IEEE Electrification - December 2022 - 63

volt-var. As shown in Figure 5, volt-
var activation can even be seen from
the utility billing meter; in this
instance, it is visible because 1) the
background customer var consumption
is low enough that it does not
mask inverter var operation and 2)
voltages exit the dead band in which
there is no var operation.
NREL and Hawaiian Electric also
have been investigating capturing
volt-watt operation, which is exceedingly
more challenging, as there is no
way of actively differentiating curtailment
and normal fluctuations in
active power production, even with a
dedicated DER generation meter. For
now, NREL and Hawaiian Electric are
using heuristic algorithms, based solely on AMI voltage
and inverter capacity, to estimate the maximum curtailment
a customer could see. This is being done to identify
potential customers who may be experiencing persistent
curtailment and to identify system changes to operation
or upgrades that could reduce customer PV curtailment.
The AMI data are collected and analyzed at a 5-minute
resolution to try to capture shorter duration voltage excursions
that would have been averaged out at the 15-minute
or hourly resolution. To date, there has been little curtailment
of any statistical significance, and individual cases
have been due to preexisting conditions and tuning of
voltage control assets rather than high voltages induced
by solar production.
AMI for Network Diagnostics
The vast quantity of data from AMI provides utilities the
potential to perform network diagnostics of potential
power quality issues (Figures 6 and 7). Doing so will
100
10
20
30
40
50
60
70
80
90
0.95
0.96
0.97
0.98
0.99
1
1.01
Per-Unit Voltage
Figure 6. The duration curve for voltages experienced by 100 AMI meters, allowing the establishment of regions of normal and tail-end voltages.
IEEE Electrification Magazine / DECEMBER 2022
63
1.02
1.03
1.04
1.05
Voltage Range That 90% of
Customer Voltages Remain Within
How might VVO, with
full visibility of gridedge
voltages, react
when simultaneously
having meters
across the circuit
above and below
character-of-service
ranges?
require robust analysis to determine
which power quality issues are normal
statistical issues present in any
given network and which are systemic
and persistent problems requiring
action from distribution engineers. A
lack of visibility of grid-edge power
quality has led to little understanding
of the frequency of normal voltage
excursions that have had little
impact on customer power quality.
With this visibility now present, engineers
will have the data to make
informed decisions about which voltage
issues must be addressed. Planners
will have the data needed to
measure the magnitude of deviation,
the duration, and the frequency (i.e.,
the persistence). Power quality issues should also be
ranked and weighted by the number of customers
impacted. Also, data may reveal characteristics of issues
that can be grouped (e.g., which issues are being experienced
on an entire feeder or on the distribution secondaries
and which issues are impacting individual
customers, which may point to issues in primary regulation
versus localized network impacts) and the appropriate
level of response to proactively plan and remediate
any power quality issues.
AMI for Operations and Volt-Var Optimization
In the longer term, AMI is also being explored for use in
operations. Integrating real-time AMI measurements into
an advanced distribution management systems (DMSs) is
seen as a potential way to enhance voltage regulation strategies,
such as volt-var optimization (VVO). Challenges of
AMI integration include the volume of data and the interpretation
of the results. An expansion to use AMI in DMSs
High Region out of
Normal Operating Region
Percent Duration (%)
Low Range out of
the Normal Operating Region
IEEE Electrification - December 2022

Table of Contents for the Digital Edition of IEEE Electrification - December 2022
