Development of an Optimal Vehicle-to-Grid Aggregator for Frequen...
需积分: 10 38 浏览量
2016-01-26
10:26:00
上传
评论
收藏 666KB PDF 举报
IEEE TRANSACTIONS ON SMART GRID, VOL. 1, NO. 1, JUNE 2010 65
Development of an Optimal Vehicle-to-Grid
Aggregator for Frequency Regulation
Sekyung Han, Student Member, IEEE, Soohee Han, Member, IEEE, and Kaoru Sezaki, Member, IEEE
Abstract—For vehicle-to-grid (V2G) frequency regulation ser-
vices, we propose an aggregator that makes efficient use of the dis-
tributed power of electric vehicles to produce the desired grid-scale
power. The cost arising from the battery charging and the revenue
obtained by providing the regulation are investigated and repre-
sented mathematically. Some design considerations of the aggre-
gator are also discussed together with practical constraints such as
the energy restriction of the batteries. The cost function with con-
straints enables us to construct an optimization problem. Based on
the developed optimization problem, we apply the dynamic pro-
gramming algorithm to compute the optimal charging control for
each vehicle. Finally, simulations are provided to illustrate the op-
timality of the proposed charging control strategy with variations
of parameters.
Index Terms—Aggregator, battery, dynamic programming, elec-
tric vehicle, plug-in hybrid electric vehicle (PHEV), regulation, ve-
hicle-to-grid (V2G).
I. INTRODUCTION
W
ITH the growing acceptance of global climate change
as a critical environmental problem, reelectrification of
automobile transportation, referred to as vehicle-to-grid (V2G),
is getting the spotlight. So far, V2G researchers have been
mainly focused on how to connect the vehicles batteries to the
power grid [11],[14],[16]. Much attention has also been paid
to prove the validity of the V2G [15], identify its feasible ser-
vices [16], and pioneer its new markets [17], [19]. Especially,
extremely fast charging rate of the battery makes the frequency
regulation acknowledged as one of the most promising and
practical services with V2G [11], [13], [19].
Currently, most of the frequency regulation is provided by
generators that bid into the market. Such frequency regulation
is carried out on an MW basis between a grid operator and
plants that operate single or a few generators [18]. Since a typ-
ical single vehicle battery could provide only 10–20 kW of the
power capacity [20], an intermediate system, called an aggre-
gator, is necessary to deal with small-scale power of vehicles
while providing the regulation service on large-scale power.
This aggregator would play an important role in charging the
Manuscript received December 15, 2009; revised February 28, 2010. Date
of publication April 12, 2010; date of current version May 21, 2010. Paper no.
TSG-00026-2009.
S. Han and K. Sezaki are with the Department of Information Science
and Technology, The University of Tokyo, Japan (e-mail: arnold@mcl.iis.u-
tokyo.ac.jp, sezaki@iis.u-tokyo.ac.jp)
S. Han is with the Department of Electrical Engineering, Konkuk University,
Seoul, Korea (e-mail: shhan@konkuk.ac.kr)
Color versions of one or more of the figures in this paper are available online
at http://ieeexplore.ieee.org.
Digital Object Identifier 10.1109/TSG.2010.2045163
batteries of more than hundreds or thousands of vehicles. The
vehicles pertaining to the aggregator are charged or discharged
alternatively to meet the requested power from the grid operator
and fulfill charging their batteries. Since the conditions of each
vehicle such as aimed or current state-of-charge, and expected
duration of being plugged in differ from each other, it would be
meaningful to design an efficient aggregator that provides the
regulation service in an optimal way as well as charge each ve-
hicle within a specific time.
Recently, for optimal operation of the V2G, some efforts have
been made to extend an existing method, called unit commit-
ment (UC), that schedules available generating units to operate
a power system efficiently [10]. However, since the UC involves
entire units in a grid, it requires much time-consuming compu-
tation and complicated numerical algorithms. The extended UC
with V2G makes the problem even more complicated, which
naturally leads to stochastic methods such as the particle swarm
optimization (PSO). In addition, vehicles are considered just
as generating units in the extended UC and are assumed to be
charged from renewable sources, which is unrealistic to be ap-
plied to entire vehicles. The optimality is pursued only from the
perspective of efficient grid operation rather than that of each
vehicle. Thus, it is intractable to attract the vehicle owners to
join the V2G voluntarily. Moreover, when it comes to the reg-
ulation, the decision strategy should be entirely revised as the
pricing mechanism of regulation is based on the available power
capacity, not the generation cost.
In this paper, we approach the V2G regulation from a strategic
perspective for the first time. The goal of this paper is to de-
sign an optimal aggregator with respect to the frequency regula-
tion. Instead of considering the entire power system, we merely
focus on the clustered vehicles under supervision of the aggre-
gator. Unlike other conventional generating units, vehicle bat-
teries have totally different features. Vehicle batteries neither
have a startup cost nor a shutdown cost while typical genera-
tors do. In addition, generation cost which is the most impor-
tant factor in the UC is trivial in frequency regulation, since the
long-term mean of regulation request is almost zero [13]. Nev-
ertheless, the energy constraint is still a matter of concern for
the frequency regulation. Furthermore, since a vehicle has both
aspects of energy consumer and regulation provider, the market
prices for those are considered simultaneously.
We begin the article with an overview of the regulation
scheme followed by several design considerations for the V2G
regulation. Section III constructs a mathematical model with
appropriate assumptions and incorporate several factors for the
V2G regulation into performance criteria. Optimal solutions are
then obtained through the dynamic programming. The solutions
1949-3053/$26.00 © 2010 IEEE
评论0
最新资源