International Journal of Applied Engineering Research ISSN 0973-4562 Volume 13, Number 21 (2018) pp.

14886-14892
© Research India Publications. http://www.ripublication.com

Optimization Algorithms for Solving Microgrid and Smart Grid Integration

Problems

William Sanchez-Huertas1, Víctor Gómez1, Cesar Hernández1*

1
Universidad Distrital Francisco José de Caldas, Technological Faculty, Bogotá, Colombia.

*
Corresponding author, 1*Orcid: 0000-0001-9409-8341

Abstract the discharge of the transmission lines. However, some aspects

still need to be improved like the continuity of the service
Smart Grid does not have a universally accepted definition, it
which is interrupted in the case of solar energy since radiation
can be described both in simple terms and in more complex
is not present during all 24 hours of the day and the speed of
forms. The evolution of conventional grids to Smart grids and
the wind is discontinuous regarding wind energy. Microgrids
the integration of distributed generation and microgrids have
(MG) are hence implemented which are electrical systems that
challenges such as generation forecasts, intelligent network
can operate in island mode, being totally independent from the
management, determining the location, size and quantity of
electric network or work in parallel to serve as backup units [4].
non-conventional sources of energy. This paper presents the
optimization algorithms such as the Colony of Ants, Genetic Microgrids are constituted by these elements: generation units,
Algorithm, Particle Swarm Optimization Algorithm, Bee distributed storage units, a power converter and a control
Colony Algorithm and Tabu Search Algorithm, as the best system [5].
solution to the presented challenges. To achieve the above, an
Distributed resources are highly sought after due to their
application research was carried out in the fields of energy
versatility and are installed in distribution systems to reduce the
efficiency, network topology and distributed energy resources
greenhouse effect. Their main disadvantage lies in the
based on these algorithms. Given that the results have been
discontinuity of their generation which is why a large-scale
satisfactory, it is proposed to continue with this line of research
power generation system is proposed to transform these
so that these technologies are more accessible to all users.
technologies into technically competitive strategies. As the
Keywords: Algorithms, integration, microgrid, optimization, demand for electric systems keeps growing, it is necessary to
smart grid investigate on the integration of DER in order to guarantee a
service with high quality, efficiency and continuity [6].
The smart grid (SG) does not have a universal definition and
INTRODUCTION
can be described both in simple and complex terms. The term
Currently, electric power systems have similar characteristics ‘smart’ implies that its operation is automation-based while the
all over the world, such as a centralized generation at a distance concept ‘grid’ is related to a network of electric conductors that
of several kilometers from the consumption centers, an provide electricity for certain points. That said, an idea can be
infrastructure that has been used from 30 to 40 years, a growing conceived on what a smart grid is. It used to be dream or an
energy demand and a constant concern for the environment [1]. idea but nowadays it is one of the most discussed topics in the
modern electric sector. This modern network is capable of
There is a worldwide energy dependency on oil, carbon and
storing, communicating and making decisions. A SG
natural gas related to a geographical distribution which has
transforms current networks into more cooperative and
prompted nations to diversify their national energy basket and responding systems within a well-organized structure [7].
seek renewable energy sources for an environment-friendly According to the strategic document of future electric networks
generation that reduces greenhouse effect emissions and
in Europe, a SG is an electric network that can intelligently
mitigates climate change [2]. Non-conventional renewable
integrate the actions from all users (generators, consumers and
energy sources include wind energy, photovoltaic solar energy,
mixed) connected to it to obtain electricity in an efficient,
biomass energy, geothermic energy, tidal energy, among affordable and secure manner [5].
others. They can generate electricity, heating and fuel as end
products. The National Institute of Standards and Technology (NIST)
proposed to divide the smart grid into seven domains which
Nowadays, conventional electric systems are facing a wide
include actors and applications. On one hand, the actors can be
range of problems such as overcharged transmission lines, high
devices, systems or programs that make decisions and
generation costs and variety in nodal voltages. In response,
exchange necessary information to perform applications. They
Distributed Generation (DG) has been successfully can be smart measuring devices, solar generators and control
implemented and its resources, named Distributed Energy systems. On the other hand, the corresponding applications can
Resources (DER), can solve or minimize some of the
be domotics, solar energy generation, energy storage and
aforementioned problems [3]. For instance, the installation of
energy management [8]. Figure 1 shows the SG model
small generation centers close to the consumption spots enables
presented by NIST.

14886
 International Journal of Applied Engineering Research ISSN 0973-4562 Volume 13, Number 21 (2018) pp. 14886-14892
© Research India Publications. http://www.ripublication.com

Additionally, an energy management system is required for there would not be a continuous and reliable service. Therefore,
residential, industrial and commercial users that adjust to the the Distributed Hybrid Energy System (DHES) offers to
needs of incorporating DG, DER, MG and SG. This integrate DER and distributed energy storage technologies
management unit is called Distribution Management System (such as battery systems) that can store enough energy and can
(DMS) and comes from the need to implement technological be used in scarcity periods [12].
support that makes decisions and enables an effective and
Management of Smart Grids
efficient integration [9].
The management of the network is analyzed from four points
The goals of the DMS are to control, optimize, analyze and plan
of view: generation, transmission, distribution, management
the network’s operation as well as design the supply chain.
and control of the electric network. The challenges found in
each aspect show some barriers that are arise in the migration
from traditional electric networks to smart grids.
Generation
When microgrids (MG) are integrated into the electrical
system, the offered energy must be planned from the dispatch
centers whether it is produced from microgrids or large
stations. In the case of Colombia, in [2] a modification of the
offering mechanisms is proposed along with the creation of an
intradaily market. The purpose is to plan paths of action
oriented towards creating adequate and favorable conditions for
the participation of MG and harness DER. This offer will be
made based on the hourly generation forecasts.
Additionally, MG that cannot generate energy over periods of
time shall receive support from power systems. These changes
in the state of MG shall be made automatically in order to
guarantee an electricity service of quality for the users [13].
Figure 1. Interaction between actors in different domains of
the Smart Grid through Secure Communication Flows and Transmission
Electric Flows. Source [8] The different nodes of the electric system shall include
interconnected DG systems that shall produce technical issues
for the transportation and distribution networks. Such issues
shall be related with difficulties in voltage control, reactive
PROBLEMS FOR SMART GRID AND MICROGRID power management, reduction of the efficacy in electric
INTEGRATION INTO THE CONVENTIONAL protection strategies plus the negative impact in energy quality.
ELETRIC NETWORK Perhaps the biggest challenge that limits the proliferation of DG
lies in the control of the voltage levels in the nodes [14][15].
This section discusses the challenges of determining the This implies that an adequate operation is necessary to remotely
location, size and configuration of a renewable power plant, the control the voltage of nodes in order to maintain the balance of
generation forecasts and the management and control in the the electric system and assure that the power parameters are
integration of SG and MG into the conventional electric within the limits set by different standards.
network. Said challenges are now described.
Distribution
Location, size and configuration of a renewable power
plant Distribution networks were designed for unidirectional energy
flows and not to ease the penetration of DG. As a result, the
To determine the location, size and configuration of renewable distribution units that are stationed in inappropriate places,
power plants, algorithms are required to optimize the available where generation does not meet the demand or surpasses the
area since one of the characteristics of renewable power plants network’s capacity, can have negative effects such as reverse
is that they take on a large extension of terrain in order to power flows, increases in voltage, degradation of the quality of
generate some megawatts (MW). Optimization algorithms can the voltage waveform, injection of harmonic components in the
determine the location and size of a power plant in places where network or destabilization of the network [16][17].
the weather information is scarce [10].
The installation of bidirectional Smart measuring devices
Generation forecast allows the identification of what has been consumed and what
The generation forecast becomes a challenge due to the has been generated, in the different periods of time for the
variability of the resource used to generate. For instance, the corresponding billing. Therefore, network operators can
periods of time when the wind flow is continuous cannot be properly charge for the real consumption of each participant of
predicted for wind energy systems or when solar radiation is the system.
optimal for photovoltaic solar systems [11]. This uncertainty in
the prediction process turns systems into inefficient because

14887
 International Journal of Applied Engineering Research ISSN 0973-4562 Volume 13, Number 21 (2018) pp. 14886-14892
© Research India Publications. http://www.ripublication.com

The distribution of energy involves the term ‘prosumer’ which

is the figure used to name the small system which is both
producer and consumer of energy and that is connected to the
electric network.
Management and control of the electric network
In every point of the system, the information from electric
variables is being registered in real time. Therefore, it must be
recollected in order to establish the users’ behaviors and the
profiles of the electric network so to optimize the network
management. This involves an evolution of the network
management based on centralized systems (SCADA) into
decentralized or distributed control systems to assure optimal
levels of efficiency, security and reliability [18]. The amount of
information is so large that it needs to be handled by
technologies such as Big Data.
In tandem with the electric network, information transmission Figure 2. Ant colony optimization. Source [21]
networks must exist that are generated by remotely controlled
units so that they can operate in case of failure or maintenance
to the infrastructure. This algorithm was developed by Marco Dorigo in 1992 [22],
and is based in the natural behavior of ants while they are
looking for food and seek an optimal path. ACO had the initial
OPERATIONAL RESEARCH purpose of solving the travelling salesman problem (TSP)
which consists on visiting several cities in the shortest possible
Operational Research is an interdisciplinary modern science time by passing only once per city, except for the origin and the
that adopts theory and special techniques to seek the optimal final destination that can be visited twice. Although the results
solution of complex problems related to the management, were good, ACO did not show a better performance than other
organization and control of the systems existing in nature and algorithms used to solve the TSP [22]. However, it performs
those created by mankind (organized systems, physical well in sequential orders, balance in the assembly line and in
systems, economic, ecological, educational, social, etc.) for fabrication and logistics problems involving programming
optimal decision-making [19]. [23].
Informally, optimizing means that something more than The kth ant moves from state x to y with the probability
improving. However, in the scientific context, optimization is described in equation (1)
the process of trying to find the best possible solution to a 𝛽
𝛼
specific problem. In an optimization problem, there are 𝑘
(𝜏𝑥𝑦 )(𝜂𝑥𝑦 )
𝑝𝑥𝑦 = (1)
different solutions and often the criterion to set them apart 𝛼 )(𝜂 𝛽 )
∑(𝜏𝑥𝑦 𝑥𝑦
consists on determining the best one [20].
Where 𝜏𝑥𝑦 is the number of pheromones deposited in the
transition to state x to y, α ≥ 0 is a parameter used to control the
OPTIMIZATION ALGORITHM influence of 𝜏𝑥𝑦 , 𝜂𝑥𝑦 is the convenience of state xy (an a priori
Next, the most relevant optimization algorithms in current knowledge, typically 1 / dx with d as the distance) and β ≥ 1 is
literature are analyzed. a parameter to control the influence of ηxy.

Ant Colony Optimization (ACO) When all ants have completed the solution, the traces are
updated according to equation (2).
In general, ants tend to leave pheromones in the environment
𝑘
when looking for food. When they manage to find food, they 𝜏𝑥𝑦 ← (1 − p)𝜏𝑥𝑦 + ∑ 𝛥𝜏𝑥𝑦 (2)
deploy even more pheromones on their way back to the nest 𝑘
with the purpose of guiding the other ants. The next ants will
Where, 𝜏𝑥𝑦 is the number of pheromones deposited for a
then optimize the path since they will take the path that requires 𝑘
the lowest effort. The pheromones gradually disappear and the transition state xy, p is the pheromone coefficient and 𝛥𝜏𝑥𝑦 is
th
most recent cells remain. Figure 3 describes the optimization the number of pheromones deposited by the k ant, often given
process with the ant colony method. by the Travelling Salesman Problem.

Genetic Algorithm (GA)

Genetic algorithms are optimization, search and learning
algorithms inspired in the processes of natural evolution and
genetic evolution. From an initial population, the most capable
individuals are chosen to reproduce and mutate, and finally

14888
 International Journal of Applied Engineering Research ISSN 0973-4562 Volume 13, Number 21 (2018) pp. 14886-14892
© Research India Publications. http://www.ripublication.com

obtain the following generation of individuals that is more with no restrictions, although it can be used to solve
adapted to survive than the previous generation [24]. combinatory problems [28].
The genes in GA are the design variables and the chromosome There are three classifications for bees: worker, spectator and
represents a potential solution. Possible solutions are saved explorer. First, the worker bees seek food and when they find
after each iteration. When calculating the value of the physical it, they return to the hive. Then, the spectator bees can trace
aptitude (that is determined through an assessment of the high quality sources of food. When the recollection ends, the
individuals of a population), two individuals are chosen to be worker bee joins the explorer bees that constantly seek food.
modified with genetic operators, including crossing and The spectator and worker bees are modelled with the
mutation, to create a new generation [25]. exploitation method while the explorer bees are modelled with
the exploration technique. The catch in the BCA method is that
Figure 3 shows the operation of a basic genetic algorithm.
it is good for exploration yet it lacks the exploitation factor [25].
Taboo Search (TS)
The taboo search (TS) introduced by Glover in 1986 is a meta-
heuristic approach that tackles combinatory optimization
problems. More precisely, TS is an extension of local search
classic methods that can be interpreted as the combination of
local search with short term memories. Two basic elements of
TS are the definition of its search space and its neighborhood
structure. The search space is defined as the space containing
all possible solutions that can be visited during the search
process. Regarding the neighborhood structure in each iteration
of the algorithm, the local transformation that can be applied to
the current solution defines a set of neighboring solutions in the
search space, i.e. the neighbor of the current solution. In fact,
the environment of the current solution is a subset of the search
space of all possible solutions obtained by adopting a unique
local transformation for the present solution [25].
Figure 3. Genetic algorithm i: Initialization, f(X):
Assessment, ?: End condition, Se: Selection, Cr: Crossing, Meta-heuristic hybrid method
Mu: Mutation, Re: Replacement, X*: Best solution.
The implementation of two or more optimization algorithms
Source [26]
carries superior advantages in comparison to the individual
implementation. Therefore, the hybrid implementation has had
a growing trend in its use. However, hybrid algorithms can be
Particle Swarm Optimization (PSO)
a hybrid of two or more meta-heuristics or a hybrid of meta-
The optimization with particle swarm, better known in heuristics and traditional optimization approaches [29].
scientific literature with the PSO acronym is a stochastic
method of evolutional computation inspired in the social
behavior of organisms such as flocks of birds or school of fishes APPLICATIONS
[26].
In [30], Roberto and Thiago present an algorithm based in ACO
Particles represent possible solutions. PSO are based on for the optimal allocation of distributed generation in medium
swarms and uses a set of particles to seek the best solutions. power electric networks. The problem is solved with the
Then, the best experience of particles and swarms is followed purpose of minimizing the real power losses, considering the
in which the algorithm recollects the best location and speed of installation costs of the generators. The ant algorithm has been
each particle. The movement of the particles is followed as the tested in a radial distribution system of 33 energized bars at
particles move until the end criterion is met, which is an optimal 12.66 kV (see Figure 4) and nine options of distributed
acceptable solution or a given number of iterations [27]. generation that can be used several times throughout all 32 bars
(without including the bar linked to the substation). The results
Bee Colony Algorithm (BCA)
show the efficiency of the methodology presented by managing
The algorithm of artificial bee colonies was proposed by Dervis to reduce the losses in active energy by a 59.66%. Furthermore,
Karaboga in 2005 and is based on the foraging behavior of bees its robustness is shown with a good quality solution,
and originally designed for numerical optimization problems significantly reducing the losses and improving the voltage
profile.

14889
 International Journal of Applied Engineering Research ISSN 0973-4562 Volume 13, Number 21 (2018) pp. 14886-14892
© Research India Publications. http://www.ripublication.com

Figure 4. System of 33 bars. Source [30]

In [31], Wang Pu plans to reduce the energy consumption of a algorithm is used to solve the optimization problem in real time,
train line between the Yizhuang and Beijing stations, which are while the limitations of the distribution system, the
separated by 2682 meters and have an execution time of 185 photovoltaic inverters and the PHEV are considered. The
seconds. An optimization method based in a golden ratio simulation results in the distribution system with 33 bars prove
genetic algorithm is proposed. In first place, the train model is the effectiveness of the proposed method in mitigating the
established. In second place, the optimal operation strategy of voltage fluctuations for a PV generation profile with strong
the subway train is analyzed with different tracks. Then, the variations.
golden ratio genetic algorithm (GR-GA) is proposed to solve
the problem, seeking the optimal energy transfer position for
the train. The results show that the algorithm has a better CONCLUSIONS
optimization result in comparison to the traditional genetic
Optimization algorithms are an important tool in the search for
approach. It improves savings by 14.2%.
technical solutions and financial improvements in different
Furthermore, in [32], Elbaset defines the optimal and efficient scenarios and integration processes of distributed generation
design of a wind energy park in a determined territory into current electric networks and smart grids. Therefore,
interconnected to the electric network. This implies several scientists are recommended to proceed with these
complex optimization problems. Multi-target functions are developments in order to make these technologies more
shown, based on PSO techniques for an optimal distribution of accessible for all users and promote the efficient use of energy.
the farm in a place near Zaafrana, in the Suez Golf. To begin
with, the target function has to find an optimal number of wind
turbines to produce a specific annual demand of energy. Then, ACKNOWLEDGEMENTS
the effect of the trail is minimized to achieve an optimal
placement of the wind turbines based on binary PSO. The authors of this article wish to acknowledge Colciencias and
Furthermore, the optimal number of suppliers is calculated in the Universidad Distrital Francisco José de Caldas for funding
order to minimize the generated cost per kilowatt-hour. The resources to develop this research.
trail loss model is represented by the model based on Jensen
and the economic model is based on the concept of annual cost
of the system. A complete numerical example is detailed within REFERENCES
an industrial framework. [1] C. Virtual, H. Sarmiento Uruchurtu, and R. Nieva
Moving on to another focus, the intermittent nature of Gómez, “" IEEE Day " Technical Chat
photovoltaic (PV) solar sources’ output generation due to " Smart Grid " Conferencia Virtual Redes
transitory clouds leads to considerable fluctuations to be Eléctricas Inteligentes Senior Member,” 2010.
expected in the voltages of distribution systems with high [2] UPME, “Integración de las energías renovables no
penetrations of distributed generation using PV energy. convencionales en Colombia,” Bogotá, 2015.
Abdelfatah Ali in [33] proposes a method to mitigate voltage
fluctuations caused by photovoltaic energy when [3] A. K. Basu, S. P. Chowdhury, S. Chowdhury, and S.
simultaneously optimizing the reactive power from Paul, “Microgrids: Energy management by strategic
photovoltaic inverters and the charge/discharge of pluggable deployment of DERs - A comprehensive survey,”
hybrid electric vehicles (PHEV). To effectively mitigate the Renewable and Sustainable Energy Reviews, vol. 15, no.
voltage fluctuations, Abdelfatah presents a new target function 9. Pergamon, pp. 4348–4356, 01-Dec-2011.
that represents the average deviation in voltage. These new
objective greatly reduces voltage fluctuations. The BCA

14890
 International Journal of Applied Engineering Research ISSN 0973-4562 Volume 13, Number 21 (2018) pp. 14886-14892
© Research India Publications. http://www.ripublication.com

[4] C. Hernandez, V. Gómez, and E. Rivas, “Distributed resources: A review of the state-of-the-art,” Renew.
Energetic Resources: Trend, Control and Operation,” Sustain. Energy Rev., vol. 14, no. 5, pp. 1353–1366,
Int. J. Eng. Technol., vol. 9, no. 5, pp. 3553–3561, 2017. 2010.
[5] F. De Ingeniería, U. Distrital, and F. José, “La Influencia [17] V. Gomez, C. Hernandez, and E. Rivas, Desarrollo,
de los Niveles de Penetración de la Generación políticas y futuro de las microredes inteligentes, Primera
Distribuida en los Mercados Energéticos The Influence Ed. Bogotá: Universidad Distrital Francisco José de
of Penetration Levels of Distributed Generation in the Caldas, 2018.
Energy Markets,” vol. 2, no. 1.
[18] C. Yuen, D. Botting, A. D. B. Paice, J. Finney, and O.
[6] and A. S. S. Chris Marnay, F Javier Rubio, “Shape of Preiss, “Cuando las redes se vuelven inteligentes
the Microgrid,” Ernest Orlando Lawrence Berkeley Automatización inteligente de las redes de distribución,”
Natl. Lab., 2000. Rev. ABB, vol. 1, 2008.
[7] B. D. Solomon and K. Krishna, “The coming sustainable [19] A. Vidal, L. Carpente, and B. Casas Méndez,
energy transition: History, strategies, and outlook,” “ALGORITMOS HEURISTICOS EN
Energy Policy, vol. 39, no. 11, pp. 7422–7431, 2011. OPTIMIZACION,” 2013.
[8] G. Locke and P. D. Gallagher, “NIST Special [20] L. A. Arito, G. Leguizamón, and M. Errecalde,
Publication 1108 NIST Framework and Roadmap for “Algoritmos de Optimización basados en colonia de
Smart Grid Interoperability Standards, Release 1.0 NIST Hormigas aplicados al problema de asignación
Framework and Roadmap for Smart Grid cuadrática y otros problemas relacionados,” 2010.
Interoperability Standards, Release 1.0,” 2010.
[21] Johann Dréo, “El camino más corto encontrado por una
[9] V. A. Gómez, C. Hernandez, and E. Rivas, colonia de hormigas,” 2006. [Online]. Available:
“Management of Distributed Energetic Resources,” vol. https://commons.wikimedia.org/wiki/File:Aco_branche
12, no. 20, pp. 9506–9514, 2017. s.svg. [Accessed: 24-Jun-2018].
[10] K. S. Perera, Z. Aung, and W. L. Woon, “Machine [22] A. Soifer and I. Loiseau, “Algoritmos de Colonia de
Learning Techniques for Supporting Renewable Energy Hormigas para el Problema del Viajante de Comercio
Generation and Integration: A Survey,” Lect. Notes por Familias y para el Problema de Ruteo de Veh´ıculos
Comput. Sci. (including Subser. Lect. Notes Artif. Intell. por Familias,” 2015.
Lect. Notes Bioinformatics), vol. 8817, pp. 81–96, 2014.
[23] C. Blum, “Beam-ACO—hybridizing ant colony
[11] B. K. Bala and S. A. Siddique, “Optimal design of a PV- optimization with beam search: an application to open
diesel hybrid system for electrification of an isolated shop scheduling,” Comput. Oper. Res., vol. 32, no. 6, pp.
island—Sandwip in Bangladesh using genetic 1565–1591, Jun. 2005.
algorithm,” Energy Sustain. Dev., vol. 13, no. 3, pp.
[24] A. de la Peña and P. Truyol, “ALGORITMOS
137–142, 2009.
GENÉTICOS.”
[12] K. Kusakana, “Optimal scheduling for distributed
[25] K. K. Castillo-Villar, “Metaheuristic algorithms applied
hybrid system with pumped hydro storage,” Energy
to bioenergy supply chain problems: Theory, review,
Convers. Manag., vol. 111, pp. 253–260, Mar. 2016.
challenges, and future,” Energies, vol. 7, no. 11, pp.
[13] H. A. Rahman, M. S. Majid, A. Rezaee Jordehi, G. Chin 7640–7672, 2014.
Kim, M. Y. Hassan, and S. O. Fadhl, “Operation and
[26] Johann Dréo, “Evolutionary algorithm.svg - Wikimedia
control strategies of integrated distributed energy
Commons.” [Online]. Available:
resources: A review,” Renew. Sustain. Energy Rev., vol.
https://commons.wikimedia.org/wiki/File:Evolutionary
51, pp. 1412–1420, Nov. 2015.
_algorithm.svg. [Accessed: 24-Jun-2018].
[14] W. El-Khattam and M. M. . Salama, “Distributed
[27] M. E. H. Pedersen and A. J. Chipperfield, “Simplifying
generation technologies, definitions and benefits,”
Particle Swarm Optimization,” Appl. Soft Comput., vol.
Electr. Power Syst. Res., vol. 71, no. 2, pp. 119–128,
10, no. 2, pp. 618–628, Mar. 2010.
Oct. 2004.
[28] T. De, M. Adán, and E. Aguilar Justo, “Un algoritmo
[15] I. Benoit, “CENTER FOR ECONOMIC STUDIES
basado en la colonia articial de abejas con búsqueda
WORKING PAPER SERIES DISTRIBUTED
local para resolver problemas de optimización con
GENERATION : DEFINITION , BENEFITS AND
restricciones.”
ISSUES Pepermans G ., Driesen J ., Haeseldonckx D .,
D ’ haeseleer W ., Belmans R . secretariat : [29] Michael Gendreau and Jean-Yves Potvin, Handbook of
DISTRIBUTED GENERATION : DEFINITION , Metaheuristics, Second Edition. 2010.
BENEFITS AND ISSUES,” Energy Policy, vol. 32, no.
[30] N. R. Ferreira and A. de C. Thiago Lopes, “Optimal
0, pp. 0–21, 2003.
Allocation of Distributed Generation using Ant Colony
[16] A. Alarcon-Rodriguez, G. Ault, and S. Galloway, Optimization in Electrical Distribution System,” 2018
“Multi-objective planning of distributed energy Simp. Bras. Sist. Eletr., pp. 1–6, 2018.

14891
 International Journal of Applied Engineering Research ISSN 0973-4562 Volume 13, Number 21 (2018) pp. 14886-14892
© Research India Publications. http://www.ripublication.com

[31] W. Pu, X. Gao, S. Ding, and H. Gao, “Optimization of

train energy saving based on golden ratio genetic
algorithm,” 2018 33rd Youth Acad. Annu. Conf. Chinese
Assoc. Autom., pp. 869–874, 2018.
[32] A. A. Elbaset, M. M. Ismail, and A. H. K. Alaboudy,
“Particle Swarm Optimization for Layout Design of
Utility Interconnected Wind Parks,” 2018 IEEE Power
Energy Soc. Innov. Smart Grid Technol. Conf., pp. 1–6,
2018.
[33] A. Ali, D. Raisz, and K. Mahmoud, “Mitigation of
Voltage Fluctuation in Distribution System Connected
with PV and PHEVs Using Artificial Bee Colony
Algorithm,” 6th Int. Istanbul Smart Grid Cities Congr.
Fair, pp. 144–148, 2018.

14892