IEEE Electrification Magazine - March 2015 - 33

Proof of concept of New Management
Solutions for autonomous Power Systems:
Laboratory Setup
Envisioning the development of advanced experimental
infrastructures for feasibility demonstration of solutions for
EV charging control and integration of renewables in electric power systems, INESC Technology and Science (INESC
TEC) has been implementing a laboratorial infrastructure
that exploits distinctive control and management solutions
for key resources such as EV and renewable-based generation units. Within this framework, the present laboratory
infrastructure was designed to support the development
and testing of solutions and prototypes both for hardware

Pmax
Pref
∆fmin
∆f0

∆fmax

∆f

Pmin

Frequency Deadband
V2G

EV Battery Charging

Figure 8. The EV charger power-frequency droop characteristic.

Frequency (Hz)

50.5
50
49.5
49

Case 1
Case 2

48.5
48

100

150

200

250

300

Time (s)
(a)
1

0.95
PEV (MW)

following and frequency control. The existing hydro units
could theoretically replace the diesel units in this task as
they are also equipped with synchronous machines. However, the high head height and long conduit impose a
higher water starting time. Therefore, the hydro units are
not frequently used to perform frequency control without
diesel generation since the frequency fluctuations tend to
be larger. The mechanical wear and tear of fast governing
actions also cautions against it.
To illustrate the potential benefits resulting from the
EV contribution to primary frequency regulation, a valley
hour scenario with a total conventional load of around
730 kW was considered. Additionally, it was assumed that
25% of the island vehicle fleet (2,000 vehicles) was
replaced with EVs and all of them adhered to smart charging schemes that would concentrate the EV load on the
valley periods. Under such conditions, the additional load
for EV charging is about 965 kW. An expansion scenario
was considered, where more wind power installed capacity was included (three wind turbines with 330 kVA). The
frequency regulation was also considered to be exceptionally provided by hydro power stations. A wind gust was
simulated in the system, leading to a wind power
decrease from 500 to 300 kW. The impacts in the system,
when the EV does not provide primary frequency regulation (case 1) and when they provide primary frequency
regulation (case 2), are depicted in Figure 9.
If the system is operated without EV participation on the
primary frequency control, and under a full renewable-based
generation dispatch, the frequency drops to a minimum
value of 48.4 Hz. In opposition, when the EV droop control is
considered, the frequency does not fall below 49.8 Hz.
There are two factors influencing this discrepancy: the
slow dynamic reaction of the hydro generation units and
the fast reaction of EVs to frequency deviations. In terms
of EV participation in frequency control, the total charging
power temporarily changes from the original 965-824 kW,
resulting in a reduction of 3.6% in the energy absorbed by
the batteries of EVs during the 300 s of simulation. The
participation of EVs not only sustains the frequency drop
but also avoids the oscillations verified in the frequency
response without EV participation.

0.9

0.85

0.8

Case 1
Case 2
0

100

150

200

250

300

Time (s)
(b)

Figure 9. The contribution of EVs to frequency regulation: (a) the frequency and (b) the active power of EVs.
IEEE Electrific ation Magazine / March 2 0 1 5

Table of Contents for the Digital Edition of IEEE Electrification Magazine - March 2015