IEEE Power & Energy Magazine - March/April 2022 - 37

EVs, solar, and battery storage together form the net
load on a feeder and the system. On the bulk system, the
concept of the duck curve is well known, where solar
and wind can result in significant decreases of the net
load during the day and trigger a dramatic ramping of
the load as solar generation declines simultaneously
with increases in electric demand. The same concept
also holds for distribution feeders and circuits, but,
since each circuit or feeder is smaller than the entire
system, the individual impacts of DERs could potent
ially be significant relative
to the capacity of the feeder
or circuit because DER adoption
from circuit to circuit can
vary greatly. The variability of
the load on feeders could also
increase. Planning for this variability
and uncertainty requires
careful investigation of the net
load and its variability circuit
by circuit rather than averaging
the system variability down to
individual feeders uniformly.
Modeling Approach-
Single-Hour Deterministic
Versus Multihour
Probabilistic
or Stochastic Models
The growth of DERs on circuits
and feeders will change the net
load shape, and, just like for the
bigger bulk system, ramping and
managing the peak load become
more important. Also, with the
growth of electrification, such as
EVs, the timing of future system
and local peaks may shift to a different
time of day or even year.
For example, with significant
electrification and, at the same
time, significant growth of solar
and storage, many California
utilities may go from summer to
winter peaking.
To appropriately plan for this
and capture both the costs and
values of DERs, we must model
more hours than just the annual
peak hour. Depending on the
system, this may mean modeling
all 8,760 h of the year or a
subset thereof. In addition, the
introduction of more DERs and
customer-sited load management
march/april 2022
systems brings new sources of uncertainty that need to be
accounted for. In the underlying load itself, uncertainty
exists regarding whether new DERs, such as batteries, the
demand response, and their control systems, will perform as
needed. There is uncertainty at the level of the distributed
generation available to offload the feeder transformer for a
given load level. Therefore, in addition to expanded timesequential
modeling, uncertainty must be factored appropriately,
ideally by applying stochastic and probabilistic
modeling techniques.
Control Systems Risk of
Individual DERs and Locations
Risk of
Underlying
Load on
Feeder or
System
Joint Risk of Net Load Taking
Covariance into Account
Net Load Risk of
Individual DERs
and Locations
Planning Load = Expected Peak Load + Sum of Expected DER Impacts +
Covariance Among Load and DERs
(a)
(b)
figure 3.The (a) load and DER risks and (b) impact of covariance for the variability
of the net load.
MW
Peak Load
Feeder Capacity
t1
t3
t2
Net Load
Deferral
t3 = Feeder at Capacity Unless Upgraded (Not Accounting for DER Impacts)
t2 = Launch Feeder Upgraded Work
t1 = Distribution Plan Identifies System Upgrade for t3
Time
figure 4.The distribution planner's dilemma.
ieee power & energy magazine
37
IEEE Power & Energy Magazine - March/April 2022

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