See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/337830891

Protection in Inverter Based Microgrids

Preprint · December 2019

DOI: 10.13140/RG.2.2.26423.68005

CITATIONS READS

0 709

1 author:

Satyaprajna Sarthak Sahoo

Indian Institute of Technology Bombay
3 PUBLICATIONS   1 CITATION   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Microgrids View project

All content following this page was uploaded by Satyaprajna Sarthak Sahoo on 09 December 2019.

The user has requested enhancement of the downloaded file.

Protection in Inverter Based Microgrids

EN 406: Seminar
by

Satyaprajna Sarthak Sahoo

Fourth Year Undergraduate
Roll No. 16D170026

under the guidance of

Prof. Suryanarayana Doolla

Department of Energy Science and Engineering

Indian Institute of Technology, Bombay
Mumbai 400 076
December 9, 2019
 Abstract
Traditional hierarchical architecture of the electrical grid and grid connected systems
are protected by mechanisms which are very carefully tuned and synchronised based on
governing behaviour of the grid itself. The aggressive penetration of renewable energy
generation, while appearing to increase the reliability of a grid network by decentralising
energy generation, has inadvertently completely altered the governing behaviour of the
grid completely, based on which the grid protection system was established. It is evident
that most traditional protection systems will fail to operate effectively in a electrical
system dominated by microgrids. This report aims to document this change in behaviour,
and proposed changes in the techniques of system protection.

i
Contents

1 Introduction 1
1.1 Overview of System Protection . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.1 Apparatus Protection . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.2 System Protection . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2 Introduction to Microgrid Protection . . . . . . . . . . . . . . . . . . . . 4
1.2.1 Types of Microgrids . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2.2 Modes of Operation . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2 Challenges in Microgrid Protection 6

2.1 Bidirectional Power Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.2 Sensitivity & Response Time . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.3 Fault Current Levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.4 Blinding of Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.5 Sympathetic Tripping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.6 Prohibition of Automatic Reclosing . . . . . . . . . . . . . . . . . . . . . 8

3 Behaviour of a Microgrid under Fault Conditions 9

3.1 Grid Connected Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.2 Islanded/Isolated Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.3 Transition phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

4 Role of Grounding in Microgrid Behaviour 12

4.1 Types of Grounding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
4.1.1 Ungrounded . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
4.1.2 Solidly Grounded . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
4.1.3 Resistance Grounding . . . . . . . . . . . . . . . . . . . . . . . . 13
4.1.4 Reactance Grounding . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.1.5 Resonant Grounding . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.2 Grounding in Microgrids . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.2.1 Types of Generators in a Microgrid . . . . . . . . . . . . . . . . . 14
4.2.2 Transformer Topologies . . . . . . . . . . . . . . . . . . . . . . . . 15
4.2.3 Microgrid Grounding Configurations . . . . . . . . . . . . . . . . 16

ii
5 Conclusion 20
5.1 Proposed Paradigms of Protection . . . . . . . . . . . . . . . . . . . . . . 20

iii
List of Figures

1.1 Section of a Radial Distribution System [16]. . . . . . . . . . . . . . . . . 2

1.2 Estimation of fault location [16]. . . . . . . . . . . . . . . . . . . . . . . . 3

3.1 Blindness of protection/loss of sensitivity in an MG interfaced grid . . . . 9

3.2 Coordination issues caused due to decentralised microgrid architecture . . 10

4.1 Current vectors of an ungrounded system while in (a)Unfaulted condition

, and (b) during an LG fault:[1] . . . . . . . . . . . . . . . . . . . . . . . 13
4.2 Common transformer winding in microgrids:[12] . . . . . . . . . . . . . . 16
4.3 Zero Sequence Equivalent circuits of transformer winding topologies . . . 18
4.4 Grounding configurations in ac systems . . . . . . . . . . . . . . . . . . . 19

iv
Chapter 1

Introduction

”The AC electrical grid is prevalent for power generation and transmission in any part of
the world, with operating voltages anywhere from 400kV to 400V. This complex system
has a wide range of apparatus, which may be protected from nature or just left in the
open. In either case such equipment, due to natural or artificial reasons tend to malfunc-
tion, which adversely affects performance, and the effects of the same may affect other
devices. Additionally, acts of nature play a far bigger role in altering the grid functioning,
endangering the fragile state of the grid, which may shut down entirely in the case of a
wrong sequence of events.

All this calls for a robust and meticulous protection system to be implemented, ensur-
ing the protection of system elements. Extensive work has been carried out in modelling
all types of system abnormalities and fault situations, and the behaviour of the grid in
such situations.

However, with the penetration of renewable sources of energy, the current system of
protection employed is at risk. Many European countries have set target for integration
of renewables that exceed 50%. India has set a target to have 33% of all power generated
by renewables. Such targets pose a significant risk to the electrical grid from a protection
perspective. This report tries to compile the behaviour of all such problems caused by
the change in behaviour of the grid, caused by the integration of renewables and solutions
proposed for handling fault conditions.

1.1 Overview of System Protection

Protection paradigms can be classified into apparatus protection and system protection.

1
1.1.1 Apparatus Protection
Apparatus protection deals with detection of a fault in the apparatus and consequent
protection. Apparatus protection can be further classified into the following:

1. Transmission line protection and feeder protection

2. Transformer protection
3. Motor protection
4. Busbar Protection
Apparatus protection is primarily designed with the aim of protection of a specific ap-
paratus, or zone of a system. Hence the parameters that apparatus protection elements
use to make tripping decisions are majorly local parameters like current or impedance
which do not change on a system level or are affected by changes in another section of a
system, hence avoiding sympathetic tripping. Existing methods of apparatus protection
are given below.

Overcurrent Protection
Traditional Overcurrent protection techniques rely on the radial architecture of distribu-
tion system. A radial architecture is characterized by a large generation source at the
centre, with power flow radially outwards, into feeder networks. Power flow is, during un-
faulted conditions, majorly unidirectional, and operational voltage decreases as we move
radially outwards.

Figure 1.1: Section of a Radial Distribution System [16].

This technique is based on the principle that all major faults tend to draw large
inrush currents during either the transient state or the steady state, or both. Hence
by using proper measurement devices (e.g., current transformers), we can measure the
value of current in any line, and subsequently trip the line in case of an overcurrent situa-
tion. It is used mainly in a radial distribution system, and it is quite simple to implement.

Overcurrent protection devices are unidirectional in nature. Hence in the figure Fig-
ure 1.1, since power flow is from the left, relay R1 can respond to both faults F1 and F2 ,

2
however R2 only responds to F2 .

Directional Overcurrent Protection

However, many situations demand more information other than current magnitude for
reliable checking of fault situations. In such cases, phase angle data (which is always rel-
ative to a reference phasor), is used. The implementation requires the use of two relays
for each line, to account for the direction of power flow.

Such systems have to be extensively used in microgrids, since the basic behaviour of
such systems are indicative of bidirectional power flow. It uses both phase angle and
magnitude information of the current flow for tripping decisions. Currently they are used
in subtransmission networks.

Distance Overcurrent Protection

Distance Protection devices work by trying to estimate the apparent system impedance
by measuring voltage and current at the sending end. In a no load condition, the line
current I = 0, & the apparent impedance of the line as seen by the relay is infinite. As
the line gets loaded, the apparent impedance decreases to some finite value (ZL + Zline ),
where Zline is the line impedance and ZL is the impedance of the load. In the presence of
a fault at a per unit distance ‘m’, the impedance as seen by the relay decreases to mZline ,
as shown in Figure 1.2.

Figure 1.2: Estimation of fault location [16].

Differential Protection
Differential Protection is one of the most reliable methods of system protection currently
in practice. [5] proposes a detection scheme for microgrid protection based on differential
protection. The principle used in the system is that in a line, or the apparatus in which
the system is implemented, during fault conditions, the sum total of the currents entering
will be different than the sum total of the currents leaving it. This method uses com-
munication between the two ends of the line and measures different properties of current

3
like phase and magnitude to influence it’s decision. They tend to be expensive to set up
and complex to implement, but have high security and reliability.

1.1.2 System Protection

System Protection devices as are evident from the name are designed to protect the system
from any instability or fault conditions by cutting off generation sources by sensing system
parameters such as voltage, or frequency. Relays are generally divided as:

1. Under frequency & over frequency relays

2. Rate of change of frequency relays
3. Under/Over voltage relays
4. Reverse power flow relays
5. Vector shift relays

1.2 Introduction to Microgrid Protection

A microgrid, is described as :
”A microgrid consists of a group of interconnected loads and DERs with clearly defined
electrical boundaries which can be operated as a single standalone controllable entity and
can be operated either grid-interconnected or grid-isolated.” [15]

1.2.1 Types of Microgrids

A common classification that is done is on the basis of power transfer. Microgrids can be
classified into:

1. AC Microgrids
2. DC Microgrids
3. Hybrid Microgrids

Since the traditional power systems are based on AC power, most microgrids are AC
microgrids. DC microgrids generally have higher reliability as they are decoupled from
the grid and not affected by it’s disturbances.

1.2.2 Modes of Operation

Microgrids may operate in grid-isolated mode or grid-interconnected mode. There is also
a transition from grid isolated operation to grid-interconnected and a transition from
grid-interconnected to grid-isolated modes.

4
 From the protection perspective, focus has to be on the following aspects of a microgrid
operation[15]:

1. Separation of the microgrid from the local electric power system due to electric
power system (EPS) de-energization, faults or other abnormal operating conditions

2. Microgrid re-synchronization to the EPS

3. Isolation of faults or other abnormal operating conditions within the microgrid,

either while operating grid isolated or grid-interconnected.

Protection within a grid-isolated microgrid includes consideration of short circuits, ab-

normal operating conditions, and power balance. When the microgrid is operating grid-
interconnected, protection issues may be divided into protection of the microgrid and the
re-synchronization protection to the EPS.

5
Chapter 2

Challenges in Microgrid Protection

Due to the methodology of microgrids being very different from the operation of electrical
grid networks, it’s operation poses many unique challenges not previously experienced or
addressed.

2.1 Bidirectional Power Flow

Traditional AC distribution systems were arborescent in nature: there were no closed
loops, and the power flow was unidirectional both in normal and abnormal states. Dur-
ing fault situations, the network could be divided into two regions for ease of analysis:
the upstream area, which is defined as the area between the source and the point of the
fault, and the downstream area, which was the region of the network between the fault
and the load.

Microgrids do not follow this hierarchical topology. They are characterized by de-
centralized sources and loads interconnected in a mesh-like architecture. This makes
estimation of short circuit current extremely difficult, as the fault current shall depend
upon components both upstream and downstream. Further, decentralised sources mean
that downstream behaviour is never constant. Micro sources may be added at any point
in time, and it’s behaviour depends upon it’s configuration.

2.2 Sensitivity & Response Time

A microgrid, especially in the isolated mode, generally has a very fast response time,
(section 3.3) and low system inertia. This is due to the lack of high reserves in DG
sources, especially of reactive power. This requires protection devices to have be able
to detect quickly and accurately, as small disturbances in system parameters can cause
catastrophic consequences.

6
2.3 Fault Current Levels
Fault current levels in a MG interfaced grid change dynamically. This is because of
the varying types of DER’s (details in subsection 4.2.1). Fault contribution is gen-
erally divided into sub-transient, transient and steady state levels. Induction motors
and generators typically contribute decaying fault currents up to four cycles. However,
inverter-interfaced systems have very unconventional fault response that pose incremen-
tal challenges to the sensitivity to conventional protection techniques.

A DER interfacing inverter can be single phase or three phase[15]. Additionally, these
inverter controls include a current limiting control that restricts the maximum current to
up to 2 pu of the rated current. The time frame is also very less, i.e., within two cycles
of the fault occurring. This leads to very high frequency transients in the fault current.
This essentially makes the inverter behaviour non linear in the faulty power system. Even
during ride through conditions, the phase angle of of the fault current fed by the inverters
may be different than the phase angle of the fault current fed by other sources (e.g., the
substation source, which is largely reactive)[15]. Since inverters are ungrounded, they
generally end up supplying positive sequence current in case of a fault.

2.4 Blinding of Protection

”The operating region of conventional over current relay is the pickup current which de-
pends on the system impedance. Pick up current of a line is calculated in such a way
that it is higher than the rated current but lower than the minimum short circuit current.
Conventional AC systems have their pickup current values rise as we go upstream from
the load to the source, enabling synchronizing the relays. This convenience is lost in a
MG where pickup current region is difficult to define, and it’s values change randomly as
we switch from line to line, which makes coordination of protection devices difficult.When
a DER is connected, the feeder equivalent impedance increases and as a result, the fault
current detected by OC relay is decreased. This issue decreases the operation zone of
relay and it cannot cover the end of its protected line.”

2.5 Sympathetic Tripping

Sympathetic tripping, or false tripping occurs when the fault current in a feeder is also
fed by the fault current generated by a DG in a neighboring feeder attached to the same
substation. In such a case, the protection devices of the neighboring feeder might discon-
nect the circuit. Another cause may also be the voltage rise in the unfaulted lines in an
ungrounded system. section 4.2.3

7
2.6 Prohibition of Automatic Reclosing
An AC network behaviour in the absence of microgrids is radial, and hence during fault
conditions it is sectioned into upstream and downstream regions of the fault. The recloser
trips and separates the upstream and downstream regions. In a microgrid, the fault is
still fed by the local DER’s and if not considered, may result in damage to equipment,
harm personnel, or damage the recloser, thereby risk the evolving of a transient fault into
a permanent fault.

8
Chapter 3

Behaviour of a Microgrid under

Fault Conditions

Generally, a microgrid should be able to operate in both the grid connected mode and
islanded mode (without grid support). The MG is connected to the grid through a
static transfer switch (STS). Any analysis done using a microgrid hence must be done by
taking to account both operating modes and with special emphasis on their difference in
behaviour, and the transition phase between the two modes.

3.1 Grid Connected Mode

In the grid connected mode, the microgrid behaviour shall depend on transformer topolo-
gies (subsection 4.2.2) and microgrid grounding (subsection 4.2.3). In the event of a fault
in the MG, a large current may be fed into the MG by the grid, which far bypasses design
specifications of many MG elements. On the other hand, from the grid perspective, the
feeding of fault current by the DG sources leads to blindness of protection. Figure 3.1
demonstrates how a DG downstream of a relay leads to loss of sensitivity. More details
in section 1.1.1.

Figure 3.1: Blindness of protection/loss of sensitivity in an MG interfaced grid

[3]

In case of a fault on the grid side, subsection 4.2.2 explains the different cases that may

9
occur based on local topologies. However, most inverter based sources are not grounded,
which means that they may not supply any zero sequence current to the fault. However,
this may lead to a large unbalanced voltage spike which may cause spurious tripping. A
figurative example of the same is given in Figure 3.2

Figure 3.2: Coordination issues caused due to decentralised microgrid architecture

[3]

3.2 Islanded/Isolated Mode

While islanded, the MG loses the voltage reference, and the grounding support of the
grid. Since inverter interfaced systems are usually not grounded, this can lead to back
feed [12], a situation where the MG would be mainly grounded by the DG’s and their
transformers. Consequently, the fault response of the system under fault conditions de-
pend on the grounding strategy. If the DERs and their transformers are not effectively
grounded, LG fault-induced over voltages on the unfaulted phases can reach the L-L
voltage level.

Another major issue during operating in islanded mode is the low system inertia in
the MG (section 2.2) . In such a situation, the systems, the system tends to respond,
very quickly, to very small fluctuations in system parameters, Hence, system reliability
decreases, and the possibility of a chain reaction increases.

10
3.3 Transition phase
Normally, during a fault, the current sensing unit(CSU)[8] ensures fault detection and
disconnects the DG from the grid by using power electronic switches. CSU is a device
which is installed in the point of common coupling(PCC), and it’s function is to mea-
sure the current flowing through it to and from the grid. After islanding, the microgrid
resumes operations, and all the system protection devices within the microgrid operate
normally using settings calibrated for a post-fault islanded scenario. It is a simple and
inexpensive method of operation and it seems to have the following advantages:

1. Low cost and complexity

2. There is no need to change protection devices and their settings
3. Synchronization is only required for the DG disconnected

However, a few problems arise that complicate the situation. A microgrid is usually
dominated by power electronic devices which are notorious for their fast response to any
fluctuations. However, traditional circuit breakers usually operate with a delay, and hence
for proper coordination, these CB’s should be made of power electronic devices, which
operate faster.

Then again, grid code regulations may require a DG to stay connected to the grid
during voltage/frequency fluctuations until a certain threshold. Further, the behaviour of
the system protection devices of the microgrid should be to coordinate with the protection
device at the PCC and to not trip before the PCC disconnection. During this time,
all sources should continue to operate, even during switching and sensing of the PCC
device [10]. To that extent, all devices of the MG should have fault ride through (FRT)
capability.
Re synchronisation of an islanded section of the MG is a significant point of concern.
The process of re-synchronisation may either be manual or automatic, and it may need
anywhere from seconds to minutes depending upon the characteristics of the system. Re
synchronisation schemes have been classified into:[10]

1. Active synchronisation
2. Passive synchronisation
3. Open-transition transfer synchronisation

Active & passive synchronisation schemes generally have high reliability as compared
to open transition transfer schemes but these schemes tend to be complex and uneco-
nomical.

11
Chapter 4

Role of Grounding in Microgrid

Behaviour

The grounding scheme of a microgrid and it’s components invariably affects the following
behaviour:[12]

1. Fault-induced overcurrent (OC) and overvoltage stress

2. Protection system design and coordination
3. Safety of personnel and equipment
4. Reliability of energy supply
5. Harmonics
6. Maintenance costs

The grounding strategies of AC MG’s, and those of the traditional ac grids are affected
by similar constraints, thus having significant similarities. However, there are differences
due to the unique behaviour of the microgrid, such as islanded operation, or the higher
reliability expectations.

4.1 Types of Grounding

AC distribution systems are grounded by the neutral, through a device (surge arrestors).
Hence transformers and source generators with Y connected windings provide a neutral
wire available for use by design. In other cases such as ∇ connected windings, a neutral
point has to be obtained by using grounding transformers (zigzag and Y-∇ type trans-
formers). To effectively gauge the difference between the different types of grounding,
we shall use the LG fault current ratio, that is the ratio of a single line to ground fault
compared to a 3-phase to ground fault.

12
 (a)
(b)

Figure 4.1: Current vectors of an ungrounded system while in (a)Unfaulted condition ,

and (b) during an LG fault:[1]

4.1.1 Ungrounded
A Ungrounded system is intentionally not connected to the ground. However, any
ungrounded system is in fact capacitance grounded due to the presence of the capac-
itance generated by the conductor and the ground, and any stray capacitances. Hence
a grounded fault current in any of the phases would amount to the total capacitance
charging current, which generally amounts to about 1% of the maximum fault current.

However, LG faults in such systems may cause overvoltages in the unfaulted phases,
which may reach the line-to-line level. Under transient conditions, the overvoltages may
exceed 3 pu, which can be damaging, since appliances can only bear high voltages for a
very short duration, if at all. Such systems pose significant challenges for fault detection
and protection, since not only are the fault currents very low, but the voltage rise at the
bus means that response has to be fast to avoid sympathetic tripping.

Figure 4.1 shows the angle of the fault current vectors in an ungrounded system. A
LG fault causes a significant shift in the L-G voltage, and since the capacitance charging
current is phase shifted, actual fault current is actually 3 times the leakage current pre-
fault.

4.1.2 Solidly Grounded

In a this system, the neutral is grounded without any impedance. Therefore, the LG
fault current ratio is at least 60%, and the transient overvoltage caused by LG faults is
less than 1.5 pu.[12]

4.1.3 Resistance Grounding

So far it was evident that there is a certain trade-off between fault currents and high
overvoltages during fault conditions. The neutral of a resistance grounded system is

13
grounded by a resistance. The value of the resistance may be set according to system
specifications, and may vary depending upon how high voltages the system can handle
and how high the fault currents must be for reliable detection of faults.[12]

4.1.4 Reactance Grounding

In this, an inductor is used to ground the system, enabling limitinf the LG fault current
without leaking real power to the ground. They are classified into high and low reactance
grounding, based on the value of the reactance. In low reactance, LG fault current ratio
ranges from 25% and 100%, generally to achieve prevention of excessive transient overvolt-
ages. Hence, transient overvoltage during LG faults generally does not exceed 2.3 p.u.[12].

4.1.5 Resonant Grounding

Resonant grounding is achieved by specifically tuning the reactance of a reactance grounded
system to resonate or match the stray capacitance so that the LG fault current is limited
and in phase with the voltage. Transient LG overvoltages may reach about 2.73 p.u. [12]

4.2 Grounding in Microgrids

Grounding in Microgrids is derived from the grounding architecture of every component
of a MG. A good place to start would be from the source itself, since it is the primary
source of fault currents.

4.2.1 Types of Generators in a Microgrid

One major property of a MG is the varying types of sources that it can have, all of which
have differing properties and grounding configurations, all of which affect fault current
estimation adversely. A rough classification of the various types of DG in a MG include
[9]:

Type 1: Synchronous Generators

These generators are directly coupled to the grid. Some examples of these kind of sys-
tems are hydro micro-turbine generators, or thermal generators with CSP systems. Such
generators are characteristic of having very high short circuit current, upwards to a max-
imum of 10 pu of the rated current. Of course the fault current depends upon whether
the generator is grounded, the reactance of the grounding, and transformer topologies.

14
Type 2: Asynchronous Generators
Asynchronous generators, like the squirrel cage induction generator (SCIG), and the
wound rotor induction generator (WRIG), are also directly interfaced to the grid. Their
short circuit behaviour is similar to that of Type-1 generators; they also can feed short
circuit currents upto 10 pu. However, the fault currents tend to decay rapidly, and may
reach values below the rated current, as the asynchronous generator does not have an
independent excitation system. Hence detection of faults for a purely Type-2 based DG
system will have to be based on transient currents only, which makes fault detection and
subsequent action difficult.

Type 3: Doubly fed Asynchronous Generators

DFIG’s generally tend to have a crowbar scheme of protection of the rotor circuit. This
scheme, when activated, short circuits the rotor circuit, and the DFIG behaves like a
conventional synchronous generator, with a large transient current that decays quickly.

Type 4: Inverter Based Systems

Inverter based DG’s pose the biggest challenge in the protection of a microgrid. This
is because the behavior of inverters are completely different from traditional rotating
machines like Type 1-3 machines. Various properties of rotating machines have to be
modelled in generators(eg; droop and inverse droop[14]) to be able to interface them
with the grid. In any case, type 4 systems are characteristic of having a limited contribu-
tion to short circuit currents of up to 4 pu of the rated current, which can go as low as 2
pu as well. This is partially the consequence of the fast response of the inverter control
for the protection of the power electronic devices.

4.2.2 Transformer Topologies

As inverter based DG’s in a microgrid are not generally grounded, transformer topologies
play an important role in deciding the short circuit current, as grounded sources provide
a major source of zero sequence current. Figure 4.3 shows the equivalent zero sequence
circuit of different types of transformer windings. Flow of power depends upon these
configurations of the transformers on the MV and LV side.

Figure 4.2 shows the effects of common transformer winding configurations used in
MG’s. In Figure 4.2a, a ∇-Yg transformer is used. In such a case, the DER is not able
to feed a zero sequence current into a fault caused on the MV side. However, there are
two concerns. First, this leads to voltage rise on the unfaulted phases. Second, while a ∇
winding may not be able to source zero sequence current, it cycles through it’s windings.
Unattended fault may cause the ∇ windings to heat up and cause insulation damage

15
 (a)

(b)

(c)

Figure 4.2: Common transformer winding in microgrids:[12]

because of extra zero sequence current cycling through it.

Figure 4.2b shows a Yg − Yg winding. This system is prone to a high short circuit
fault current on the LV side, which may damage equipment on the LV side, which may
not be designed to handle grid side fault current levels, which may be as high as 30-50
pu of the rated current. This system is also prone to blinding and backfeed.

Figure 4.2c is an example of a Yg -∇ winding transformer. Here the transformer acts

as the source of the zero sequence current, and acts as a semi protection for the DER in
the case of a fault. However, this leads to loss of sensitivity of the protection devices on
the MV side.

4.2.3 Microgrid Grounding Configurations

The manner in which the microgrid is interfaced with the grid is crucial, and it affects
MG behaviour. Figure 4.4 shows the different configurations, which are explained in the
subsequent sections.

Three wire ungrounded configuration

This configuration is not grounded. Grounding support from the grid is lost, which re-
duces the fault current, making locating LG faults extremely difficult. Further, voltage

16
based protection techniques nonselective. Also, there is the risk of undetected or unat-
tended LG faults evolving into LL faults.

Three wire uni-grounded configuration

This configuration does not rely on any grounding device itself, but is connected to a
three wire unigrounded utility system. Such systems don’t necessarily suffer from OV
and fault detection problems in grid connected mode. However, when islanded, the MG
loses grid grounding support and all the problems associated with ungrounded systems
are switched in.

Four wire uni-grounded configuration

This configuration overcomes the problems of a three wire unigrounded configuration, as
the neutral continues to provide the grounding support to the MG even during islanding.
However, the MG may experience neutral voltage rise and large voltage imbalance if the
neutral is not solidly grounded or if the feeder is long.[12]

Four wire multi-grounded configuration

This configuration has many advantages which include:

1. Improved voltage transients

2. Optimal size of overvoltage protective devices.
3. Low zero-sequence impedance w.r.t. unigrounded config.
4. Safety

Such configurations are mainly used in cases where there are a lot of single phase loads.
They suffer from ground potential rise, large LG fault currents, and ground fault protec-
tion coordination issues, since there are many closed loops the zero sequence equivalent
circuit.

17
Figure 4.3: Zero Sequence Equivalent circuits of transformer winding topologies
[12]

18
Figure 4.4: Grounding configurations in ac systems
[12]

19
Chapter 5

Conclusion

It is evident that the current system of protection will fail in the case of large scale
penetration of DG and microgrids. Even in renewables, solar power has seen the max-
imum impact in fields of both scale and ease of implementation, and it is expected to
lead the growing penetration of renewables. The reason for solar power being singled out
is the fact that pv generation is integrated to the grid using inverters which have the
least similarity as compared to other sources of renewable generation, and it’s behaviour
is atypical and hence may need separate attention. Inverter based systems have many
unique properties and their behavior varies based on factors like grounding and type of
faults. That and the bidirectional flow of power means that microgrid protection systems
need additional functionality like directional support and communication between peers
to enable any basic functionality. Various operating modes have to be set in devices on
both sides of a microgrid to ensure that the relays operate on thresholds that is applicable
for the current situation.

5.1 Proposed Paradigms of Protection

Analysis of the aforementioned content indicates that the current systems of protection
have to be thoroughly revised when DG’s enter the scene. This prompts one to take a
closer look at which system parameters give reliable information on the case of faults.
[5] researches the different variables including sequence currents, voltages and harmonic
properties, to evaluate which have the highest accuracy in the case of the fault. The eval-
uation is done using a decision tree-based approach (random forest technique). Various
literature sources have concluded that for effective protection of micro grids, commu-
nication between devices is essential. [7] presents an adaptive algorithm for measuring
fault current in an inverter interfaced MG. [11] presents a wavelet and data mining based
signal processing approach to detect fault current threshold.

Another approach to understand the behaviour of a microgrid is to try to model it’s

characteristics. [6] attempts to model the small signal behaviour of a droop controlled

20
inverter based microgrid. However, literature also suggest using voltage based schemes
for protection. [2] suggests a voltage based relaying scheme for fault detection. [9],[13],[4]
document other paradigms of protection microgrids. ”

21
Bibliography

[1] Ieee recommended practice for grounding of industrial and commercial power sys-
tems. IEEE Std 142-2007 (Revision of IEEE Std 142-1991), pages 1–225, Nov 2007.
[2] H Al-Nasseri and MA Redfern. A new voltage based relay scheme to protect micro-
grids dominated by embedded generation using solid state converters. In 19th Inter-
national Conference Electricity Distribution, pages 1–4, 2007.
[3] Belwin J Brearley and R Raja Prabu. A review on issues and approaches for micro-
grid protection. Renewable and Sustainable Energy Reviews, 67:988–997, 2017.
[4] G Buigues, A Dysko, V Valverde, I Zamora, and E Fernández. Microgrid protec-
tion: Technical challenges and existing techniques. In International Conference on
Renewable Energies and Power Quality, volume 1, pages 222–227, 2013.
[5] E. Casagrande, W. L. Woon, H. H. Zeineldin, and D. Svetinovic. A differential se-
quence component protection scheme for microgrids with inverter-based distributed
generators. IEEE Transactions on Smart Grid, 5(1):29–37, Jan 2014.
[6] E. A. A. Coelho, P. C. Cortizo, and P. F. D. Garcia. Small-signal stability for
parallel-connected inverters in stand-alone ac supply systems. IEEE Transactions
on Industry Applications, 38(2):533–542, March 2002.
[7] Y. Han, X. Hu, and D. Zhang. Study of adaptive fault current algorithm for micro-
grid dominated by inverter based distributed generators. In The 2nd International
Symposium on Power Electronics for Distributed Generation Systems, pages 852–
854, June 2010.
[8] Seyed Amir Hosseini, Hossein Askarian Abyaneh, Seyed Hossein Hesamedin Sadeghi,
Farzad Razavi, and Adel Nasiri. An overview of microgrid protection methods and
the factors involved. Renewable and Sustainable Energy Reviews, 64:174–186, 2016.
[9] Patrick Tendayi Manditereza and Ramesh Bansal. Renewable distributed generation:
The hidden challenges–a review from the protection perspective. Renewable and
Sustainable Energy Reviews, 58:1457–1465, 2016.
[10] Aushiq Ali Memon and Kimmo Kauhaniemi. A critical review of ac microgrid pro-
tection issues and available solutions. Electric Power Systems Research, 129:23–31,
2015.

22
 [11] D. P. Mishra, S. R. Samantaray, and G. Joos. A combined wavelet and data-mining
based intelligent protection scheme for microgrid. IEEE Transactions on Smart Grid,
7(5):2295–2304, Sep. 2016.

[12] J. Mohammadi, F. Badrkhani Ajaei, and G. Stevens. Grounding the ac microgrid.

IEEE Transactions on Industry Applications, 55(1):98–105, Jan 2019.

[13] H. Nikkhajoei and R. H. Lasseter. Microgrid protection. In 2007 IEEE Power

Engineering Society General Meeting, pages 1–6, June 2007.

[14] Nagaraju Pogaku, Milan Prodanovic, and Timothy C Green. Modeling, analysis and
testing of autonomous operation of an inverter-based microgrid. IEEE Transactions
on power electronics, 22(2):613–625, 2007.

[15] Power System Relaying and Control Committee Subcommittee C Working Group
C30. Microgrid Protection Systems. IEEE Power Energy Society, July 2019.

[16] S.A. Soman. Digital protection of power systems. https://nptel.ac.in/courses/

108101039/.

23

View publication stats