ADAPTIVE NETWORK PROTECTION IN MICROGRIDS Alexandre Oudalova, Antonio Fidigattib a

ABB Switzerland Ltd., Corporate Research, Segelhof 1, CH-5405 Dättwil, Switzerland

INTRO
Distributed Energy Resources (DER)
environmental and economic benefits for end customers, utilities and society.
Renewable Energy Resources (RES)
micro-sources close to the load has the advantage of reducing transmission losses as well as preventing
network congestions.
power supply interruption of end-customers is diminished due to islanded mode
1. In the first case the protection system should isolate the microgrid from the main grid as rapidly
as necessary to protect the microgrid loads.
2. In the second case the protection system should isolate the smallest part of the microgrid when
clears the fault
problems related to selectivity (false, unnecessary tripping) and sensitivity (undetected faults or delayed
tripping) of protection system may arise.
two main issues, first is related to a number of installed micro-sources in the microgrid and second is
related to an availability of a sufficient level of short circuit current in the islanded operating mode
operating conditions of microgrid are constantly changing because of the intermittent micro-sources
(wind and solar) and periodic load variation.
Also a network topology can be regularly changed aimed at loss minimization or achievement of other
economic or operational targets and a loss of relay coordination may happen setting parameters of relays
must be checked/updated periodically

In modern digital relays, a tripping short-circuit current can be set for a wide range, e.g. 0.6-15*CB rated
current

A use of power electronics (PE) interfaces leads to a number of challenges in microgrid protection,
especially in the islanded mode.
F1
CB1 should operate if no MV tripping, sensitivity problems microgrid contribution to SCC 1.1-
1.2*IDERrated A directional OC relay acting on CB1 is a feasible solution the setting has to be continuously
monitored and adapted when microgrid generation undergoes considerable changes (related to the number
and type of connected DERs)
Distribution Network Operators (DNO)

F2
a distribution transformer OC protection clears the fault by opening CB0. CB1 is opened simultaneously
by “follow-me” function (hardware lock) of CB0. In case of hardware lock failure, a possible fault
sensitivity problem can arise as in the case of fault F1. Typical solutions are similar to the F1 case,

F3-Grid Connected
disconnect the smallest possible portion of the LV feeder by CB1.2 and CB2.1
CB1.2 is opened due to a high level of short-circuit current supplied by the main MV grid
CB 1.1 is the backup for 1.2
If there is a synchronous DER between CB1.1 and fault the fault current can become smaller. (protection
blinding) may result in a delayed CB1.1 tripping
CB1.2 operates first then part is islanded with the fault CB2.1 can have sensitivity issue should have a
“follow me” function for CB1.2 and CB2.1

F4 -Grid Connected
High fault current CB2.4 trip, if not SWB2 isolated by CB2.5 and DER is cut off

F3-Islanded
Absence of high-level SCC, directional OC required to detect the fault location
There are two possible ways to address the problem of absence of a high level short-circuit current:
1. Install a source of high short-circuit current (e.g. a flywheel or a supercapacitor) In case such
source of high short-circuit current is installed, it would be typically connected to the LV bus bar.
2. Install an adaptive microgrid protection using on-line data on microgrid topology and status of
available micro-sources/loads.

F4-Islanded
Low SCC with only DER contribution CB2.4 settings selected for the main grid connected mode are just
slightly higher than rated load current It assures that the end-customer site will be disconnected even if
only DERs with PE interfaces are available in the microgrid.

Adaptive Protection
Use of numerical directional OC relays
several settings groups that can be parameterized locally or remotely automatically or manually.
Use of new/existing communication infrastructure

microgrid central controller (MCC)

Pre-calculated information during off-line fault analysis of a given microgrid

A set of meaningful microgrid configurations as well as feeding-in states of DERs (on/off) is created for
off-line fault analysis and is called an event table.
On-line operating block
MCC monitors the microgrid state by polling individual directional OC relays.

Directional interlock
if The fault current direction is opposite to the present interlock direction
then Keep present port assignment
and Relay sends locking signal to the present interlock direction
if The fault current direction is similar to the present interlock direction
then switch output/input ports, i.e. interlock direction
And relay sends locking signal to the new interlock direction (opposite to the present interlock
direction)

Simulation
Assumptions
Over current protection trip curves have similar shape for each breaker
L part - protection against overloads
S part – protection against short circuit with short time
delay trip
I part – instantaneous protection against short circuit
ts is changed to provide selective operation
The instantaneous tripping part I is removed from all
curves for a simplification purpose
Implementation of Dijkstra’s Algorithm in a Dynamic Microgrid for Relay Hierarchy Detection Taha
Selim Ustun, Cagil Ozansoy, Aladin Zayegh

Disadvantages of above
implements an algorithm which includes a look-up table. This is a large set-back because it requires the
knowledge of all possible microgrid configurations beforehand, plus human input for the preparation of
this table ,any kind of a new deployment, which is very common to microgrids, requires that the whole
selectivity table should be re-written

SELECTIVITY ISSUES AND RELAY HIERARCHY

Selectivity is a well-known protection concept which means isolating the fault with the nearest relay in an
effort to minimize its effect on the rest of the system
In conventional protection for passive systems – downstream relays operate first
With DG penetration, downstream and upstream concept changes with microgrid status

Selectivity hierarchy changes with the change in microgrid topology eg. Case 1 CB 5 open – two
branches, case 2CB 4 open CB5 closed – single branch
CB 2 should operate for faults in the utility grid, need two different time delays one for upstream and one
for downstream

tmax max fault clear time

n number of selective levels

t base = tmax/n
t relay of nth level =t base x S relay

tassigned = t relay – t comm delay

DIJKSTRA’S ALGORITHM AND ITS IMPLEMENTATION

FOR RELAY HIERARCHY DETERMINATION

N value is found with dij for each node (breaker) and time delay is calculated

ENs. IEC 61850, for example, is an international standard for substation automation

ISSN 1816-949X
Fredman-Tarjan algorithm
Graph theory based, minimum spanning tree,
Conclusion – Dijkstra is faster than fredman tarian algo

BACKEND DATABASE AIDED ADAPTIVE PROTECTION IN MICROGRID

P.Balamurugan1, V.Ananthakrishnan2, Dr. J.L. Febin Daya3 , Dr.O.V.Gnana Swathika4
VIT University Chennai
gnanaswathika.ov@vit.ac.in

Overcurrent fault is the most common fault in microgrids.

The Genetic Algorithm

The relay coordination problem is stated as an optimization problem.

1. Simulating the microgrid test system in MATLAB, and noting the current values between all buses, for
different faults.
2. Using the current values from the above step to find TMS values for the relays using Genetic
Algorithm.

Relay Coordination in Microgrid

Pallavi P.Bagul#1,Prof.Sonali M.Akolkar*2
K.K.Wagh Institute of Engineering
Education & Research Nasik
422003,India

must have the specified operation sequence

Whenever the minimum fault occur in the system the primary relay must operate first depending on the
location of the fault, and then the backup relays must trip only when the primary relays are not operated
correctly.
time interval between the primary and backup protection is called the coordination time interval (CTI)
usually considered as 0.2 s to 0.5 s.
can be applied to all protection devices such as fuses and reclosers

tbackup − tprimary ≥ CTI

introduction of DG increase fault current, reduces the CTI may cause to unnecessary removal of a larger
portion of the system

Therefore, the operating parameters of OCRs must be correctly adjusted and that is carried out by the
adaptive algorithm implementation
1) Power Grid without DG.
2)Power Grid with all DG.
3) Power Grid with different capacity of DG.
4) Power Grid with different fault location.

Not completed……
IEEE Elect r i f icat ion Magazine / MARCH 2014
Adaptive Protection System for Microgrids By Liang Che, Mohammad E. Khodayar, and Mohammad
Shahidehpour
IEEE Standard 1547 requires DER units to stop energizing the distribution system when the system is de-
energized due to faults.

Fault Current-Level Modifications

grid-connected mode, utilities contribute to microgrid fault
islanded mode, microgrids’ potential fault currents are lower.
lower fault currents would not trip overcurrent (OC) relays
DER units, especially generators with rotating prime movers, have higher contributions to
fault currents when compared with DERs with power electronic interfaces.

Impact of DERs on the Operation of Protective Devices

DER units have plug-and-play characteristics
xx Malfunction of PDs due to downstream faults
xxSympathetic tripping

Impact of Microgrid Topology on the Coordination of Protective Devices

fault current in a loop is divided between two parallel paths.

Possible solutions

Synchronous generators, including permanent magnet synchronous generators (PMSGs), and fly wheels
are suitable choices for increasing fault currents in microgrids.
Differential protection schemes, either centralized, or localized