ETA 2021 Strategic Plan - Flipbook - Page 77
breakthroughs in computer science and AI that
enable scientists to characterize this stability and
to identify changes to that stability due to errors
in system behavior or interrogation.
Detailed Approach
The Benefits of Integrated Energy Systems
A rise in grid-interactive buildings, a near
complete transition of the on-road vehicle
fleet to the electric grid, and increased vehicleto-vehicle and vehicle-to-infrastructure
connectivity will all lead to a rapid increase in
interconnectedness. New methods are needed
to reveal the potential energy, carbon, and cost
savings that coordinated optimization offers.
In the first year, we will develop a scaled down
simulation “testbed” that is suitably expansive
to estimate the carbon-minimizing benefits of
integration but not so unduly complex that it
prevents us from exploring the vast range of
potential economic, reliability, and operation
interactions. The testbed will enable us to
identify the computational breakthroughs
needed to envisage a real-time control system
for an integrated system, and presents a
platform for comparing energy controls schemes
that balances risk-aversion and risk-taking
decisions for integrated energy infrastructure.
The testbed will consist of a simulation
environment that includes detailed building
physics models, distributed energy resources,
and the bulk power system. It will include (1) an
IEEE34 prototypical test feeder to represent the
electrical distribution network; (2) the Council on
Large Electric Systems (CIGRE) or similar network
to represent the transmission and generation of
electric power; (3) a building physics simulator
with controls for heating, ventilation and airconditioning (HVAC), lighting, and dynamic
façades; (4) behind-the-meter generation from
PV; (5) behind-the-meter stationary battery
storage; (6) single and fleet electric vehicles; (7)
detailed driving patterns with spatial movement;
(8) behind-the-meter smart inverters to regulate
voltage; and (9) local control loops for battery
and buildings. Various model fidelities (details
and dynamics) will be available to explore
the impact on global system objectives. The
simulation environment will be interfaced
through Python directly or through Berkeley
Lab’s Functional Mock-up Interface. Simulations
will be developed on local workstations and
final large-scale simulation environments run on
Berkeley Lab’s Lawrencium high-performance
computing cluster.
Even for this condensed system the control,
and optimization, of the coupled systems will
present fundamental computational hurdles
due to the nonlinear interactions across these
systems. These interactions give rise to the
need for new control strategies that cannot
be discovered from the study of individual
components in isolation. We must think of
interdependent system parameters as giving
rise to new, emergent controls that jointly
modify the behavior of the integrated energy
system, manifesting in reduced-order models
in certain circumstances. Examples of such
emergent controls are rich in synthetic biology,
where tightly coupled genetic and metabolic
components are jointly modulated through
E TA S t r a t e g i c P l a n 2 0 2 1 - 2 0 3 0
|
75
