IEEE Wireless Communications - April 2017 - page 60

IEEE Wireless Communications • April 2017
58
1536-1284/17/$25.00 © 2017 IEEE
A
bstract
The development of smart grid brings great
improvement in the efficiency, reliability, and
economics to power grid. However, at the same
time, the volume and complexity of data in the
grid explode. To address this challenge, big data
technology is a strong candidate for the analysis
and processing of smart grid data. In this article,
we propose a big data computing architecture for
smart grid analytics, which involves data resourc-
es, transmission, storage, and analysis. In order to
enable big data computing in smart grid, a com-
munication architecture is then described con-
sisting of four main domains. Key technologies
to enable big-data-aware wireless communication
for smart grid are investigated. As a case study of
the proposed architecture, we introduce a big-da-
ta-enabled storage planning scheme based on
wireless big data computing. A hybrid approach
is adopted for the optimization including GA for
storage planning and a game theoretic inner opti-
mization for daily energy scheduling. Simulation
results indicate that the proposed storage plan-
ning scheme greatly reduces the cost of consum-
ers from a long-term view.
I
ntroduction
Smart grid is an innovation of power grid with a
high integration of advanced monitoring, sens-
ing, communication, and control technologies
in order to provide sustainable, economic, and
secure power services to customers. With the
rapid development of smart grid, large amounts
of smart meters (SMs) and sensors have been
deployed with huge coverage. As a result, a
large number of multi-sourced heterogeneous
smart grid data has been produced. Enormous
value can be extracted from these smart grid data
which can not only increase the quality of smart
grid from the view of utility companies, but also
provide better service for different types of cus-
tomers than traditional power grid. Therefore, the
application of big data technology can bring huge
benefits for smart grid. For instance, IBM utilized
4 petabytes of climate and environmental history
data; a wind turbine location model was designed
that can determine the best installation location
of fans. As a result, the efficiency of wind turbine
production has been improved and the service
life has been extended [1].
However, big data computing in smart grid
brings stringent requirements for wireless com-
munication technologies due to the vast volume,
variety, and velocity of smart grid data. Compared
to traditional wired communication technology,
wireless communication technologies have some
unique advantages in deployment, expansion, and
cost efficiency. Moreover, with advanced tech-
nologies, wireless communications are also char-
acterized by high data rate and reliability. Much
research has been conducted on wireless com-
munication technologies applied in smart grid.
For example, Yu
et al.
[2] studied the cognitive
radio technology applied in smart grid to achieve
optimization of spectrum resource management.
Overall, wireless technologies provide promising
support for big data computing in smart grid.
Wireless big data computing has wide appli-
cations in smart grid, which can be generally cat-
egorized into four aspects: customer profiling,
demand response, network planning, and pricing.
As a case study of the proposed smart grid big
data computing architecture, we introduce a resi-
dential energy storage planning mechanism in this
article. Some work has been done previously on
the planning problem of energy storage devices
owned by utility companies [3]. However, to the
best of our knowledge, there is still little research
on the planning of consumer-side energy storage
devices. In this article, we focus on the planning
of energy storage devices deployed at each con-
sumer’s home based on historical energy con-
sumption data. A genetic algorithm (GA)-based
approach is introduced to obtain the optimal
capability to deploy energy storage devices for
each consumer considering long-term total cost.
In addition, an energy scheduling scheme is pro-
posed based on game theory for the daily opti-
mization in energy storage planning. As a unique
advantage of game theocratic approaches, the
Nash equilibrium derived from the proposed
energy scheduling game can give all consumers a
satisfactory reduction in billing.
This article focuses on the issue of wireless big
data computing in smart grid. We first investigate
the consistency between the characteristics of
big data and smart grid data. To this end, a big
data computing architecture is proposed. Four
levels are introduced in the architecture: data
resource, data transmission, data storage, and
data analysis. The data transmission level plays an
important role in bridging the other three levels
as well as connecting the network within each
level. To address the data transmission in smart
grid big data computing, a hierarchical big-da-
ta-aware wireless communication architecture is
then described. Possible wireless technologies
are enumerated for different domains among
the architecture. To enable big-data-aware wire-
K
un
W
ang
, Y
unqi
W
ang
, X
iaoxuan
H
u
, Y
anfei
S
un
, D
er
-J
iunn
D
eng
,
A
lexey
V
inel
,
and
Y
an
Z
hang
W
ireless
B
ig
D
ata
C
omputing
in
S
mart
G
rid
S
mart
G
rids
Kun Wang, Yunqi Wang,
Xiaoxuan Hu, and Yanfei Sun
are with Nanjing University
of Posts and
Telecommunications.
Der-Jiunn Deng (corre-
sponding author) is with the
National Changhua University
of Education.
Alexey Vinel is with
Halmstad University.
Yan Zhang is with the
University of Oslo and also
with Simula Research
Laboratory.
Digital Object Identifier:
10.1109/MWC.2017.1600256WC
1...,50,51,52,53,54,55,56,57,58,59 61,62,63,64,65,66,67,68,69,70,...132
Powered by FlippingBook