Machine Translated

Translated by
by Google
Google

Evaluation of the Requirements for Battery Energy

Storage System to Postpone the Investment of
Distribution System
Chanikkanan Wannasut1 and Komsan Hongesombut2

Department of Electrical Engineering, Faculty of Engineering, Kasetsart University

50 Phaholyothin Road, Chatuchak, Bangkok 10900, Thailand. Goal. 0-2797-0999 Ext. 1577
E-mail: 1 2
[email protected], [email protected]

Abstract—This paper proposes the application of the battery energy storage VSPPs to postpone the investment for the new distribution line construction
system (BESS) to be installed in the distribution system for the purpose of
and to improve the efficiency of electricity usage by allowing the BESS to
postponing the investment of the distribution reinforcement in the short term
power distribution planning. A case study of the distribution system including store excessive the power during the period of the generated power greater
the photovoltaic system and the BESS obtained from the Provincial Electricity than the power demand and to release the power to the power system during
Authority (PEA) is used to evaluate the requirements of the BESS to be installed in
the period of the generated power less than the power demand. By this
the distribution system by considering four types of loads which are household,
small enterprise, industry and other types. It reveals from the study that the installation of the
strategy, it is possible for PEA to accommodate in time with the increased
BESS can effectively extend the time of the new distribution line construction. number of connected

VSPPs.

II. THE BATTERY ENERGY STORAGE SYSTEM

Keywords—Battery Energy Storage System, Distributed
Generator, Very Small Power Producer, Smart Grid The BESS can be classified into many types such as the lead-acid,
NaS, super capacitor, flywheel, etc. These types have long storage period

I. INTRODUCTION and can store the large amount of energy but slow response and short life
time. Therefore, they are suitable for the provisional power supply in case of
At present, the global energy crisis and related issues become critically
the interruption of electricity. The differences between various types of
important Many current researches are focusing on energy management
batteries can be shown in Table I [3]. In this study, the lead-acid battery is
issues and smart grid technologies. Due to the continuously increasing of
chosen due to its general availability and inexpensive price. The installation
electricity demand and the promotion of renewable energy sources connected
of the battery into the power system with DG can extend the investment
to the power systems, many electric utilities are urged to pay more attention
period of the construction of the new distribution power system such as
to analyze, to design and to improve the existing power system suitable for
extending the investment period of changing to the larger sizes of the
the future change [1]. In Thailand, the Provincial Electricity Authority (PEA)
distribution feeders by which the battery will store the excessive power during
is one of the authorized bureaus realizing the importance of the smart grid.
the light load and release the power to the power system during the peak
Recently, PEA constantly endeavors to adopt the smart grid by creating the
load [4].
PEA smart grid roadmap having the implementation project for 15 years [2].
In the first period (2012-2017), there are many pilot projects such as the
smart grid development, the smart meter installation, the microgrid
TABLE I
development and the connecting plan of very small power producers (VSPPs).
THE PROPERTIES OF EACH TYPE OF BATTERIES.
Being focused on VSPPs, PEA needs to plan for extendable distribution
Battery Rated Power Price Life Time Efficiency
systems in order to accommodate the increased power generated from
VSPPs in the future satisfying both funding and the installation setup period Types (kW) (USD/kWh) (Times) (%)

for long term and short term. Lead Acid 1– 50,000 13 1,500 85

I don't know Several kW - 65 5,000 75-86

MW

Super 1 - 100 130-520 50,000 95

The application of the BESS is an alternative to assist the distribution Capacitor

utilities with the increased number of connected
Flywheel 5- 1,500 130-390 30,000 90

978-1-4799-0545-4/13/$31.00©2013IEEE

Authorized licensed use limited to: CONACYT NATIONAL COUNCIL OF SCIENCE AND TECHNOLOGY. Downloaded on January 12, 2025 at 21:40:09 UTC from IEEE Xplore. Restrictions apply.
 Machine Translated
Translated by
by Google
Google

III. DEMAND FORECAST

and
is the tolerance error of the simulation model.
t

This paper considers the energy consumption of the electricity It can be noticed that the equation does not depend on the

customers by using the demand forecast in the PEA small area load temperature as in the two previous cases.

forecast in order to comply with the short term distribution planning in four D. Other type

years [5]. The loads of electricity customers can be divided into four types The simulation model of this type of user can be expressed
ace:
as follows: 1) household, 2) small enterprise, 3) industry and 4) other type.
yy++=ÿ ÿ ÿ11 ttt
and

(3)

The simulation model of the demand forecast can be

Where
calculated from (1) to (3).
th
and
t is the power demand in the t month,
A. Household and
is the simulation constant,

The simulation model of this type of electricity customers and

is the variable coefficient in the simulation model,

can be expressed as: and

is the tolerance error of the simulation model.
t

++= + + + ÿt
and
Tty 13412312max,1 and t and you
and

and you
and

and yt and

(1)
IV. SAMPLE SYSTEM

Where The distribution system shown in Fig. 1 is used for the evaluation. This
th
ty is the power demand in the t month, power system consists of the following
th
components:
t Tmax, is the average maximum temperature in t
month,
PV1 PV2
and is the constant that can be calculated from:
22 kV
( m 1 +ÿÿ121121 ,
ÿÿÿÿ ÿÿÿÿ ÿ ÿÿ 1
) substation
0

4.5 km 4.5 km 4.5 km 4.5 km

and
1
+ÿÿ is 2121
and
,

and es 1, Load 1 Load 2 Load 3 Load 4

2

es
and
3
12 , Fig. 1. A study distribution system.
and
4
is ÿ 121 ,

A. Distribution system
and
t is the tolerance error of the simulation model.
The grid power can be calculated from the following

B. Small enterprise equation:

The simulation model of this type of electricity customers

ÿ+++= PPPPPPP
can be expressed in a similar way as in Section 3A. LLLLL ( )4321 PV (4)

C. Industry
The simulation model of this type of electricity customers can be
Where
expressed as:

PG is the grid power,

and ++++= ÿ yyy ÿ13312211
t ÿÿÿÿÿ t and

tttttt (2) PPPP

4321 ,,,
MMMM are the loads of electricity customers of

1st, 2nd, 3rd, and 4th type, respectively,

Where
th PPV is the power generated from the photovoltaic power
and
t is the power demand in i month,
station.
and is the constant which can be calculated from:
The maximum length of the distribution system is 18 km with the wire
m( 1ÿÿÿÿ
+ÿÿÿÿÿ) 121121
ÿ ,
of SAC 185 mm2 according to the rated standard value of the PEA's
es
1 1,
planning criteria [6]. The construction cost of the second circuit of the 22-
es
2 12 ,
kV distribution line is 43,333 USD/km [7].
ÿÿÿ3
is ÿ 121 ,

Authorized licensed use limited to: CONACYT NATIONAL COUNCIL OF SCIENCE AND TECHNOLOGY. Downloaded on January 12, 2025 at 21:40:09 UTC from IEEE Xplore. Restrictions apply.
 Machine Translated
Translated by
by Google
Google

the power of 17,210 kW 16,510 kW and 16,460 kW consecutively that

B. Photovoltaic power test
is over the line rating but the time duration of each period is short about
The forecasting of PV power output is based on the assumption
1 hour. In 2015, 10:00 am-16:00 pm, the distribution line must be able
that there is the construction of the first photovoltaic (PV) power plant
to accommodate the power that is above the line rating for about 6 hours.
of 1,000 kW and the later construction of double power capability each
The maximum power demand of 1,800 kW occurs in the 10:00 am-11:00
year until reaching the capacity limit of 8,000 kW. Therefore the new A.M.

construction of the PV power plant has the capacity from 1,000 kW to However, if the PEA decides to construct the new distribution line
8,000 kW. Fig. 2 shows the PV power profile based on the forecasting
to solve the problem of excessive power that is the new construction of
result.
the 2nd circuit on the same pole, the cost will be 43,333 USD/circuit

km. If the line length is 18 km, the total cost will be 780,000 USD and it

requires one year for the construction. On the other hand, if the BESS
8,000

7,000 is adopted instead, in 2012-2015, the size of 3,500 kWh having the cost
6,000 of 45,500 USD can be used to solve the same problem in the short
5,000
term. Therefore, the adoption of the BESS is cheaper
kW4,000

3,000

2,000
than the new construction of the new distribution line. Table II

1,000
summarizes the evaluation results of the power and energy required for
0 h
BESS.
10:00

11:00

12:00

13:00

14:00

15:00

16:00

17:00

18:00

19:00
00:00

01:00

02:00

03:00

04:00

05:00

06:00

07:00

08:00

09:00

20:00

21:00

22:00

23:00

24:00

2012 2013 2014 2015 600

500
Fig. 2. PV power profile.
400

C. Forecasting load profiles of electricity customers kW300

The load analysis of the 5th feeder of the Pattaya substation is 200

carried out and is adjusted according to the load growth in the central 100

area of Chonburi province [8]. By using (1)-(3), the forecasting of the h

0
electricity demand of each type within 4 years (2012-2015) can be
10:00

11:00

12:00

13:00

14:00

15:00

16:00

17:00

18:00

19:00
00:00

01:00

02:00

03:00

04:00

05:00

06:00

07:00

08:00

09:00

20:00

21:00

22:00

23:00

24:00
plotted as shown in Fig. 3-6. 2012 2013 2014 2015

D.Battery energy storage system

Fig. 3 The forecasting load profile for household customers.
The BESS installation can be used to postpone the
investment for the construction of the new distribution lines so as 300

to solve the problem of over ratting of the distribution lines in short term 250

planning. In this paper, the lead-acid battery is

200
considered as the BESS that has the lifetime of 1,500 times or
150
kW
4 years and costs about 260-650 USD/kWh.
100
V. RESULTS OF USING THE BESS TO POSTPONE THE
INVESTMENT OF DISTRIBUTION SYSTEM
50

h
From the PEA's planning criteria, it is found that the 22- 0
10:00

11:00

12:00

13:00

14:00

15:00

16:00

17:00

18:00

19:00

kV distribution line of SAC of 185 mm2 can accommodate the

00:00

01:00

02:00

03:00

04:00

05:00

06:00

07:00

08:00

09:00

20:00

21:00

22:00

23:00

24:00

power up to 16,000 kW. From Fig. 7, it is obvious that in 2014, 10:00 2012 2013 2014 2015

am-11:00 am, 1:00 pm-2:00 pm, and 3:00 pm-4:00 pm, the distribution
Fig. 4. The forecasting load profile for small enterprise customers.
line must be able to accommodate

Authorized licensed use limited to: CONACYT NATIONAL COUNCIL OF SCIENCE AND TECHNOLOGY. Downloaded on January 12, 2025 at 21:40:09 UTC from IEEE Xplore. Restrictions apply.
 Machine Translated
Translated by
by Google
Google

TABLE II
2,000
THE EVALUATION RESULTS OF POWER AND ENERGY REQUIRED FOR BESS.
1,800

1,600

1,400 No. Duration (hours) Maximum power over Energy (kWh)

2014 2015
1,200 the line rating (kW) 1,212
kW1,000 513
-
800 460 606
-
600 1 1 359
-
400 2 230
-
200 3 1.40 1,800 1,110
0 h -
45 1 1,008 2,426
1.30 4 Total
10:00

11:00

12:00

13:00

14:00

15:00

16:00

17:00

18:00

19:00
00:00

01:00

02:00

03:00

04:00

05:00

06:00

07:00

08:00

09:00

20:00

21:00

22:00

23:00

24:00
1,196 3,536

2012 2013 2014 2015

VI. CONCLUSIONS

Fig. 5. The forecasting load profile for industrial customers.

With the construction of the new distribution line, it is found that it is more

expensive solution since the excessive power may occur only in 1-6 hours in a

300 day. The installation of the BESS can be used to postpone the new distribution line

construction so as to solve the short term planning problem and save the cost of
250
investment. However, this paper does not consider the optimal location of the
200
installation of the BESS.
kW150

100
Finally, in order to bring the analysis results more complete, the future work should
50
be done by considering the optimal location and the power flow analysis for all

0 possible scenarios should be verified by using the specific analysis software

h
10:00

11:00

12:00

13:00

14:00

15:00

16:00

17:00

18:00

19:00
00:00

01:00

02:00

03:00

04:00

05:00

06:00

07:00

08:00

09:00

20:00

21:00

22:00

23:00

24:00

package.

2012 2013 2014 2015

REFERENCES

Fig. 6. The forecasting of load profile for other types of customers. [1] Z. Xue-song, C. Li-qiang and M. You-jie, "Research on smart grid technology", Proc. of the
2010 International Conference on Computer Application and System Modeling (ICCASM),
vol.3, pp. 599-603, Oct. 22-24, 2010.

20,000
[2] Website: http://www.pea.co.th, "PEA smart grid roadmap", 2012.
18,000
16,000 [3] H. Liang, M. Fan, J. Li, "Research on optimal configuration for combined
14,000
storage in microgrid", Proc. of CIGRE 2011 Bologna Symposium, 13
12,000
10,000 Sep. 2011.
kW 8,000
[4] A. Oudalov, D.Chartouni, C. Ohler, G.Linhofer, "Value analysis of
6,000
4,000 battery energy storage applications in power systems", Proc. of Power
2,000
Systems Conference and Exposition (PSCE'06), pp. 2206-2211,
0 h
10:00

11:00

12:00

13:00

14:00

15:00

16:00

17:00

18:00

19:00

Oct. 29 – Nov. 1, 2006.

00:00

01:00

02:00

03:00

04:00

05:00

06:00

07:00

08:00

09:00

20:00

21:00

22:00

23:00

24:00

2012 2013 2014 2015 [5] Technical Report, "Study on small area load forecast", Provincial

Electricity Authority, 2010.

Fig. 7. The power flow in the distribution line.
[6] Power system planning criteria, Provincial Electricity Authority, 2010.

[7] Unit construction cost, Provincial Electricity Authority, 2010.

[8] Website: http://www.pea.co.th, "Load profile", 2012.

Authorized licensed use limited to: CONACYT NATIONAL COUNCIL OF SCIENCE AND TECHNOLOGY. Downloaded on January 12, 2025 at 21:40:09 UTC from IEEE Xplore. Restrictions apply.