IEEE Power & Energy Magazine - January/February 2020 - 64

Unlike a traditional DMS, ADMS applications have access to
information and controls from multiple systems and are well
suited for improving system observability and controllability.
emerging electric power distribution systems and identifies
future research directions.
Traditionally, power distribution systems have been de-
signed for passive operation with limited observability and
controllability. Potential issues are addressed at the plan-
ning stage, where the system is usually designed and built
to cope with the anticipated worst-case scenarios. However,
distribution systems worldwide are rapidly transforming into
complex and stochastic networks that support large pen-
etrations of distributed energy resources (DERs), responsive
loads, and distributed sensing and decision-making units.
A critical aspect of these rapid transformations is the large
amount of demand and generation uncertainty introduced to
the system, leading to unprecedented operational challenges.
These dynamically varying operating conditions cannot be
easily managed in the planning stages; rather, they call for
advanced approaches that oversee the evolving needs of
power distribution systems.
Unlike a traditional DMS, ADMS applications have access
to information and controls from multiple systems and are
well suited for improving system observability and control-
lability, which are necessary to address rising operational
complexities. We categorize ADMS applications into three
stages based on the degree of advancements:
✔ stage 1: applications currently being adopted by
the industry and at the demonstration and deploy-
ment stages
✔ stage 2: near-term applications at the R&D stage and
expected to be at the deployment stage within the next
one to five years
✔ stage 3: long-term applications at the early research
concept stage and expected to be in the deployment
stage within the next five to 10 years.

Overview of ADMS Applications
An ADMS leverages data- and control-rich environments that
facilitate the management of power distribution systems using
advanced analytics. Conceptually, an ADMS is an advanced
version of a DMS, which includes all the functionality of a
DMS, while incorporating the future planning and opera-
tional requirements of evolving power distribution systems.
The advancements brought by an ADMS to the distribution
systems depend upon the capabilities of the applications sup-
ported by the ADMS. This section details ADMS applica-
tions currently supported or envisioned by industry and the
research community alike in the near or long term as well as
their dependencies on data, automation, and advanced analyt-
64

ieee power & energy magazine

ics. Depending on the maturity of the existing technology and
the additional development needed to support the proposed
applications, we categorize ADMS applications into the three
stages previously described; Table 1 presents the characteris-
tics of some representative applications within each category.

Stage 1: Applications Currently
Being Adopted by Industry
These are the ADMS applications currently being adopted
by a majority of distribution companies and made available
by most ADMS vendors. The large-scale deployment of sen-
sors, controllers, and actuators has been a key enabler in the
development and rapid adoption of stage 1 applications such
as active voltage management, direct load control methods,
and utility-scale DER control and coordination. Similarly,
applications related to fault location, isolation, and site res-
toration (FLISR) are, today, more the norm than the excep-
tion for any distribution company with a basic ADMS. A
common feature of stage 1 applications is the use of sim-
pler analytics and primarily rule-based/heuristic methods.
Nonetheless, the growing complexity of power distribution
systems calls for better/smarter analytics that are currently
under development.

Stage 2: Near-Term Applications
The development of these applications is largely driven by
the need to optimize feeder operations by coordinating the
state of heterogeneous loads and DERs in a complex, active
distribution system. With the added complexity and hetero-
geneity of proactive and controllable assets, rule-based and
local control methods may become inefficient or even inad-
equate. This is where centralized model-based optimization
and control methods can be beneficial to a range of distribu-
tion system applications. The availability of network models,
real-time data, remote-control capabilities, and advanced
analytics are the key enablers for stage 2 applications. These
advanced analytics, however, may involve nonlinear mod-
els and the use of optimization techniques, such as optimal
power flow (OPF) methods. The requirement for added com-
putational power and the need for scalable problem formula-
tions (e.g., using linearization) are the major challenges to
adopting these applications.

Stage 3: Long-Term Applications
Finally, we envisage applications expected to be ready for
distribution companies within the next five to 10 years.
These applications are largely driven by the need for
january/february 2020
IEEE Power & Energy Magazine - January/February 2020

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - January/February 2020
