IEEE Power & Energy Magazine - March/April 2018 - 72

200

Emergency Mode Level 1
52.8 MW/min

(MW)

150
Fast-Start CT 1
24.9 MW/min

100

Fast-Start CT 2
23.7 MW/min

50
0

20 25
Time (min)

figure 5. An implementation rate comparison for emergency mode level 1.

included thousands of voltage regulators. aDMS provided a
dynamic model of the grid and all its devices, thus eliminating the need to maintain such a list. It also provided tools
for system operators to efficiently manage voltage regulators
that needed to be excluded from emergency voltage reduction due to required maintenance or other issues. with the
addition of line regulators to emergency voltage reduction,
temporary and permanent switching that changed circuit
topology became an additional factor and could be accounted
for if accurately modeled by an operator within aDMS. this
capability would likewise prove to be beneficial with the
future introduction of automated circuit reconfiguration by
self-healing networks. Furthermore, the combination of a
dynamic model and control panel with a definable system
hierarchy enabled the potential for emergency voltage reduction activation on smaller geographic areas, even down to
the individual t/D substation level, in the event of localized
issues such as transformer overloading.
It is important to note that, while the aDMS model provided these advantageous enhancements, emergency mode
continued to be based on the concept of nonoptimized voltage reduction and was therefore designed to operate independently of the quality of the model's LF and SE results. the
nonoptimized premise and overall objective of fast implementation speed for immediate load reduction were also
underlying factors in the decision for the mode not to perform any type of command validation. aDMS did become
responsible for processing and executing the necessary commands for starting (and stopping) the mode and periodically
sending the fail-safe commands required to keep the voltage reduction active in the regulator control. this was by
no means a simple task, considering the speed requirement
and the scope of the previously discussed changes realized
as part of the mode's strategic development. the simplest
yet probably most revealing example of this is a comparison
between the legacy system and emergency mode with regard
to the total number of commands required for activation.
72

ieee power & energy magazine

DEP estimates that it required approximately 600 commands to start level 1 emergency voltage reduction prior to
the start of the DSDr project. the total increased to approximately 6,800 commands at the time of aDMS deployment,
primarily due to the addition of almost 3,000 line regulators
and two additional commands for each.
the mode was also designed with configurable options
for sending commands to disable local voltage override
automation of feeder capacitor banks prior to starting voltage reduction, which DEP utilizes as part of its emergency
mode level 2 to avoid conflicting and unnecessary device
operations. this pushed the potential worst-case total number of commands required for starting emergency mode to
more than 9,000. there was an obvious driver for extensive evaluation and optimization of SCaDa hardware and
infrastructure to ensure that command throughput was not
a limiting factor, accompanied by continual testing after
going live with aDMS to verify that the implementation
speed met nErC requirements.
a comparison of data captured from an emergency mode 1
activation and the dispatch of two of DEP's fast-start Ct
units proved that the result was a success. as seen in Figure 5, emergency mode was able to reach its maximum load
reduction (shown as an increase for the comparison) well
within the required time frame and-perhaps even more
impressively-at a rate over twice as fast as the ramp rate
achieved by the two generation units (Ct 1 and Ct 2).

2014 Polar Vortex
In January 2014, several areas of north america experienced
a weather phenomenon-known as a polar vortex-that produced the coldest winter temperatures experienced for
nearly two decades. temperatures on the East Coast dropped
20-30 °F below the normal seasonal average, creating a surge
in energy demand that set new all-time peaks on 6 and 7 January. Due to the wide-reaching and extended duration of the
event, the entire U.S. power grid was strained by record high
demand and significant generation capacity issues that arose
from the impacts of the extreme cold. the temperatures in
several areas fell below the design limit of generators and
associated equipment, resulting in many frozen units.
In addition, the firm demand for natural gas-which is
typically supplied to generators through real-time delivery-
increased to a level where there was no transportation capacity
for nonfirm purchases, causing a lack of fuel for many other
generators. these unavailable units put system operators in
an extremely challenging situation and forced many to rely on
alternative options for maintaining the continuity of the grid,
including public appeals for energy conservation and demandside management programs. One southeastern utility was even
forced to shed firm load, which is typically a last resort. the
conditions for DEP were not far from that point, but its operators also had another resource at their disposal.
DEP, in anticipation of a record peak on the morning of
7 January 2014, proactively deployed several of its available
march/april 2018

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - March/April 2018