Advances in Electrical and Computer Engineering, Vol/Issue: 12(1). Pp. 77-84, 2012.
[22] Coteli, R., et al. “Three-level Cascaded Inverter Based D-STATCOM Using Decoupled Indirect Current Control”,
Iete Journal of Research, Vol/Issue: 57(3). Pp. 207-214, 2011.
[23] Afshar, M.H. “Large scale reservoir operation by Constrained Particle Swarm Optimization algorithms”, Journal of
Hydro-Environment Research, Vol/Issue: 6(1). Pp. 75-87, 2012.
[24] Almeder, C. and L. Moench. “Metaheuristics for scheduling jobs with incompatible families on parallel batching
machines”, Journal of the Operational Research Society, Vol/Issue: 62(12). Pp. 2083-2096, 2011.
[25] Bhaskaran, K., et al. “Dynamic Anycast Routing and Wavelength Assignment in WDM Networks Using Ant
Colony Optimization (ACO)”, in 2011 Ieee International Conference on Communications, 2011.
[26] Shi, B., et al. “A Hybrid Genetic &Ant-colony Algorithm for Fuzzy Petri Net Parameter Optimization Problems”, in
Mechanical, Industrial, and Manufacturing Engineering, M. Ma, Editor. Pp. 565-568, 2011.
[27] Benbouzid-Sitayeb, F., et al. “An Integrated ACO Approach for the Joint Production and Preventive Maintenance
Scheduling Problem in the Flowshop Sequencing Problem”, 2008 Ieee International Symposium on Industrial
Electronics, Vol. 1-5. Pp. 1541-1546, 2008.
[28] Chandrasekaran, C., et al. “Metaheuristics for solving economic lot scheduling problems (ELSP) using time-varying
lot-sizes approach”, European Journal of Industrial Engineering, Vol/Issue: 1(2). Pp. 152-181, 2007.
[29] Li, A. and F. Bai. “The R-TSP model and its application in the VLSI floor plan”, in Proceedings of the Fifth
International Conference on Information and Management Sciences, H.T. Nguyen, X. Zhao, and J. Peng, Editors.
Pp. 585-587, 2006.

Optimal Placement of D-STATCOM Using Hybrid Genetic and Ant Colony Algorithm to Losses Reduction

More Related Content

What's hot (20)

Similar to Optimal Placement of D-STATCOM Using Hybrid Genetic and Ant Colony Algorithm to Losses Reduction (20)

More from IJAPEJOURNAL (20)

Recently uploaded (20)

Optimal Placement of D-STATCOM Using Hybrid Genetic and Ant Colony Algorithm to Losses Reduction