IEEE Power & Energy Magazine - July/August 2018 - 42

We report results from a recently published initial analysis
conducted by NREL that simulated widespread electrification
from present day through 2050 in the United States.

changes to rate structures, end-use technologies, weather, and
other demand-side changes. as a result, the peak-to-average
demand ratio remains constant through time in this scenario,
with peak load 70% higher than average load (59% load factor).
in Figure 5, the "peak" is defined as the top 40 load hours.
this simple definition does not capture all implications, as
discussed later, but illustrates the large impact that electrification can play on load shapes. in contrast, the peak-toaverage ratio decreases substantially through time in both
the high-electrification and high-combined scenarios. By
2050, peak demand is less than 9-13% higher than average demand in these scenarios (89-91% load factor). this
result is driven in large part by our bounding assumption that
electric vehicle charging and hydrogen production loads are
highly flexible. specifically, we assume zero charging during the top 40 load hours of the year and that load associated
with hydrogen production is completely flexible within a day.
our implicit assumption is that rate structures incentivize
smart charging and the existence of sufficient infrastructure
to avoid vehicle charging during the peak hours and adequate hydrogen storage and other infrastructure exist to shift
hydrogen production to nonpeak hours. given that load associated with hydrogen production represents approximately
50% of total annual transportation electricity consumption
under this scenario (in 2050), the diurnal flexibility ascribed
to hydrogen production results in highly flexible load in the
vehicle sector. our analysis does not evaluate the costs to
achieve this level of flexibility, which is highlighted as a
future research need. Without intelligent policies and consumer behavior, the peak-to-average demand ratio is likely
to remain higher than shown in the outer years of the analysis, with important implications for build-out and operation
of the grid.
How electrification drives electricity consumption patterns can have other impacts to electric system planning and
operations beyond resource adequacy. For example, the flattening of demand profiles can have important implications for
the amount and type of supply-side resources; flatter demands
might drive the preference for technologies that provide energy
resources over capacity resources (e.g., resources designed for
efficient performance and low-cost energy but with higher
capital costs; these have traditionally been referred to as baseload resources). in another example, changing demand profiles
can impact renewable integration challenges and opportunities. the capacity and energy values of wind and solar generation are closely tied to the correlation (or lack thereof)
42

ieee power & energy magazine

between variable generation profiles and demand. the amount
of renewable curtailment can also be affected by how electrification impacts load profiles. Whether regulations, market
designs, and rates enable electrification to support efficient grid
evolution such as through greater demand-side participation is
an important research area.

Electricity Supply-Side Evolution
with Widespread Electrification
the impacts of u.s. electrification depend on how the
demand side might evolve but also on the supply-side future
of the electricity system. We develop various electricity supply-side scenarios-responding to the demand-side changes
envisioned-by employing the nrel regional energy
Deployment system (reeDs) model. the model simulates
the operation and expansion of the u.s. power system,
including power plants, transmission, and storage from present day through 2050 by choosing the cost-optimal mix of
technologies. (utility stationary storage, as opposed to storage in vehicles, is treated as a resource in the modeling.) the
least-cost solutions found by the model are constrained to
meet all regional electric power demand requirements (with
and without electrification), planning and operating reserve
requirements, technology resource constraints, and any policy requirements. in all scenarios, we use technology cost
and performance assumptions from the mid-cost projection
of the nrel 2016 annual technology Baseline and reference case fuel price assumptions from the eia 2016 annual
energy outlook.
Figure 6 shows the annual generation and installed
capacity from four scenarios modeled using reeDs. in the
reference scenario, which includes only current policies,
demand for new capacity resulting from business-as-usual
growth in load and end-of-life retirement of existing generators is met predominantly by new wind, solar, and natural
gas generation. total generation in 2050 under such a scenario is 5,300 tWh, of which about 33% comes from wind
and solar (compared to about 7% in 2016), 28% from natural
gas, and 22% from nuclear, hydropower, and other renewable
technologies. While coal generation declines in the long run,
it still provides about 17% of 2050 generation in this reference scenario. on a capacity basis, we find 16 gW of average
net annual additions of wind and solar capacity from 2017 to
2050 and about 9 gW of net annual natural gas capacity additions, figures that are roughly in line with u.s. trends since
2010. in other words, without widespread electrification
july/august 2018

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - July/August 2018