Available online at www.sciencedirect.

com
Available online at www.sciencedirect.com
ScienceDirect
AvailableScienceDirect
Available onlineatatwww.sciencedirect.com
online www.sciencedirect.com
Energy Procedia 00 (2017) 000–000

ScienceDirect
Energy Procedia 00 (2017) 000–000
ScienceDirect
www.elsevier.com/locate/procedia
www.elsevier.com/locate/procedia

Energy
EnergyProcedia
Procedia141 (2017) 000–000
00 (2017) 428–431
www.elsevier.com/locate/procedia

4th International Conference on Power and Energy Systems Engineering, CPESE 2017, 25-29
4th International Conference September
on Power and Energy
2017, Systems
Berlin, Engineering, CPESE 2017, 25-29
Germany
September 2017, Berlin, Germany
Literature ReviewSymposium
The 15th International of Power System
on District Blackouts
Heating and Cooling
Literature Review of Power System Blackouts
Assessing the
Yuan-Kang Wuaa*, Shih
feasibility of using
Ming Changthe
a
, Yi-Liang Huaa
a heat demand-outdoor
Yuan-Kang
National WuUniversity
Chung-Cheng *, Shih Ming
, No.168, Chang
University Road,, 62102
a
Yi-Liang Hu
Chiayi, Taiwan
temperature National
function forUniversity
Chung-Cheng a long-term district
, No.168, University Road, 62102 heat
a
demand forecast
Chiayi, Taiwan

a,b,c a a b c c
Abstract I. Andrić *, A. Pina , P. Ferrão , J. Fournier ., B. Lacarrière , O. Le Corre
Abstract
a
IN+ Center for Innovation, Technology and Policy Research - Instituto Superior Técnico, Av. Rovisco Pais 1, 1049-001 Lisbon, Portugal
Blackouts in a power systemb can occur in several ways. How to prevent a blackout is an important issue in power systems. The
Veolia Recherche & Innovation, 291 Avenue Dreyfous Daniel, 78520 Limay, France
reasons to in
Blackouts cause blackouts
acDépartement
power system can
canbeoccur
the overloading of transmission
in several ways. How to prevent lines,a blackout
ice coating onimportant
is an lines, failure
issueofinprotection or control
power systems. The
Systèmes Énergétiques et Environnement - IMT Atlantique, 4 rue Alfred Kastler, 44300 Nantes, France
systems,
reasons toand causeothers. Blackout
blackouts can can
be thebe overloading
prevented through appropriate
of transmission control
lines, strategies
ice coating on in a system
lines, failure toof prevent from
protection or an N-1
control
contingency,
systems, and maintain load-generation
others. Blackout balance, and
can be prevented cascade
through to more fault
appropriate controlcontingencies.
strategies in This work to
a system reviews
preventblackouts
from an from
N-1
different countries
contingency, maintainby considering
load-generation theirbalance,
causes andand solutions,
cascade toas morewell as
faultthecontingencies.
discussion thoughts fromreviews
This work many researchers. The
blackouts from
survey
different
Abstractcancountries
provide significant
by consideringreferences to improve
their causes and the grid protection
solutions, as well strategies.
as the discussion thoughts from many researchers. The
© 2017can
survey Theprovide
Authors. Publishedreferences
significant by Elsevier Ltd.
to improve the grid protection strategies.
© 2017 The Authors. Published by Elsevier Ltd.
Peer-review
© 2017 The
District under networks
Authors.
heating responsibility
Published
are byof Elsevier
the organizing
commonly Ltd. committee
addressed in theofof CPESEas
literature 2017.
one of the most effective solutions for decreasing the
Peer-review under responsibility of the scientific committee the 4th International Conference on Power and Energy
Peer-review
greenhouse under
gas
Systems Engineering. responsibility
emissions from of the
the organizing
building committee
sector. These of CPESE
systems 2017.high investments which are returned through the heat
require
Keywords:
sales. Due Blackout,
to thePower System,
changed Protection,
climate Contingency.
conditions and building renovation policies, heat demand in the future could decrease,
prolonging the investment return period. Contingency.
Keywords: Blackout, Power System, Protection,
The main scope of this paper is to assess the feasibility of using the heat demand – outdoor temperature function for heat demand
1.forecast.
Introduction
The district of Alvalade, located in Lisbon (Portugal), was used as a case study. The district is consisted of 665
1.buildings
Introduction
that vary in both construction period and typology. Three weather scenarios (low, medium, high) and three district
In recent scenarios
renovation years, numerous blackouts
were developed in theintermediate,
(shallow, world havedeep). been Tooccurred,
estimate such as 2003
the error, North
obtained heatAmerican blackout,
demand values were
In recent
2006 European
compared years,
with numerous
blackout,
results from aand blackouts
2013heat
dynamic in
Indian theblackout.
demand world
model,haveThose been
previously occurred,
blackouts
developed such
caused
and as 2003
great
validated by North
financial American
losses andblackout,
the authors. brought
2006 European
The results
inconvenience showedtoblackout,
that when
people. andonly
There 2013
are Indian
weather
many changeblackout.
reasons Thosethose
is considered,
to cause blackouts
the margin ofcaused
blackouts, error great
could
such financial
acceptablelosses
asbetransmission for and
some
line brought
applications
tripping and
(the error in annual
inconvenience
overloading, failure demand
to people. was lower
There
of protection than
areormany 20%
reasons
control for all
to weather
systems, cause scenarios
those
voltage considered).
blackouts,
collapse, cybersuch However,andafter
as transmission
attack, introducing
others. line renovation
Thetripping and
experience
scenarios,
overloading,
learned fromthefailure
error value
those increased
of protection
historical up
ortocontrol
blackouts 59.5%
provides(depending
systems,
significanton thereferences
voltage weather
collapse, and renovation
tocyber scenarios
attack,
improve the and combination
others.
existing considered).
The grids.
power experience
This
The value of those
slope coefficient
learned
work from
reviews the current powerincreased
historical system on
blackouts average significant
provides
blackouts inwithin the range
the world of 3.8%toup
references
and investigates to 8% per
improve the decade,
the reasons existing that corresponds
power
that caused thegrids. to the
This
blackouts.
decrease in the number of heating hours of 22-139h during the heating season (depending on the combination of weather and
work reviews this
Additionally, the current
work also power system
proposes blackouts
several in the world
corresponding and investigates
improvement schemesthe reasons
to avoidthatpowercaused the blackouts.
system
renovation scenarios considered). On the other hand, function intercept increased for 7.8-12.7% per decade (depending on the
Additionally, this work also proposes several corresponding improvement schemes to
coupled scenarios). The values suggested could be used to modify the function parameters for the scenarios considered, and
avoid power system blackouts.
improve the accuracy of heat demand estimations.

© 2017 The Authors. Published by Elsevier Ltd.

Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and
* Cooling.
Corresponding author. Tel.: 886-939500016.
* Corresponding
E-mail address: allenwu@ccu.edu.tw
author. Tel.: 886-939500016.
E-mail address:
Keywords: allenwu@ccu.edu.tw
Heat demand; Forecast; Climate change
1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review
1876-6102 ©under
2017responsibility
The Authors. of the organizing
Published committee
by Elsevier Ltd. of CPESE 2017.
Peer-review under responsibility of the organizing committee of CPESE 2017.

1876-6102 © 2017 The Authors. Published by Elsevier Ltd.

Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the scientific committee of the 4th International Conference on Power and Energy
Systems Engineering.
10.1016/j.egypro.2017.11.055
 Yuan-Kang Wu et al. / Energy Procedia 141 (2017) 428–431 429
Yuan-Kang Wu et al./ Energy Procedia 00 (2017) 000–000

The main purpose of this investigation is to obtain potential protective mechanisms against blackouts in power
systems, and to help power systems design their prevention plans.

2. Historical blackouts in the world

This section introduces several power system blackouts around the world. The experience learned from those
blackouts provides significant references for improving existing power systems. Many power system blackouts have
occurred in large power systems, such as the blackout in North America or Europe. On January 24th 2006, a blackout
occurred on the Greek island of Kefallonia [1]. The main reason of that accident is the towers collapse and the
greatly loads with ice on overhead line conductors. The conductor surface of the high-voltage tower was coated with
an ice-sleeve approximately 15 cm in diameter. The consequences of the blackout caused the cut of electricity by
approximately 3000 MWh. Since the blackout occurred, additional substations have been commissioned in
Kefallonia Island. Additionally, the distance between towers has been re-established to endure the climate changes.
New and existing lines have been upgraded and derived into several anti-icing methods.
On August 13th 2003, a cascade blackout occurred throughout Ohio, Michigan, Pennsylvania, New York,
Vermont, Massachusetts, Connecticut, New Jersey, and Ontario, Canada [2]. The blackout of 61,800 MW lost lasted
around 4 to 7 days before the power fully restored, which affected around 50 million people. During the observation
of the blackout, several key causes had been identified. For instance, the state estimator (SE) and contingency
analysis of the Midwest ISO’s (MISO) were not under full automatic operation; they did not solve line outage
effectively. Additionally, the managements of load rising, increasingly limited generation, and reactive power
imbalance were inadequate at MISO. Inadequate Tree Trimming also causes this blackout; as long as the trees
around the high-voltage conductors are trimmed to a certain distance, short circuit could be avoided. After this
accident, the traditional network has been upgraded using smart grid’s technologies. The traditional power systems
used system control and data acquisition (SCADA) systems. However, SCADA systems cannot provide the view on
the whole system; moreover, the coordination for transmission is extremely slow and the system communications is
based on telephone calls. By contrast, the smart gird technologies provide advanced communication, monitoring,
pervasive control and digitalized systems. Notably, even though smart grids provide better control and monitoring
systems, the cyber-security brings the challenge to system operators.
One of the blackouts in the Indian recently is on July 30th and 31st 2012, two blackouts occurred and lasted
several hours causing over 620 and 700 million people, respectively, with the electricity lost. The first blackout was
caused by an overloading event that happened on the 400 KV Gwalior-Bina transmission double lines (one of the
double lines was under maintenance). Therefore, the overloading caused a cascade of tripping through the network,
leading to approximately 32 GW of generation shortage. Due the higher demand with the lack of generation and the
inadequate response to the issue, the system failed again on 31st of July. The reasons that cause the blackouts include
[3]:
 Planned maintenance work is not adequate.
 The forecasts about the changes of sudden atmospheric conditions are not adequate.
 Absence of control mechanisms (such as the static and dynamic reactive power controls) in HVDC for stabilizing
power grids.
 Inadequate response from State Electricity Boards to control voltage and frequency.
 Transmission congestion that leads to the malfunction of distance relays.
As presented above, the Indian blackout was mainly caused by the insufficient management of the power quality,
outage system, and peak load. In the past, the Indian grid measured the system parameters by remote terminal unit
(RTU). It is suggested to improve the real time monitoring from RTU to phasor measurement unit (PMU), aiming to
monitor the power system measurements at a rapid speed. The measurement by RTU is slow, and only provides
RMS values without phase angles; furthermore, there is no time correlation for RTU measurements. The PMU
measurements can improve the above mentioned technologies. Integration of distributed energy resources may be
another way to prevent blackout incidents. Small distributed generations could prevent the chain reactions for power
loss.
One of the most severe fault failures in Vietnam happens on May 22nd 2013. A fault on a 500 kV line tripped and
separated the Southern grid from the North and Central grid of Vietnam system. The blackout fault happened when a
430 Yuan-Kang Wu et al. / Energy Procedia 141 (2017) 428–431
Yuan-Kang Wu et al./ Energy Procedia 00 (2017) 000–000

truck moving a tree bumped onto a 500 kV line. This incident cascaded more tripping to generators and transmission
lines, causing voltage collapse in the Southern grid of Vietnam. Since this incident, several new protective scheme
for relay systems, including over-current, distance protection, out-of-step, over-excitation limiter, and under-
frequency load-shedding relays have been proposed [4]. According to the simulation results in [4], the over-
excitation limiter relay plays an important role to avoid the power system blackout in 2013 Vietnam system.
Voltage collapse is one of the main causes of blackouts in the world [5, 6]. Following a loss of a power system
element (lines, generators, transformers, or others), the other elements of the system would be overloaded, leading to
low voltage level at system buses. At that situation, the distance relays on transmission lines may trip. In addition,
overloaded components would be tripped, which propagates the overload problem to other elements of the system.
As several system elements are tripped, the lack of reactive support becomes more severe; therefore, the over
excitation limiter (OEL) relays of generators may be tripped as well, resulting in low frequency problem. Voltage
collapse does not generally occur by seconds, which depends on the system strength. It can happen in time frame of
minutes. This gives time for system to response to the voltage variation in voltage dynamics.
On November 1st 2014, Bangladesh Power System (BPS) had a nationwide blackout in a whole day. It was noted
[7] that a HVDC station had a sudden outage and spinning reserve did not respond; moreover, some generators in
the western region of BPS system were under-maintenance, which causes this blackout. In this incident, the actual
load shedding amount is much lower than the expected one. For instance, as the frequency falls to 48.9Hz, the
expected shedding load is 69MW but the actual one is only 25MW. This load shedding process went on for 5 stages
and eventually caused the blackout. It is suggested [7] that the BPS system should maintain the critical operating
inertial reserve to be greater than 9s, improve the under-frequency protection scheme by considering the rate of
change of frequency, and controlling grid islanding techniques to establish the fast response Bangladesh power
system.
Several power systems have started to develop smart grids, including the functions of load balancing, peak
curtailment, sustainability, demand-response support, two way communications, and computing technologies. Those
smart grid developments allow the grid to be reliable and easy to control. However, they also face security issues.
For example, internet is an efficient communication network to file important databases but it is also easy for
hackers to access to these personal databases. On December 23rd 2015, Ukraine faced a cyber-attack that caused a
blackout that affected approximately 225,000 customers for several hours. The main cause for this blackout is the
cyber-attack; the attack hijacked the SCADA network. The attack injected false data to the network that caused the
system to collapse. Those attackers have the knowledge of power system operation and the system features. They are
also capable of manipulating meter measurements as well as the knowledge to the network topology, system
electrical parameters, specifics of the SCADA network devices and bad data detection scheme. The only solution is
to strengthen the defensive system to prevent blackouts, disturbance or faults from a cyber-attack. According to [8,
9], they provide good protective methods to prevent cyber-attacks in a smart grid.
On September 8, 2011, an outage in the Pacific Southwest caused the whole San Diego to lose power for 12 hours.
This blackout not only affected 1.5 million customers in San Diego, but also affected parts of Arizona, southern
California, and Baja California, Mexico, with a total of approximately 2.7 million customers to lose power. The
main reason to cause this blackout is line trip during the peak load. The electric transmission network in southern
California and Arizona was designed to be able to sustain a single line trip of 500 kV. However, the tripping of the
Hassayamapa-North Gila 500 kV line (H-NG), a major transmission that connects the power from Arizona through
Imperial Irrigation District (IID) to San Diego area, during the peak load caused its surrounding lines to overload and
tripped. The study in [10] indicated that the San Diego blackout was caused by insufficient load shedding. According
to the simulation results from [10], the expected load shedding amount should be larger than the actual one.

3. Potential improvement schemes to avoid power system blackouts

A power system can be protected in many ways; however, different models and systems may have different
methods to optimize the protection. Additionally, many power grids have been changed by installing new
components or control schemes; therefore, the original protection scheme and grid codes should be revised,
including the revision of the design of relay systems, the special protection system, the load shedding mechanism,
the grid coordination mechanism, and the measured and control system. Furthermore, renewable energy has been
 Yuan-Kang Wu et al. / Energy Procedia 141 (2017) 428–431 431
Yuan-Kang Wu et al./ Energy Procedia 00 (2017) 000–000

increased remarkably; therefore, the corresponding revision on the protection strategies for incorporating renewable
energy is also important. Generally, the causes of power system blackouts include dynamic or static stability loss,
voltage collapse, voltage instability in transmission networks, inappropriate load shedding, multiple tripping of
overloaded lines and others. The potential improvement schemes would consist of grid reinforcement, increase of
tie-line, re-design of load shedding strategy, planning of suitable locations for new generators, installing more
measured and control systems, optimization of reactive power and voltage controls, installing advanced protection
relays, and automatic operations and alarm during severe incidents. The modern power system should be capable of
self-healing and prevention against blackouts, automatic re-closing of transmission lines, and component
reintegration. After a power system blackout, the power system should also be restored automatically and quickly.
Additionally, modern power systems could increase their stability and avoid blackouts using state-of-the-art
controllers, sensors, and products, such as high voltage direct current (HVDC) elements or novel special protection
schemes.

4. Conclusion

This work reviews blackouts from different countries by considering their causes and solutions. These countries
include Greece, North America, Indian, Vietnam, Bangladesh and Ukraine. According to this investigation, the
power system blackout in different countries was caused by different reasons. However, if the power system can be
enhanced, or cascading failures can be predicted, the probability of power system blackout would become lower.
This work reveals that dynamic or static stability loss, voltage collapse, voltage instability in transmission networks,
inappropriate load shedding, and multiple tripping of overloaded lines would cause power system blackouts.
Furthermore, this work summarizes important methods to improve system stability and avoid power system blackout.

Acknowledgements

This work is financially supported by the Ministry of Science and Technology (MOST) of Taiwan under Grant
105-2221-E-194-032-. Project title: Study on the Preventive Control Strategy for an Offshore Island System by
Considering System Flexibility, Renewable Power Forecasting and Demand Response.

References

[1] S.D. Anagnostatos, C.D. Halevidis, A.D. Polykrati, P.D. Bourkas, C.G. Karagiannopoulos “Examination of the 2006 blackout in Kefallonia
Island, Greece” Int J Electr Power Energy Syst, 49 (2013), pp. 122–127.
[2] J.E Chadwick, "How a smarter grid could have prevented the 2003 U.S. cascading blackout," IEEE Power and Energy Conference at Illinois
(PECI), UIDC, Feb. 2013
[3] S.Sarkar, G. Saha, G. Pal, T. Karmakar “Indian Experience on Smart Grid Application in Blackout Control” in 2015 National Systems
Conference (NSC), 2015 1-6
[4] H. Nguyen-Duc, H. Cao-Duc, C. Nguyen-Dinh, V. Nguyen-Xuan-Hoang, "Simulation of a power grid blackout event in Vietnam". IEEE In
Power & Energy Society General Meeting, pp. 1　5, 2015
[5] C. W. Taylor, Power System Voltage Stability. McGrawHill, 1994.
[6] S. Abe, Y. Fukunaga, A. Isono, and B. Kondo, “Power system voltage stability,” IEEE Trans. Power App. Syst., vol. PAS-101, no. 10, pp.
3830–3840, Oct. 1982 G. Morison, G.Bao, and P. Kundur, “Voltage stability analysis using static and dynamic approaches,” IEEE
Transactions on Power systems, vol. 8, no. 3, pp. 1159–1171, 1993.
[7] A. Kabir, M. Hasan Sajeeb, N. Islam, H. Chowdhury “Frequency Transient Analysis of Countrywide Blackout of Bangladesh Power System
on 1st November”. 2015 International Conference on Advances in Electrical Engineering (ICAEE), 2015, pp. 267-270.
[8] NCCIC/ICS-CERT, “Cyber-attack against Ukrainian critical infrastructure,” released 25 February 2016.
[9] E-ISAC and SANS, “Analysis of the cyber attack on the Ukrainian power grid: Defense use case,” released 18 March 2016.
[10] E.C. Portante, S. F. Folga, J. A. Kavicky, L. T. Malone, “Simulation of the September 8, 2011, San Diego Blackout,” Simulation Conference
(WSC) 2014, pp. 1527 – 1528.