IEEE Electrification - December 2022 - 59

accurate and time-interval billing,
and the ability to provide feedback on
customer energy consumption.
AMI: Beyond Billing
Unlocking the true potential of utilities'
AMI investments means business
process changes that go beyond
the basic AMI infrastructure for meter
reading and software measurement,
validation, verification, and customer
billing presentment of energy consumption.
Most modern meters, alongside the integrated
energy delivered and received (or kilowatthours), can also
measure the voltage, current, and reactive power. In addition,
metering time resolution, while traditionally hourly,
can be down to minutes or seconds. Metrics need not just
be integrated averages but can be the maximum, the minimum,
or counters of key statistical metrics. The potential
of AMI has led to utilities collecting myriad metrics and
statistical quantities and operating at different time resolutions.
AMI can be much more than an asset for smart
billing alone (e.g., the time-of-use, demand charges, and
zero export), and it has the capability to enable greater
insight into network planning and operation.
Smart metering is an
enterprise-wide tool
that is enabling
visibility when and
where it has been
most needed.
Utilities frequently now have
design choices regarding which data
they wish their AMI infrastructure to
collect. These choices involve which
physical metrics to log, which statistical
metrics to capture, and at which
time resolution to collect the data (see
Figure 2). The design choices come
down to how AMI data will be used
beyond billing. These design choices
also include how many meters to
deploy on a single customer premise
in the presence of DERs, which we discuss later.
Important decisions, such as the time resolution at
which the data are collected, impact the usability of the
data for specific applications. Particularly with powerflow
metrics, time resolution can have large impacts on
results. This is mainly due to the large discrete fluctuations
that are characteristic of load consumption (Figure 3).
High-resolution data can be important for estimating
the peak load when sizing protection equipment and
network assets as well as when estimating network
power-flow data (e.g., in Figure 3, the maximum load
measured at 60-minute intervals is less than 64% of
that measured at 1-minute intervals). One primary
example is estimating losses on the distribution network.
Because losses are nonlinear (i.e., I2R), having
fewer sampling data of the integrated average power
flow can underestimate losses by as much as 40%. The
obvious tradeoff with high-frequency data collection is
the volume of data being stored. While collecting 1 second
data may seem unreasonable, it could be argued
that hourly data lose too much fidelity for many analysis
purposes.
For most utilities, AMI has a potential value add-on
beyond billing and outage management systems that
has not yet been realized. However, computational,
analytical, and implementation hurdles must be overcome
to realize this value. For utilities, AMI has spurred
the transition from a desert of grid-edge data to information
overload. For a single distribution network,
utilities potentially have thousands of customer datasets,
with tens of metrics per meter and potentially
more than a million data points per meter per year.
Utilities now must work out the means to process and
extract meaning from these data to enhance network
planning and operation. Doing so requires merging
AMI datasets with geographical information systems
(GISs), customer information systems (CISs), and DER
technical data; processing huge datasets; and synthesizing
these datasets into actionable metrics. In addition,
the ability to aggregate power-flow metrics per
secondary, per feeder, and per substation will help distribution
planners assess network health and determine
which networks need readjustments to operation
or potential upgrades.
IEEE Electrification Magazine / DECEMBER 2022
59
©SHUTTERSTOCK.COM/ANDREY_POPOV
http://www.SHUTTERSTOCK.COM/ANDREY_POPOV
IEEE Electrification - December 2022

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