IEEE Wireless Communications - April 2017 - page 13

IEEE Wireless Communications • April 2017
To achieve reliable and accurate knowledge of the
grid condition, the distribution system state estima-
tion (DSSE) is of key importance. The benefit of the
state estimation is that it can take into account all
types of available measurements, thus reducing the
investment costs into the required measurement
infrastructure. Further, DSSE provides estimation of
the grid state also on the grid nodes where mea-
surement devices are not located. As the measure-
ment locations are placed all the way down to the
prosumer level, the shared cellular networks seem
to serve as an efficient and viable solution for com-
municating between measurement devices and the
back-end system. In general, DSSE performance
depends on location density, type, accuracy, and
reporting interval of the available measurement
infrastructure in the grid, as described below.
•Phasor measurement units (PMUs) are dedi-
cated devices with common time reference pro-
vided by a very high precision clock, which allows
for time-synchronized phasor (that is, synchropha-
sor) estimations at different locations. Combin-
ing high precision and high sampling rate (up to
50/60 Hz) measurements of voltage or current
phasors on all three phases from multiple PMUs
allows for a comprehensive view of the state of
the entire grid interconnection. Figure 1 depicts
a fully embedded micro PMU (
PMU) proto-
typed within the SUNSEED project that enables
three-phase voltage/current synchrophasor mea-
surements at medium/low voltage of the distribu-
tion grid. In addition to dedicated measurement
circuitry and signal processing, it also features a
Linux-enabled application processor, LTE, Ethernet
and low-power radio communication interfaces,
secure element, and a GPS-based reference clock.
•Power measurements and control (PMC)
devices allow for three-phase power quality mea-
surements (such as real/reactive/apparent power,
frequency, voltage, current, total harmonic distor-
tion) and control of end connection points (via
on/off relays or serial line protocol). Within the
SUNSEED project, the devices were designed in
exactly the same form factor as the
PMU, reus-
ing the application and connectivity boards and
introducing a measurement and control board. A
1 s reporting period was considered for devices
deployed at major grid buses and important pro-
sumer locations to support state estimation, while
a request-response mechanism was considered to
support demand-response services described next.
•Smart meters (SMs) for standard billing mea-
surement are assumed to be deployed at each
prosumer. SMs are based on 1 min or 15 min
reporting interval. In the future, SMs may be used
for power measurements as well. However, this
requires lower reporting intervals, e.g., down to 1 s
as with PMC devices.
Regardless of the challenges that need to be
addressed, the main benefits of making the grid
highly observable can be summarized as follows:
• Disturbances on the lower voltage level can be
locally detected, cleared, and eliminated before
they affect other parts of the system.
• DSOs will be able to identify grid model defi-
cits, and allow them to construct accurate mod-
els suitable for detailed analysis and planning.
• DSOs will be able to analyze how installed and
planned generation will affect the grid, enabling
short, medium, and long term planning.
• Continuous grid observation will pave the way
for real-time grid control.
The smart distribution grid enables advanced
features in demand monitoring, analysis, and
response. Importantly, the monitoring and control
activities will not only reside in operation centers,
but can also be distributed across the entire grid
by enabling control of consumption and produc-
tion flexibility in consumer and prosumer loca-
tions. Having both sides of the distribution grid
participating in demand-response can lead to a
win-win outcome. The DSO benefits from efficient
network control, and the consumers benefit from
optimized use of energy. For example, consumers
at the demand side will be able to manage their
own consumption by changing the normal elec-
tricity consumption patterns over time. Both cen-
tralized and decentralized approaches are seen
in the literature [3, 4]. Decentralized techniques
usually have reduced computational complexity.
However, undesired communication overhead
may be expected. We note that the quality of
service (QoS) requirement for demand-response
is relatively relaxed compared to DSSE, with the
measurements inter-arrival time in the order of
minutes to hours, and the data transmission laten-
cy requirements around 1 s [5]. The communi-
cation burden can be further reduced with the
use of approximated information in the neighbor-
hood-wide consumption scheduling as proposed
in [6]. In this way, the scheduling is done by the
individual consumers while their actual consump-
tion is observed by the SMs and the PMC units.
The role of the security framework is to protect
the smart grid assets against unfriendly attacks.
An assessment of potential attack scenarios has
resulted in the identification of four high-level
security objectives:
• Insure availability of the services offered by the
smart grid (resilience to cyberattacks to insure
• Insure privacy of communications within the
smart grid (avoid spying).
• Prevent damage to equipment or infrastructures
(resilience to cyberattacks to insure equipment
or infrastructure safety).
Illustration of
PMU that has been developed for the SUNSEED project.
Application and
connectivity board
Measurement and
control board
(PMU and PMC type)
LTE modem
GPS location and
timing module
Secure element
1...,3,4,5,6,7,8,9,10,11,12 14,15,16,17,18,19,20,21,22,23,...132
Powered by FlippingBook