IEEE Power & Energy Magazine - November/December 2019 - 97

Active
Power (MW)

✔ as a guarantee that frequency does not vary too quick-

ly, allowing for devices providing frequency support
to make accurate frequency measurements
✔ to enable system studies, specifically to make sure that
phasor-domain simulations are valid in most cases.
However, these slower time-scale dynamics do not need to be as
slow as they currently are in systems with SGs! Depending on

50.5
50.3

(Hz)

at the dc side of the converter, and its size will depend on the
speed of response of other devices on the grid.
The line between the need for traditional phasor-domain
and detailed EMT power system simulations becomes less
distinct with the increase in IBPSs. At its core, every converter
used to interface a generation source to the power system is
a voltage-source converter (VSC). However, by nature of the
present IBPS control, the power system perceives these converters as virtual current sources, and the converters themselves follow the grid voltage angle and frequency. As system strength decreases, the fast controllers of the IBPSs can
experience stability issues, which are presently not observable with the converter models in phasor-domain power system simulations. Conducting a detailed EMT simulation is
the only avenue to study the phenomena (the blue curve in
Figure 7). However, through recent research conducted at
Arizona State University and EPRI, a new generic phasordomain IBPS model was developed for low system strength
conditions to capture the oscillatory behavior observed in
detailed EMT simulations (the green curve in Figure 7).
With this model, the impacts of low system strength conditions on the behavior of IBPSs and possible mitigation measures can now be studied efficiently.
It is important to note that the GFM controls, even though
they are capable of a high-amplitude fast response, must
lower their bandwidth for the following reasons:
✔ to ensure that the system is stable, without inverters
reacting faster than the network dynamics (since the
network is used to convey the synchronization information through the voltage angle)

50.1
49.9
49.7
49.5
0

1

2

(s)

3

4

5

Frequency in Hertz
With Zero-Cycle Delay
With One-Cycle Delay
With Three-Cycle Delay
With Five-Cycle Delay
With 10-Cycle Delay
With 15-Cycle Delay

figure 6. The University of Strathclyde study results showing
the effects of FFR with various delays in the control loop.

50

EMT
State-of-the-Art Positive Sequence
New Positive Sequence

0
-50

Reactive
Power (Mvar)

4
150
100
50

4.4

4.6

4.8

5

4.4

4.6

4.8

5

4.4
Time (s)

4.6

4.8

5

EMT
State-of-the-Art Positive Sequence
New Positive Sequence

0
4

Voltage (pu)

4.2

2
1.5
1

4.2

EMT
State-of-the-Art Positive Sequence
New Positive Sequence

0.5
0
4

4.2

figure 7. A comparison of EMT, the present state-of-the-art phasor domain model, and an improved phasor domain
model when a three-phase bolted fault is applied at an IBPS point of interconnection in a low short circuit area.
november/december 2019

ieee power & energy magazine

97
IEEE Power & Energy Magazine - November/December 2019

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - November/December 2019
