IEEE PES T&D Conference & Exposition 2022 - 112

Case Study: Southern Company
Distribution System Design
Within the industry, a common distribution system design
is a single transformer with multiple feeders. If system load
dictates, the design can include multiple banks serving multiple
busses and associated feeders. However, doing so adds
increasing complexity to both the construction and testing
of these systems as they often have complicated throw-over
schemes to mitigate customer impacts to system events.
Within Southern Company, a gas and electric utility
holding company headquartered in Atlanta, GA, there was
interest in what the implications would be if this standard
industry design was converted to a digital substation. The
development of a digital substation standard has the potential
to significantly reduce costs in terms of wiring, testing,
and maintenance. In addition, implementing a digital substation
design would enable system-level testing as previously
discussed in this paper. This level of testing is often impractical
for a large station as it is often difficult, if not impossible,
to coordinate equipment outages to facilitate the testing.
Instead, testing a system this complex requires multiple,
isolated tests to check individual components of the system.
Another advantage of the digital substation approach is
the ability to use IEDs to handle multiple protection functions.
This further reduces the footprint and wiring requirements,
and it simplifies to an even greater degree the logic
required to implement these complex restoration schemes, as
the number of IEDs needed is reduced.
Utilizing a single IED for multiple zones is not common
within Southern Company, so multiple topologies are being
designed and tested. With one vendor, a hybrid approach is
being used where there is a mix of single and multi-zone IEDs,
while a single, centralized IED design is being evaluated with
another vendor. For this article, the focus is on the design and
testing of the centralized IED design shown in Figure 9.
Even in a centralized IED design, there is a need to
have system-level coordination for complex schemes like
those seen in a throw-over (TO). To evaluate this type of
system-level test, logics shown in Figure 10 and Figure 11
were used. For an enabled system, if there was a fault in one
bank, the system would try to initiate the TO as long as there
was no abnormal condition on the receiving bus. The abnormal
conditions were monitored in Figure 11.
If the TO conditions were met, then the bus tie breaker
would close, picking up the load from the failed bank as
soon as the failed bank was successfully isolated. However,
if the load of the receiving bank exceeded its limit post-TO,
then it would initiate a load-shed procedure.
Case Study: Test Set-up and Results
System-Based Tests of the Protection IEDs
This section presents the system-based tests performed during
the proof-of-concept for the project described in the previous
section and shown in Figure 9, involving the protection
IEDs of Bank 1 and Bank 2. The test approach is like the one
explained previously and illustrated in Figure 7. The system
under test is modeled within the grid editor of the testing
simulation tool as shown in Figure 12.
The 115 kV side is modeled as a source while the 13.8 kV
feeders are modeled as loads. The model does not require
much information to be created, just basically some nameplate
data for the power transformers plus CT and VT ratios.
Sources and feeder loads can be easily estimated. All merging
units (PIUs in Figure 9) are added as simulated devices. The
PIU SCL (cid) files are imported for each simulated device
so the test set can publish the respective SVs. SCL (cid) files
from the protection IEDs are also imported to configure the
GOOSE signals to be monitored (subscribed) by the test set.
In total, 8 sampled values streams were needed for the
1002
IED
Bank 1
PIU
1001
PIU
420
PIU
416
PIU
426
figure 9. Case Study Application.
112
ieee power & energy magazine
PIU
436
PIU
446
IED
Bank 2
defined test cases: Two streams from each of the bank PIUs
(CT HV side + CT LV side and bus voltage); two streams from
tie breaker 420; and one stream from each feeder 416 and 436.
Feeders 426 and 446 were not included in the tests presented
in this paper and therefore their breakers were kept open in the
model. To simulate all eight streams, two test sets were used
that connected to the communication
network and time-synchronized from
the substation precision time protocol
(IEEE 1588) clock. The computer
running the simulation software
was connected to both test sets via a
USB connection.
An extract of all test cases defined
for this project, along with the
expected results, is shown in Table 2.
Figure 13 shows the test results
PIU
2001
of the first test case from Table 2.
Trip of breakers 1002 (HV side) and
feeder 416 were measured with the
expected time delays of 23ms and
25.3ms, respectively, as observed on
top of Figure 13, The trip delay of the
April 2022 Show Issue
2002
IEEE PES T&D Conference & Exposition 2022

Table of Contents for the Digital Edition of IEEE PES T&D Conference & Exposition 2022
