IEEE Power & Energy Magazine - March/April 2022 - 34

The analysis looked at several DER scenarios to see how
DERs could impact system performance and analyzed mitigation
needs based on two alternative views:
✔ Business-as-usual: Increasing electric demand from
DERs, EVs, and general growth would be met by upgrades
to substations, distribution lines, and secondary
transformers.
✔ DER support: The capabilities of DERs are taken
into account to help manage load and reduce or defer
investments by utilizing battery storage capacity, demand
response, energy efficiency, and the managed
charging of EVs.
Key Results
The IDRP study revealed several important factors that
will impact distribution planning. A failure to account for
these findings could lead to an overinvestment in the distribution
system and unnecessary rate-payer costs. Our
study shows that a significant increase in DERs can be
accommodated without major cost increases as long as the
growth in DERs is balanced and well managed. For example,
SMUD's long-term 2030 Zero-Carbon Plan adopted in
2021 anticipates EVs in the Sacramento region to grow
from about 20,000 to nearly 300,000 by 2030 and all-electric
homes to increase from about 50,000 today to about
150,000 in 2030. These dramatic changes must be managed
carefully to maintain low rates and continued high
reliability. In particular, four results stand out, each of
which has implications for planning and operations and is
discussed in the following sections.
Unless Carefully Managed, EVs Could
Constitute 10% of Peak Demand by 2030
The number of EVs is expected to grow dramatically in
California over the next decade and, this will be the most
significant driver of load growth. Not surprisingly, unless
we manage that load carefully, EVs could wreak havoc
on the distribution system and constitute as much as 10%
of our peak load by 2030. However, if managed properly,
the strong growth of EVs (and even higher volumes) could
be accommodated within the existing infrastructure. The
Probability Distribution
of EV Charging at
the Peak Hour
P10
Peak Load
Expected EV Load
figure 1. The expected vehicle load impact versus
probability outcomes.
34
ieee power & energy magazine
P99
key is to avoid charging during peak hours and manage
the charging.
To successfully manage EV charging, we must not
only have the right equipment-communication between
the grid and vehicles, ideally combined with a DERMS
and an ADMS-but we must also understand where the
load will show up. This means learning which customers
are the most likely to buy an EV, their charging habits,
and which parts of the distribution grid they will impact.
This entails developing detailed locational forecasts of EV
charging demand.
For planning purposes, we must know not just the ex -
pected impacts on each feeder but the probability distribution
of the EV charging load. This is necessary to anticipate
the likelihood that system components will be overloaded.
Merely using the expected values based on a load profile is
too simplistic. We risk overloading the equipment if the load
goes higher, thus putting reliability at risk. Conversely, if we
use an equally simplistic worst-case scenario for planning
[for example, the tail end of a probability distribution where
there is a 99% (P99) chance that the load would be lower
than the indicated level], we risk overinvesting, which results
in high costs. Each utility must also define its risk tolerance
and approach to maintaining overall reliability. Figure 1
illustrates the expected peak load on a circuit and a probability
distribution for EV charging that highlights the wide
range of possible outcomes.
Due to the potentially significant impact of the load
growth from EVs, EV charging needs to be coordinated to
avoid surprises. This can be done through managed charging,
either directly using a DERMS or working with an aggregator
or indirectly through tariffs and incentives, which create
uncertainty, as illustrated in Figure 1. The vertical bars
labeled, respectively, P10 and P99 indicate the probability
levels of the distribution. At P99, there is a 99% chance that
the load will be at or below this line. At the P10 level, there
is only a 10% chance that the load will be at or below the line
(and, conversely, a 90% chance that the load will exceed the
P10 level).
Due to the uncertainty of when and where the EV charging
load appears on the grid, having flexibility on the distribution
system also has value. For example, having an
integrated distribution grid rather than radial lines allows
the flexibility of switching the load between circuits to more
efficiently manage the demand.
Solar PVs and Batteries Are Not Significant
Drivers of Hosting Capacity Constraints
The second key finding is that solar PVs and batteries are
not putting capacity constraints on the distribution system.
Because the peak load typically occurs between 5 and
7 p.m., solar PVs have little impact on a circuit's net peak
load. Similarly, the distributed batteries used by commercial
customers to mitigate demand charges help reduce the peak
load. This contributes to improved hosting capacity for new
march/april 2022
IEEE Power & Energy Magazine - March/April 2022

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - March/April 2022
