IRJET- Demand Response Optimization using Genetic Algorithm and Particle Swarm Optimization

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 06 | June 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 1340
Demand Response Optimization using Genetic Algorithm and Particle
Swarm Optimization
---------------------------------------------------------------------***----------------------------------------------------------------------
Abstract – Due to increase in population growth
demand is also increasing and it is the main issue for
utility company to balance consumption and production.
In the past problem with insufficient capacity in the grid
often affected the supply side only and it was solved by
capacity additions, but due to increasing power
generating units its affected both environment and
economy. And the nightmares will come when peak period
will come the demand side response come into play. So,
when peak period will come the demand side response
comes into play. In demand side response consumer will
take part and reduces the load at peak period of time.
Smart meter is adopted on demand side so that it gives
information to the consumer will take part and reduces
the load at peak period of time. Smart meter is adopted in
demand side response comes into play. In demand side
response consumer will take part and educes the load at
peak period of time. Smart meter is adopted on demand
side response comes into play. In demand side response
consumer will take part and reduces the load at peak
period of time. Smart meter is adopted on demand side so
that it gives information to the consumer of their
electricity used and accordingly they reduce their load. In
its total load is divided into two part-elastic and inelastic,
and by controlling elastic load total load will be controlled
and reduce peak demand and also reduce total fuel cost
and electricity bills of consumer. In it one day electricity
used by customer is assumed to conduct simulations. As
seen from the result, it is found that through elastic load
the use of time interval changes, minimum tariff objective
can be reached. And by comparing the result for both GA
and PSO it is seen that GA gives better results as compared
to PSO.
Keywords: Demand Side Management, Optimization,
Genetic Algorithm, Particle Swarm Optimization
1. INTRODUCTION
Peak load demand situations in grid-based energy supply
systems, such as electricity, natural gas and district
heating, present particular challenges for the generation
and transmission of the energy demanded. Each system
has to be designed to give uninterrupted service to
consumers, within the terms of the particular
agreements and tariffs chosen.
In this thesis, electricity use and supply is the major
interest. The key components of a liberalized electricity
system can be clarified in a following way:
1. Electricity is generated by Producers.
2. The Network Companies (also called ‘network
owners’) are transmitting the electrical energy from the
production plants to the Consumers (also here called
‘Customers’) via the national grid, the regional networks
and the local networks. The regional networks transmit
power from the grid to the local networks and
sometimes to large consumers, e.g. industries. The local
networks distribute power to the consumers.
3. The financial transactions are undertaken by Suppliers
(also called ‘Supply Companies’); these competing
companies buy energy from Producers and sell to
Consumers. Each of the Supplier and Network Company
may be also called a Utility.
4. National grid operator, is responsible for the national
grid and has the role of system operator. This means
ensuring that production/imports correspond to
consumption/exports and that the power system works
in a reliable way.
Load demand is an especially sensitive factor in the
electricity supply system. Demand (consumption) and
supply (production) should be constantly balanced in
order to avoid supply interruptions with all their
negative technical, economic and social consequences.
If electricity storage is connected in the system (e.g. as
pumped hydroelectricity, compressed air storage), then
this is a load when being charged, and a supply when
being discharged.
Problems with insufficient capacity in the grid in the past
often were addressed at the supply side (energy
generators and suppliers) and solved by capacity
additions. According to Bellarmine (2000), “In
generating power the concept has been straightforward.
If the society demanded more power, the power
companies would simply find a way to supply users even
by building more generation facilities. This concept of
doing business has been labelled as supply-side
management.” This was a wide-spread opinion within
the energy industry. However, such continuous
expansion nowadays would hardly be compatible with a
target of sustainable energy systems.
The liberalization of electricity markets brought up new
concerns about generation and transmission capacities
in many countries. Growing competition among power
producers forced them to optimize production and

decrease their internal costs. After the liberalization of
electricity markets, the focus for solving peak load
problems has moved more from the supplier/utility side
towards the demand side/consumer.
When discussing load reduction activities, we can note
that different actors in the electricity market have
different interests in peak load demand reduction, seen
from technical, economic, environmental and social
perspectives.
2. LITERATURE REVIEW
This literature survey intended to provide information to
understand the context of this research. Proposed
architectural model, control strategies and different
methods discussed briefly.
Yamille Del Valle [2008]: In this paper presents a
detailed overview of the basics concepts of PSO and its
variants .Also, it provides a comprehensive survey on the
power system applications that have benefited from the
powerful nature of PSO as an optimization technique
.For each application, technical details that are required
for applying PSO, such as its type, particle formulation
,and the most efficient fitness function are also
discussed.
Jian Jiao [2010]: In this paper present a overview of the
basics concepts of PSO according to continuous PSO and
discrete PSO. The difference between single objective
PSO and multi objective PSO is presented. At the same
time an implementation of PSO in multi objective
optimization is discussed. To overcome the limitations of
PSO, hybrid optimization algorithms are proposed.
Zhiyu You [2010]: In this paper presents an adaptive
weight Particle Swarm Optimization algorithm with
constriction factor is proposed combined with an
analysis of convergence of Particle Swarm Optimization
algorithm .The value of the inertia weight is set
according to dynamic information about the changes in
the objective function value, as to effectively balance the
advantages of global optimization against the shortage of
local optimization .
M. Marwan [2006]: In this paper presents a demand
side response scheme, which assists electricity
consumers to proactively control own demands in such a
way to deliberately avert congestion periods on the
electrical network. The scheme allows shifting loads
from peak to low demand periods in an attempt to
flattening the national electricity requirement. The
scheme can be concurrently used to accommodate the
utilization of renewable energy sources, that might be
available at user’s premises. In addition, the scheme
allows a full capacity utilization of the available
electricity infrastructure by organizing a wide –use of
electricity vehicles.
Marwan Marwan [2008]: This research aimed to
develop consumer demand side response model to assist
electricity consumers to mitigate peak demand on the
electrical network. The model developed demand side
response model to allow consumers to manage and
control air conditioning for every period, it is called
intelligent control. The result indicates the potential of
the scheme to achieve energy savings, reducing
electricity bills to the consumer and targeting best
economic performance for electrical generation
distribution and transmission.
Duy Long Ha [2008]: This paper focuses on Demand
side load management applied to residential sector. A
multi-scale optimization mechanism for demand side
load management is proposed. It compose the Agent
Management of Energy, it carries out the distribution of
the energy of the housing by proposing a dynamically
threshold of total energy consumption will be applied to
each household .The home automation system integrated
in each household plays the role of controlling all the
energy consumption in the housing by using service
flexibilities, which have the possibilities to be modified
and controlled.
Chao-Rong Chen [2013]: This paper proposes a method
of minimizing tariff for customers through changing
elastic load use time intervals where customers
electricity use time is divided into inelastic and elastic
intervals by electricity use characteristics. By use of
genetic algorithm it is found that through elastic load use
time interval changes, minimum tariff objective can be
reached, and feasibility of the proposed method is
verified.
Abaravicius J. [2006]: This study aims to discuss the
possibilities and the benefits of using interval (hourly)
metering data from residential consumers. Through the
analysis of strengths and weaknesses of different load
analysis tools, this paper defines the knowledge they
could give, how applicable they are and what value they
could have both for the utility and for the residential
customer. The study is exemplified with ten cases of
households with electric space heating in Southern
Sweden.
Abaravicius J. [2007]: This paper reports about a study
conducted with the objective of developing a detailed
load demand analysis for commercial buildings. This
study was performed in collaboration with IKEA and E.
ON and contributes to an ongoing IKEA energy efficiency
programme. Two sample department stores in Sweden
were selected and analysed within this project. The
demand data analysis covers almost three years period,
2004-2006.

Abaravicius J [2005]: The objective of this study was to
experimentally test and analyse the conditions and
potential of direct load management from customer and
utility viewpoint. Techno economic and environmental
aspects as well as customer experiences were
investigated. Space heating and hot water systems in ten
electric-heated houses were controlled by the utility
using an existing remote reading system.
Pyrko J. [2003]: The objective of this study is to
investigate the extent to which a Load Demand
Component, included in electricity pricing, can influence
energy use and load demand in residential buildings.
This paper investigates the impact of the new tariff on
the utility and different types of typical residential
customers, making comparisons with the previous tariff.
Abaravicius, J. [2006]: The key objective of this study is
to discuss the possible environmental benefits of load
management and evaluate their significance, primarily
focusing on CO2 emissions reduction. The analysis is
carried out on two levels: national – the Swedish
electricity market, and local – one electric utility in
southern Sweden.
Shockman Ch. [2006]: This study examines the limits
and possibilities of environmental decision making by
local managers of multinational operations such as IKEA
managers. Some definitions were provided to general
store employees to determine their reaction to a
program called “demand response”. The power and
authority of local managers to respond quickly to
potential social problems is examined. The formal
decision-making process is explicated and projections
about future avenues of approach for environmentally
desirable projects are included. This study provides
insight for other socially desirable environmental
projects that face adoption difficulties in large, complex
organizations.
3. ECONOMIC LOAD DISPATCH
Economic load dispatch is one of the basic optimization
problems in power system analysis. The main focus of
ELD is to get out the optimal arrangement of power
generations is equivalent to total power demand at least
possible cost while maintain the power generators and
system constraints. The cost of generation, especially in
thermal power plants, is more, hence suitable
arrangement of unit outputs can give to outstanding
saving in operating cost.
The optimization of economic load dispatch problem
involves the answers of two different problems. The first
one is the unit commitment or pre dispatch problem
wherein it is required to choose the more desirable way
out of the accessible generating sources to run to
assemble the expected load and give a define margin of
operating reserve over a define period time. Another one
feature of economic dispatch is the on line economic
dispatch whereas it is need to scatter load among the
generating units indeed paralleled with the system in
such way as to reduce the total cost of supplying the
moment to moment requirements of the system.
4. RESULTS AND DISCUSSION
For optimize our laod we used two optimization
technique GA and PSO and compare the result by using
these two technique.
We discuss case study on three customer schedule 24
hour power use. In this section we draw four graph for
each customer .1) Load curve before optimization . 2)
cost curve before optimization. 3) Load curve after
optimization. 4) cost curve after optimization . When we
analysis all these graph it is clearly shown how how load
and cost is reduces after optimization and it directly
reduces our tarrif also.
Some assumption in Load curve:
Maximum load :1.6MW
Red line : Inelastic load
Green line : Elastic load
Blue line : Total load (Elastic+Inelastic laod)
Case 1: Electricty Scheduling of Customer by using
Genetic Algorithm
Figure 4. 1 Electricity Price Curve
Electricity Scheduling of Customer 1
Figure 4. 2 Load Curve Before Optimization

Figure 4. 3 Cost Curve Before Optimization
Figure 4. 4 Load Curve After Optimization
Figure 4. 5 Cost Curve After Optimization

Electricty Schedule of Customer by using Genetic
Algorithm
Custo
mer
Maxi
mum
Elasti
c
Load
Befor
e
Opti
mize
Maxi
mum
Elasti
c
Load
After
Opti
mize
Total
Maxi
mum
Load
Befor
e
Opti
mize
Total
Maxi
mum
Load
After
Opti
mize
Maxi
mum
Cost
Befor
e
Opti
mize
Maxi
mum
Cost
After
Opti
mize
1 1.68 0.24 1.70 0.68 0.31 0.100
9
2 1.62 0.38 1.75 0.70 0.31 0.103
5
3 1.70 0.39 1.78 0.70 0.325 0.105
0
Table 4. 1 Electricity Schedule Before and After
Optimization
Case 2: Electricty Scheduling of Customer by using PSO
Figure 4. 14 Electricity Price Curve

Electricty Schedule of Customer by using PSO
Custo
mer
Maxi
mum
Elasti
c
Load
Befor
e
Opti
mize
Maxi
mum
Elasti
c
Load
After
Opti
mize
Total
Maxi
mum
Load
Befor
e
Opti
mize
Total
Maxi
mum
Load
After
Opti
mize
Maxi
mum
Cost
Befor
e
Opti
mize
Maxi
mum
Cost
After
Opti
mize
1 1.58 0.38 1.75 0.72 0.312 0.152
2 1.69 0.48 1.78 0.75 0.324 0.152
3 1.66 0.40 1.76 0.72 0.318 0.111
8
Table 4. 2 Electricity Schedule Before and After
Optimization
COMPARISION BETWEEN GA AND PSO:
So, by comparison of table 1, table 2 and table 3 we can
say that both methods optimize in better way. But GA
gives better result as compare to PSO, in all cases and for
all elastic load, total load and cost. So we prefer GA
method for optimization.
Customer 1: table 1 shows the maximum elastic load
before and after optimize, total maximum load before
and after optimize and maximum cost before and after
optimize between Genetic Algorithm method and
Particle Swarm Optimization method.
Maxi
mu
m
Elast
ic
Loa
d
Befo
re
Opti
mize
Maxi
mu
m
Elast
ic
Load
After
Opti
mize
Tota
l
Maxi
mu
m
Load
Befo
re
Opti
mize
Tota
l
Maxi
mu
m
Load
After
Opti
mize
Maxi
mu
m
Cost
Befo
re
Opti
mize
Maxi
mu
m
Cost
Befo
re
Opti
mize
GA 1.68 0.24 1.70 0.68 0.31 0.10
09
PSO 1.58 0.38 1.75 0.72 0.31
2
0.15
20
Table 4. 3 Comparison Table for Customer 1

Maxi
mu
m
Elast
ic
Loa
d
Befo
re
Opti
mize
Maxi
mu
m
Elast
ic
Load
After
Opti
mize
Tota
l
Maxi
mu
m
Load
Befo
re
Opti
mize
Tota
l
Maxi
mu
m
Load
After
Opti
mize
Maxi
mu
m
Cost
Befo
re
Opti
mize
Maxi
mu
m
Cost
Befo
re
Opti
mize
G
A
1.62 0.38 1.75 0.70 0.31 0.10
35
P
S
O
1.69 0.48 1.78 0.75 0.32
4
0.15
2
Maxi
mu
m
Elast
ic
Loa
d
Befo
re
Opti
mize
Maxi
mu
m
Elast
ic
Load
After
Opti
mize
Tota
l
Maxi
mu
m
Load
Befo
re
Opti
mize
Tota
l
Maxi
mu
m
Load
After
Opti
mize
Maxi
mu
m
Cost
Befo
re
Opti
mize
Maxi
mu
m
Cost
Befo
re
Opti
mize
G
A
1.70 0.39 1.78 0.70 0.32
5
0.10
5
P
S
O
1.66 0.40 1.76 0.72 0.31
8
0.11
18
5. CONCLUSION
In this thesis we made a model for the economic dispatch
problem integrated with stochastic demand side
management. Our model proposes a solution for the
economic dispatch problem and reduces the load by
providing the chance to the user to engage and supervise
their load according to their requirement by combine the
demand side management.
REFERENCES
[1] M. Marwan and F. Kamel, “Optimum Demand
Side Response of Smart Grid with Renewable
Energy Source and Electrical vehicles”. 2006.
[2] Pyrko J., Sernhed K., Abaravicius J., Perez Mies
V., Pay for Load Demand. Electricity Pricing with
Load Demand Component, ECEEE Summer
study 2003 Proceedings, Cote d’ Azur, France.
[3] Ashan Ashfaq, Sun Yingyun and Akif Zia Khan,
“optimization of Economic Dispatch Problem
Integrated with Stochastic demand Side
Response”, IEEE International Conference on
Intelligent Energy and power System (IEPS)
2014.
[4] Biljana Trajanovska Utrechr, “Consumer’s
acceptance of Smart Meters”, 31 May 2013.
[5] Riccardo Poli, James Kennedy, Tim Blackwell,
“Particle Swarm Optimization an Overview”,
Springer Science + Business Media, LLC 2007.
[6] Hu X., & Eberhart, R.C., “Adaptive Particle Swarm
Optimization: detection and response to
dynamic systems”, In Proccdings of the IEEE
Congress on Evolutionary computation (CEC),
PP.1666-1670, Honolulu, HI. Picataway: IEEE,
2002.
[7] Sinha Nidul, Chakrabarthi R. and Chattopadhyay
P.K., “Evolutionary programming techniques for
dynamic systems” IEEE Transactions on
Evolutionary computation: Vol-7, pp.83-94,
2003.
[8] J.E. Baker, “Adaptive selection methods for
genetic algorithms. “In J.J. Grefenstette, editor,
Proceedings of the First International
Conference on Genetic Algorithms, pages 101-
111. Lawrence Erlbaum Associates, 1985.
[9] Mehrdad Dianati, Insop Song and Mark Treiber,
“An Introduction to Genetic Algorithms and
Evolution Strategies University of Waterloo,
Waterloo, Ontario, N2L 3GI, Canada,2002.
[10] Chao- Rong Chen, Ming-Jen Lan, Chi- Chem
Huang and Steven H. low, “Demand Response
Optimization for Smart Home Scheduling using
Genetic Algorithm”, IEEE International
Conference on System, Man and Cybernetics, 2013.
[11] K.F. Man, K.S. Tang and S. Kwang, “Genetic
Algorithms: Concepts and Applications”, IEEE
Transactions on Industrial Electronics, Vol 43
No. 5 October 1996.
[12] Marwan Marwa, Gerald Ledwich and Arindam
Ghosh, “Smart Grid Demand Side Response
Model for Optimization Air Conditioning”, 2008.
[13] Duy Long Ha, Frizon de Lamotte and Quoc Hung
Huynh, “Real time dynamic multilevel

optimization for Demand side load
management”, IEEE Conference 2013.
[14] Ioannis Lampropoulous, Wil L. Kling, Paulo F.
Ribeiro and Jan van den Berg. “History of
Demand Side Management and Classification of
Demand Response Control Schemes”, IEEE
Conference 2013.
[15] Yamille del Valle, Ganesh kumar
Venayagamoorthy, Salman Mohafheghi, Particle
Swarm Optimization: Basic Concepts, Variants
and Applications in Power Systems. IEEE
Transaction on Evolutionary Computation, Vol.
12. No. 2, April 2008.
[16] Jian Jiao, Xianjia Wang and Liubo Zhang,
“Applying Particle Swarm Optimization in
Multiobjective and Hybrid Optimization”.
Second International Conference on
Computational Intelligence and Natural
Computing, Vol. 12 No. 2, April 2008.
[17] Zhiyu You, Weirong Chen, Guojun He and
Xiaogiang Nan, ‘Adaptive Weight Particle Swarm
Optimization Algorithm with Construction
Factor”, International Conference of Information
Science and Management Engineering, 2010.
[18] Abaravicius J., Pyrko J., Load Management from
an Environmental Perspective, Energy &
Environment, Volume 17, No. 4, 2006.
[19] Shockman Ch., Abaravicius J., Piette M.A.,
Adoption Issues for an Automatic Demand
Response Program by a Multinational Company,
ACEEE Summer Study Proceedings, Pacific
Grove, California, 2006.
[20] Juozas Abaravicius, “Demand Side Activities for
Electrical Load Reduction” Lund University
June2007.
[21] C.L. Wadhwa Electrical Power System.
[22] Shubham Tiwari, Ankit Kumar, G.S. Chourasia,
G.S. Sirohi economic load dispatch using particle
swarm optimization Volume 2, Issue 4, April
2009.

IRJET- Demand Response Optimization using Genetic Algorithm and Particle Swarm Optimization

More Related Content

What's hot (20)

Similar to IRJET- Demand Response Optimization using Genetic Algorithm and Particle Swarm Optimization (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET- Demand Response Optimization using Genetic Algorithm and Particle Swarm Optimization