IEEE Power & Energy Magazine - May/June 2021 - 40

Microgrids, like all power systems, require protection
that de-energizes and isolates faults before they can harm
health or property.

control systems lead to IBRs having specific fault current signatures. These unique features include the following.

but have different time constants, they may interact with
one another.

Limited Fault Current Magnitude

Different Sequence Components

To protect semiconductor switches, most inverters include
both hardware and software switch current limiters that keep
the IBRs' fault current level low. The maximum steady-state
fault current produced depends on design parameters and
varies from one IBR to another. Once a fault is applied, the
inverter current exhibits a very short transient (lasting fewer
than two 60-Hz cycles) and then returns to a steady-state
fault current that is typically around 1.1-1.5 pu of the inverter's current rating. This is generally true whether an inverter
is grid forming or grid following.
The fault current transients depend on the inverter controllers as well as the thermal rating of power electronics
switches. Some inverter manufacturers are working on
increasing the magnitude and duration of the fault current
(especially for islanded grid-forming inverters), but even at
2-3 pu fault current, it can be difficult to distinguish in-rush
or motor-start events from a fault condition.

In grid-following IBRs, the control loops are designed to
suppress the negative-sequence current component, resulting in negative-sequence currents that are typically less than
10% of the positive-sequence value, which can cause protection systems to behave unpredictably. Grid-forming inverters can generally inject higher levels of negative-sequence
currents for unbalanced grid operation, but not all inverters
in the microgrid may be controlled this way, which can make
predicting inverter fault current contribution and protection
design challenging.

Control-Determined Current-Voltage Phase Angle

When a fault occurs that causes the IBR terminal voltage
to become low, the magnitude of the fault current produced
by grid-forming and grid-following inverters is essentially
the same. Both types will reach the inverter's current limit
and hold there, as explained earlier, but the current-voltage
phase angles may be significantly different.
For example, a grid-forming inverter will increase the
magnitude of its output current as part of its effort to reacquire its voltage setpoint, and, at the same time it will in
general move toward higher current-voltage phase angles
(higher reactive power). In contrast, a grid-following inverter
increases its output current magnitude as it attempts to reacquire its active power setpoint, but while it is still on-grid,
the current-voltage phase angle will be set by the IBR's reactive-power programming, which may set a limit anywhere
from zero to the IBR's reactive power limit.
When a grid-following inverter is isolated on a faulted
system, the IBR will change its frequency as it attempts
to adjust its current-voltage phase angle, and this frequency change will typically be sufficient to trip the
grid-following IBR, but it may also interfere with fixedfrequency root mean square measurements. Furthermore,
if multiple IBRs all are providing reactive power support
40

ieee power & energy magazine

Nonlinear Fault Current Contribution

Inverter fault current limiters also make the fault currents
nonlinear. This can impact the conventional phasor-domain
short circuit analysis, which relies on the linear-equivalent
Thevenin model of sources. Moreover, many inverters limit
their peak output currents by manipulating the pulsewidth
modulation to lower the peaks of the output current, resulting in a flat-topped waveform. The unwanted harmonics on
the voltage and current values can lead to challenges in calculating their fundamental values.
Importantly, grid-forming inverters used for microgrid or
off-grid applications may be designed to produce negativeand zero-sequence currents. Unlike typical grid-following
PV inverters, to support unbalanced microgrid applications,
grid-forming inverters control voltage and frequency instead
of limiting negative-sequence injections. For example, in
Figure 4, for a single-line-to-ground fault, conventional protection designs expect that zero-sequence current (I 0) equals
positive-sequence current (I 1) equals negative-sequence
current (I 2). For a grid-following inverter, there is no zero
sequence current and only limited negative-sequence current
during the fault because of the inverter controls, but the gridforming inverter responds as expected for more conventional
sequence current injections.
Similar experimental tests were performed for the gridforming inverter for double-line-to-ground faults, where the
inverter I 1 injection equaled the sum of I 0 and I 2, as expected,
and line-to-line faults, where the inverter I 1 current equaled
I 2 as expected. This fact can be especially challenging
when designing microgrid protection schemes because
the inverter-based generation may switch between control
may/june 2021

IEEE Power & Energy Magazine - May/June 2021

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - May/June 2021