International Journal of Pure and Applied Mathematics

Volume 119 No. 12 2018, 1851-1864

ISSN: 1314-3395 (on-line version)
url: http://www.ijpam.eu
Special Issue
ijpam.eu

Local Search Optimization Algorithm Based

Monitoring and Controlling of Virtual Power Plant
for Distribution Network
T.Kesavan1, Dr. K. Lakshmi2 ,Dr.S.SheebaRani Gnanamalar3,R.Kavin4
1Asst.Professor,

Department of Electrical and Electronics Engineering

Sri Krishna College of Engineering and Technology
Coimbatore, Tamilnadu, India
E-mail : t.kesavan87@gmail.com
2Prof. and Head

Department of Electrical and Electronics Engineering

Sri Krishna College of Engineering and Technology
Coimbatore, Tamilnadu, India
E-mail : klakshmi01@gamil.com
3Professor

Department of Electrical and Electronics Engineering

Sri Krishna College of Engineering and Technology
Coimbatore, Tamilnadu, India
E-mail : sheebaranis@skcet.ac.in
1Asst.Professor,

Department of Electrical and Electronics Engineering

Sri Krishna College of Engineering and Technology
Coimbatore, Tamilnadu, India
E-mail : kavinr@skcet.co.in

Abstract
Virtual power Plant (VPP) is one of the emerging concept for
integrating of various Renewable Energy Source(RES) like Photo
Voltaic (PV), Wind Turbines (WT) or combined heat and power
generators with grid as a single power plant. This paper proposes
Local Search Optimization Based monitoring and controlling of
Virtual power Plant in distribution network. The proposed algorithm is
used to manage the electrical energy in distribution network in order
to minimize the purchased energy from grid .This objective is achieved
by the optimal selection of renewable based distributed
generators,optimal planning and controlling of the loads and optimal

1851
International Journal of Pure and Applied Mathematics Special Issue

operation of energy storage elements. In this proposes , two main

Renewable Energy Source of Wind power and solar power are
integrated with grid to manage the energy in VPP .In this Local
Search Optimization Algorithm is used to forecasting the generate
power from wind mill and solar cell based on wind circulation and
earth temperature respectively and also calculating power demand
depend on the load condition. The MATLAB software is used to model
the VPP and the performance analysis of generating power of sources
and power demand of load.

Keywords: Virtual Power Plant; Renewable Energy Sources; Wind

Power Generation System; Distributed Energy Resources

1. Introduction

Electrical energy is essential for all home appliances, commercial and

industrial loads. Electrical power is generating from various conventional Energy
sources and renewable energy sources. Generated power [1] from various energy
sources are not fully transferred and distributed to load side. Conventional
Electrical Management System (CEMS) is used to combine some of the renewable
energy sources with grid for distribute the power to load side without waste of
electrical energy for continuous power supply[2]. But conventional EMS facing the
challenge of continuously screening, scheduling and balancing the load in
distribution side. In early conventional energy management system is used to
integrate (RES) renewable energy sources with energy storage elements and
connected to load. Because all RES are not generating power at all time .but
power is required at all time in load side. In this time, grid power supply is
connected to load side to reach the power demand[3]. Controlling of generate
power from RES is one of the major issue in CEMS. Normally conventional
management is needed in continuously monitoring and controlling is required
distribution side depends on load condition. Generating power from RES may be
connected to energy storage element when load demand is reached level[5]. In this
paper, Virtual power Plant (VPP) is proposed to overcome the drawback of
CEMS.VPP is mainly used for continuously monitoring, controlling and scheduling
of DER in load side[6].
The main objective of this research to reduce the usage of grid power is in
load side and decrease the amount of purchasing the power from grid supply. This
operation is achieved by correctly select the RES to load based on the demand. In
this paper, Local Search algorithm is proposed in VPP for forecasting and
distributing the generate power to load side without losses. Generating power from
RES is connected to Energy Storage Element, when the power is more than load
demand.
In this paper, Section II consists of General Bock diagram and operation of
Virtual power plant. Section III describes scheduling and monitoring concept of
LSO. Section IV explains Simulation and Results of LSO based VPP for 24 hours.
Section V has the conclusion of the Paper.

1852
International Journal of Pure and Applied Mathematics Special Issue

2. Virtual Power Plant

The Virtual Power Plant (VPP) is a new idea for Energy Management System
in Distribution Side.VPP is used to integrate the various Energy sources of wind,
solar, grid power and Energy storage Element.VPP is acting as single Power plant
to the load[7].The general block diagram of VPP as shown in Fig.1.1 The proposed
VPP can be classified into source side VPP and Load side VPP based on the
operation.

Fig. 1.1 The general block diagram of VPP

Source side VPP consists of some of Renewable Energy Sources like Solar PV
panel, Wind Power Plant, Battery Storage and grid power supply. Load side VPP
consists of various loads VPP and Load side VPP based on the operation[8] .In this
paper domestic loads and industrial loads are considered in load side VPP. The
source side VPP is used for monitoring the all RES, Battery and grid power supply
and Generating power of all RES can be forecasted in separately based on different
environmental condition [9].
The wind Plant is not generating constant power at all time. Generating
Power from wind is varied depend on the wind circulation on the place. In wind
plant, source side VPP is used to forecast the generating power for next 24 hours
based on the previously observed values of wind circulation in area, height of the
tower, number and dimensions of blades. Generating power from is continuously
monitoring and distributed load side based on the load requirement. The wind
power may be connected to battery when generating power from wind is more than
load requirement. The source side VPP also used to forecast the generating power
from solar panel based on the measured values of inclination angle of solar panel,
climatic condition and number of panels. The solar power is monitored by VPP and
distributes to load side. Solar power may be connected to energy storage element
when generating power is more than load demand. Battery is considered for
charging and discharging based on the operation Renewable sources and Load
requirement. The battery is selected based on the renewable sources and load
requirement. The battery is charging when renewable sources generating more
power compare to power requirement from load. The battery is discharging when
load requirement is more than generating power from renewable source [10].

1853
International Journal of Pure and Applied Mathematics Special Issue

The grid power supply may be connected to load side when generating power from
renewable sources and discharging power from battery is not sufficient to meet load
requirement [11]. The minimum amount power only may be received from grid
supply based on load requirements. The grid supply may be received power from
renewable sources by VPP when RES generating more power compare to load
requirement power[12].
The load side VPP consists of various types of load and it calculates the
power requirement of load. In this paper, Domestic Loads and Industrial loads are
considered in load side VPP. The load side VPP is used to predict the power
requirement of domestic loads and industrial loads for coming 24 hours .The power
requirement of loads can be predicted by load side VPP based on the previously
observed details of load. Various renewable sources, battery and grid power supply
may be connected to load by load side VPP based on the power requirement of
load[13].
The Load side VPP is used to continuously monitoring the power
consumption and demand of load. If the power demand is high, the load side VPP is
used connects more number of RES to load based on the power requirements. if the
power demand is less, the load side VPP is used to connect minimum number of
RES to load and other renewable sources may be connected to battery and grid
power supply [14].

3. Local Search Optimization Algorithm

Local Search Optimization (LSO) Algorithm is proposed in VPP for

forecasting the generating power from various RES and predict the power
requirement based on the load variation. The main objective of LSO is optimally
select the RES to place the Load. The working of LSO algorithm is drawn as
flowchart as shown in Fig 2.1. In flowchart, Pl is power required for load side, Pw is
the generating power from wind mill, Ps the generating power from solar panel,Pg1
is the generating power in first stage, Pg1 is the generating power in second stage
and Pg1 is the generating power in third stage.
In this paper, domestic loads and Industrial loads are considered in VPP and
power requirement of both loads can be predicted based on the previous details.

= (2.1)

The total power demand of industrial load and domestic load as expressed in
equation 2.1.Where is the Total amount of power required in distribution side,
is the power required for industrial load and is the power required for
domestic load. In LSO algorithm, Power requirement of load can be solved by
Renewable sources and which one is available by very nearest to load. In this
procedure, wind mill is considered as nearest Power plant to both loads. If
generating Power from wind mill is meet the power demand of load, there is no need
to search another RES and Grid Supply.

1854
International Journal of Pure and Applied Mathematics Special Issue

= (2.2)
If the equation 2.2 satisfied, wind power is meet requirement of both loads
demand. Other renewable source of solar power is directly connected to energy
storage element. If and equation 2.3 shows power saved in energy
storage element and it is possible only when the generating power from wind is
more than the load demand. The total power demand of industrial load and

start

Gene_power Pg1 = power_wind Pw

If
No Pg1 = Pl Yes

Yes If Pg1 > No

Pl
Load demand met

Pg 2 =Pw + Pg1
Battery charging Load demand met

No Yes
If
Pg1 = Pl

Yes
If Load demand met
Pg2 > Pl
No

Battery charging Load demand met Pg 3 =Pgrid+ Pg 2 pPw

No Yes
If
Pg1 = Pl

No
Yes
If
Pg2 > Pl Load demand met

1855
International Journal of Pure and Applied Mathematics Special Issue

Battery charging Load demand met

stop

domestic load as expressed in equation 2.1.Where is the Total amount of

Fig 3.1. Flow
power required in distribution side, chart of LSO algorithm
is the power required for industrial load and
is the power required for
domestic load. In LSO algorithm, Power requirement of load can be solved by
Renewable sources and which one is available by very nearest to load. In this
procedure, wind mill is considered as nearest Power plant to both loads. If
generating Power from wind mill is meet the power demand of load, there is no
need to search another RES and Grid Supply.

= (2.2)

If the equation 2.2 satisfied, wind power is meet requirement of both loads
demand. Other renewable source of solar power is directly connected to energy
storage element. If and equation 2.3 shows power saved in energy
storage element and it is possible only when the generating power from wind is
more than the load demand.

= (2.3)

If the equation 2.2 is not satisfied, the power demand of load is higher than
the power generating from wind. In this time, LSO search another local
renewable source which is nearest to load side. The solar panel is considered
another local RES for both loads.

(2.4)

From equation 2.4, generating power from solar Panel and wind mill are
connected to load side for power requirement .if the equation 2.4 satisfied,
combined renewable sources are meet the demand of load power and grid power
supply is not required to load side. Equation 2.5 shows the generating power is
connected to energy storage element when generating power is more
than requirement of load power.

1856
International Journal of Pure and Applied Mathematics Special Issue

= (2.5)

The battery supply may be connected to load, If generating power of any

renewable source are suddenly varied or reduced due to some environmental
condition. If the Equation 2.4 is satisfied and combined power of wind mill, solar
panel and storage element are fulfilled required power for load.

(2.6)

The grid power supply may be considered to the load when the equation 2.
is not satisfied and combined power of wind mill and solar panel are not fulfilled
the power demand of load.

(2.4)

Where is the generating power from wind mill, is the

generation power from solar panel, is the energy stored in energy storage
element and is the power getting from grid supply. In equation 2.4, combined
power from wind mill, solar panel and grid power supply may be connected to
load. In LSO algorithm, power demand of load can be solved by renewable sources.
But grid power supply may be connected to load side when power requirement of
load is more than generating power from renewable sources. In this time, based on
the load demand minimum power only may be connected from grid supply.

4. Results and Discussion

In this paper load flow study has completed for domestic loads and
industrial load for 24 hours by VPP . The distribution network of 33 nodes as
shown in Fig 4.1.General electric board source is connected to all nodes. There are
three loads are considered in this network. Load 1 is located in node 10 and it is
considered as domestic load.

Fig 4.1. The distribution network of 33 nodes

1857
International Journal of Pure and Applied Mathematics Special Issue

Load 2 and Load 3 are located in node 22 and 30 respectively and both are
considered as Industrial load. Power requirement of Load 1 is predetermined and
1.25 MW power is averagely needed in load 1 for 24 hours. Load 1 is required
maximum power of 2.6 MW at time of 17 and minimum power of 1.1 MW from the
time of 0 to1 and 4 to 5 as shown in Fig 4.2. Power demand of Load 2 is calculated
and 1.4 MW power is averagely needed for 24 hours. Load 2 is required maximum
power of 2.5 MW at time of 17 and required minimum power 1.1 MW at time of 5.
Power demand of load 3 is predetermined as shown in Fig 4.2.and 1.7 MW Power
is averagely needed for 24 hours. The maximum power of load 3 is 2.7 MW and
minimum power is 1.4 MW. These three loads are predetermined for 24 hours by
load side VPP.

load vs Hour

3
Load 1
Variable loads demand (MW)

2.5 Load 2
Load 3
2

1.5

0.5

0
0 5 10 15 20 25
Time (Hour)
Fig 4.2. Power demand curve of various loads

The generating power of Wind power plant and Solar PV panel are
forecasted for 24 hours by source side VPP. The wind power plant is generating
average power of 1 MW for 24 hours. In Fig 4.3. shows power output of wind power
plant from 0 to 24 hours. In morning time, the wind power plant generating the
power of above 1 MW from 0 a.m. to 5 a.m. and below 0.5 MW from 6 a.m. to 8
a.m. and the maximum power of 2 MW has generating from 10 a.m. to 12 p.m.
The wind power plant has generated below 0.2 MW at evening and above 0.5 MW
at night time.

1858
International Journal of Pure and Applied Mathematics Special Issue

2 Wind Power

Wind Power(MW)
1.5

0.5

0
0 5 10 15 20 25
Time (Hour)

The characteristic curve

Fig 4.3. of solar panelcurve
Characteristics in a summer
of windday as shown
turbine unit in Fig 4.4.
The solar PV panel has generated average of 0.5 MW power for 24 hours. In
summer day, the solar panel has generating the power from the time of 5 to 19.30
and it produced maximum power of 1 MW from the time of 10 to 12.

1.5

Solar Power
Solar Power(MW)

0.5

0
0 155 20 10 25
Time (Hour)
Fig 4.4. Characteristic curve of solar panel in a summer day.

1859
International Journal of Pure and Applied Mathematics Special Issue

load vs Hour
3
Load1

Variable loads demand (MW)

2.5 Power

1.5

0.5

0
15 20 25 0 5 10
Time (Hour)
Fig 4.5. Required and generated power of Load 1
The power demand of three loads has compensated by load side VPP as
shown in figures. In Fig 4.5 shows required power of Load 1 has satisfied load
side virtual power plant. In Load 1, LSO is used to check every minutes of load
demand and compare with source side VPP. The power demand of Load 1 is
1.2MW at 12 a.m. and wind plant has generate the power of 0.75 MW at same
time. Load 1 is not Load 1 is not fulfilled by wind power plant and 0.45 MW

3
Variable loads demand (MW)

Load 2
2.5
Power
2

1.5

0.5

0
0 5 10 15 20 25
Time (Hour)
Fig 4.6. Required and generated power of Load 2

1860
International Journal of Pure and Applied Mathematics Special Issue

power is required more from wind power plant and Solar panel is also not may be
connected grid power supply for 0.45 MW and same procedure has followed to
other load requirements. The wind plant and Solar panel has generating the
power of 2 MW and 1 MW respectively at 10 a.m and the power requirement of
load is 1 MW for same time. Energy storage element gets power of 1 MW from
load side VPP. Load 1 is required power demand of 2.2 MW at 12 p.m and wind
plant generates 0.5 MW and solar panel produced 0.7 MW for same time. In this
instant, insufficient power of 1 MW has received from Energy storage element and
the same procedure has followed for 24 Hrs.

In Fig 4.6 shows required power of Load 2 has satisfied load side virtual
power plant. In Load 2, LSO algorithm is used to check every minutes of load
demand and compare with source side VPP. The power demand of Load 2 is 1.25
MW at 12 a.m. and wind plant has generated the power of 0.75 MW at same
time. Load 2 is not fulfilled by wind power plant and 0.5 MW power is required
more from wind power plant and Solar panel is also not may be connected grid
power supply for 0.5 MW and same procedure has followed to other load
requirements. The wind plant and Solar panel has generating the power of 2 MW
and 1 MW respectively at 10 a.m and the power requirement of load is 1.7 MW for
same time. Energy storage element gets power of

3
Variable loads demand (MW)

Load 3
2.5 Power
2

1.5

0.5

0
0 5 10 15 20 25
Time (Hour)
Fig 4.7. Required and generated power of Load 3

1.3 MW from load side VPP. Load 2 is required power demand of 2.2 MW at 12
p.m and wind plant generates 0.5 MW and solar panel produced 0.7 MW for same
time. In this instant, insufficient power of 1 MW has received from Energy storage
element and the same procedure has followed for 24 Hrs.

1861
International Journal of Pure and Applied Mathematics Special Issue

In Fig 4.7 shows required power of Load 3 has satisfied by load side virtual
power plant. In Load 3, LSO algorithm is used to check every minutes of load
demand and compare with source side VPP. The power demand of Load 3 is 1.4
MW at 12 a.m. and wind plant has generated the power of 0.75 MW at same
time. Load 3 is not fulfilled by wind power plant and 0.65 MW power is required
more from wind power plant and Solar panel is also not may be connected grid
power supply for 0.65 MW and same procedure has followed to other load
requirements. The wind plant and Solar panel has generating the power of 2 MW
and 1 MW respectively at 10 a.m and the power requirement of load is 2.6 MW for
same time. Energy storage element gets power of 0.4 MW from load side VPP.
Load 3 is required power demand of 2.4 MW at 12 p.m and wind plant generates
0.5 MW and solar panel produced 0.7 MW for same time. In this instant,
insufficient power of 1 MW has received from Energy storage element and grid
power supply and the same procedure has followed for 24 Hrs. The requirement
power of all three loads can fulfilled without losses and very fast response. In LSO
algorithm is programmed in load side VPP to retaining generating the power of
above 0.25 MW compare to required power of load side.

5. Conclusion

The local search optimization algorithm based monitoring and controlling of

Virtual power Plant in distribution network is presented in this paper. The
electrical energy of distribution side has managed and purchasing energy from
grid supply has minimized. In this paper, wind power plant and solar PV panel
has integrated with grid to manage the energy in VPP. Renewable sources are
optimally selected and placed in distribution side based on the required power of
load. The selection procedure of renewable in LSO algorithm has discussed with
flow chart. The performance curve of generation power from wind and solar panel
are monitored and power demand of various loads are satisfied using load side
VPP.

References

1. Loganathan, N., Dr .K. Lakshmi and Arun Venkatesh, J, “ A Robust

Energy Management System for Smart Grid”, International
Journal of Advances in Intelligent Systems and Computing,
Springer Series, Vol. 467, pp.423-435,2017.
2. Mahmoud M. Othmana, Y.G. Hegazyb, Almoataz Y Abdelaziza,
“Electrical energy management in unbalanced distribution
networksusing virtual power plant concept”, Electric Power
Systems Research 0378-7796/© 2017 Elsevier.
3. Liang, Zheming, et al. "Risk-Constrained Optimal Energy
Management for Virtual Power Plants Considering Correlated
Demand Response." IEEE Transactions on Smart Grid (2017).

1862
International Journal of Pure and Applied Mathematics Special Issue

4. LB Nikonowicz, J Milewski, “Virtual Power Plants- general review: structure,

application and optimization”, Journal of Power Technologies 92 (3),
135,2012.
5. Loganathan,N.and Dr.K.Lakshmi , “Demand Side Energy
Management for Linear Programming Method” TELKOMNIKA
Indonesian Journal of Electrical Engineering, Vol. 14, pp.72-
79,2015.
6. Loganathan,N.and Dr.K.Lakshmi , “ANN Based Linear
Programming Approach for Managemen Demand Side Energy”
International Journal of Innovative Research in Advanced
Engineering, Vol.2,pp.46-51, 2015.
7. Dr. Lakshmi, K. and Vasantharathna,S, “Hybrid Artificial Immune
System Approach for Profit Based Unit Commitment Problem”,
Journal of Electrical Engineering and Technology- Korean IEEE,
Vol.8, pp.959- 968, 2013.
8. Dr. Lakshmi, K. Karthickkumar, K and Vasantharathna, S,
“Renewable Energy Sources Scheduling with Thermal Unit under
Uncertainty in Deregulated Power System”, International Journal
of Advanced Research in Electrical and Electronics Engineering,
Vol.1, pp.49-57, 2013.
9. Dr.Lakshmi, K. and Vasantharathna S, “Improved Genetic
Algorithm for Constrained Generation Scheduling in Restructured
Electricity Markets” International Journal of Engineering Research
and Studies, Vol.1, pp. 71-82, 2010.
10. P. Lombardi, M. Powalko, K. Rudion, “Optimal operation of a
virtual power plant, in: Power and Energy Society General
Meeting”, 2009.
11. Felix A Farret, M. Godoy Simoes, “Integration of Alternative
Sources of Energy”, John Wiley & Sons, Inc., Hoboken, New Jersey ,
2006, ISBN-13: 978-0-471-71232-9
12. ŁukaszNikonowicz_ ,JarosławMilewski, “Virtual Power Plants –
general review: structure, application and optimization”, Vol.92(3),
pp.135- 149,2012.
13. Nerea Ruiz, IñigoCobelo, and José Oyarzabal, “A Direct Load
Control Model for Virtual Power Plant Management”, IEEE
Transactions on Power Systems. Vol.24, No.2, pp.959-966,2009
14. S.Sheeba Rani.2017.Unique PWM cast for Induction motor drives
with embracing non-deterministic characteristics and evaluation
using FPGA ” in Int. Journal of Pure and Applied Mathematics-
Special Issue on Electronics Communication and Networking.Vol
117 No. 10, 2017, 25-29,

1863
1864