IEEE Power & Energy Magazine - March/April 2015 - 68

quantity and higher temporal resolution than is usually
available on typical distribution circuits, to validate the
distribution circuit models used in the model-based analysis described above. Voltage-related data from the installed
DMUs allowed the validation of the circuit's load profile and
of the switched capacitor controller's "on" and "off" voltage set points. For modeling purposes, the distribution of
the load on the circuit is often simply allocated based on
the phase currents measured at the substation by the utility's
SCADA system. In this case, because a considerable amount
of the circuit's load is served by the installed PV systems
instead of the substation, a more advanced load allocation
method was employed that allocated load proportionally
to distribution transformers in zones between line current
sensor locations; it also corrected for the amount of current
injected by the interconnected PV systems, as appropriate.
Second, the collected data were used to directly observe
the impacts on the distribution circuit caused by the variability of the interconnected PV systems. Table 1 shows the results
of an analysis of the number and severity of voltage variations
observed at a location near the 3-MW and 1.5-MW interconnected PV systems on the Fontana, California, study circuit for
the month of July 2013. Voltage sags and swells were defined
and sorted into event types based on the recommended trip
ranges for abnormal voltages in IEEE 1547a. The higher the
event type number, the faster the recommended trip time setting
and the more severe the impacts would be on general power
quality on the distribution circuit. The observed voltage sags
and swells indicate the frequency and severity of the service

table 1. Voltage sags, swells, and second-over-second
voltage deviations observed at a 500-kVA
distribution transformer located near the 3-MW
and 1.5-MW PV systems interconnected on the
Fontana, California, study circuit.
Type

Sag

Swell

Voltage Deviation

192

1.5

0.5
0

MVar

1.5

12
Time of Day

0.4
0.2
0

12
Time of Day

12
Time of Day
(b)

0.2

0.5

0.4

0.5
0

MVar

California, study circuit, at points that were advantageous
for directly measuring the impact of the interconnected
high-penetration PV systems. The LineTrackers had to be
installed on overhead portions of the circuit, which somewhat limited their placement.
The impact of the aggregate of 4.5 MW of PVs installed
on the circuit is demonstrated in Figure 4(b), which shows a
considerable reduction of substation-sourced current due to
the generation of the PV system during a clear-weather day.
The other LineTracker systems show how the load current is
distributed on the system. These data were used to improve
the load distribution model and thus the specific circuit-level
impacts seen at every point on the circuit for the modeling
efforts described in the previous section.
The data collected by the SCADA link with SCE, the
DMUs, and the LineTrackers had two primary uses within
the project. The first was to provide data, in much greater

PFref
PFcalc
0

12
Time of Day
(a)

0.5
0

PFref
PFcalc
0

figure 5. The real and reactive power produced by a 1.5-MW PV system operating at a power factor of 0.95 (capacitive):
(a) operation on a clear day and (b) operation on a day with variable weather.
68

ieee power & energy magazine

march/april 2015

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - March/April 2015