IEEE Power & Energy Magazine - January/February 2020 - 51

✔

applications. Furthermore, it should embrace a timely GIS update cycle to reflect the as-built and even
the proposed distribution network configuration.
Subsequently, the DNOM can be kept up to date with
the nominal configuration of the physical distribution network.
In parallel to the commitment to keep the as-built
model up to date, the utility must ensure that business processes are in place to keep the as-operated
status as current as possible. The SCADA-monitored
status changes and tags will be incorporated into the
DNOM automatically as they occur. Utility practices
often need to become stricter when it comes to incorporating manual field changes into the DNOM in a
timely manner. Manual changes of switchgear and
fuses as well as temporary jumpers and cuts should
be captured so that the as-operated state of the distribution network model follows the physical system in
the field. This practice improves situational awareness
and safety and also means that restoration field work
will be incorporated quickly into the outage management function, thus improving the accuracy of performance indices.
A consistent observation from utilities that implement
an ADMS in conjunction with grid digital transformation initiatives (such as AMIs, SCADA-automated
feeder switches, widespread sensors, and intelligent
fuse saving devices) is that operations personnel in the
control room spend less of their time trying to infer
the sources and extent of unplanned outages. Because
more elements can be observed and the advantages
of having advanced applications work with automated
switchgear, the outages are located and often largely restored before crews are even dispatched. When crews
are dispatched, they can travel directly to the problem
location and work on repairs and further customer
restoration with manual switching. The utilities interviewed recommended that any separate outage management and DMS/SCADA organizational silos within
the operations department be eliminated so that all parties are consistently using and updating the single asoperated model shared by both functions.
Users should plan to stay engaged with the ADMS
vendor to take advantage of product functionality
enhancements related to emerging operational challenges and opportunities. This is especially important
with respect to the rapid growth of DER penetration in
distribution networks. Photovoltaics, electric vehicle
charging, storage systems, and the evolving inverters
connecting these entities to the distribution network
are all driving a continuous stream of functional updates in the advanced applications, which are needed
for ADMS viability.
A program should be provided with a strong training
environment, such as a DOTS, enabling new users to

january/february 2020

interact with the ADMS in a controlled and safe environment rather than waiting to experience challenging
events on the job.

Vendor Requirements for
Successful Implementations
Of course, vendors must provide ADMS products and services
that make it possible for a utility to obtain and maintain a viable ADMS. A collection of functional characteristics, capabilities, and best practices that contribute to a successful ADMS
with advanced applications should include the following:
✔ The underlying model for the advanced applications
must provide a good physical representation of the real
distribution power system, matching its operational
states and construction characteristics. The solutions
of distribution applications like power flow, FLISR,
and especially IVVO can be more accurate and valuable when the model can realistically represent the
physical world. The inclusion of DERs in the model
is essential.
✔ As mentioned, the advanced applications must be numerically robust so that they can digest inevitable model
and measurement data errors and still provide converged
solutions. This is most critical in the initial commissioning builds of the DNOM but remains an important characteristic in production as well, since the as-built model
evolves continuously. The applications must be forgiving.
✔ Continuously running functions like distribution power flow, state estimation, and IVVO need to operate
nominally in a hands-off manner, without operator intervention. Planned switching and unplanned outages
requiring manual restoration work will continue to occur in the field, so system operators will have only a
short time to devote to advanced applications.
✔ The applications should be able to consume more and
more near-real-time observations (inputs) beyond the
traditional manual entries for a planning power flow.
SCADA, AMIs, DER inverters, aggregator interactions,
phasor measurements, and forecasts of load and DER
production are all currently available and should be leveraged as input sources for the advanced applications. Taking advantage of the growth in sensors is an area very
worthy of future development in the distribution state
estimation space.
✔ When an ADMS with advanced applications is running well, many people across the organization will
want to have access to the results. The vendor should
provide interfaces to allow a broad audience in the enterprise secure access to ADMS results.
✔ The applications should be adaptable in functional areas, such as evolving regulatory reporting requirements,
without custom software and system down time.
✔ Vendors should provide a technology roll-out program that does not require a utility to do large-scale
system replacements, thereby protecting the utility's
ieee power & energy magazine

IEEE Power & Energy Magazine - January/February 2020

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