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

net/publication/301231659

Past, Present and Future Challenges of the Marine Vessel's Electrical Power
System

Article · April 2016

DOI: 10.1109/TTE.2016.2552720

CITATIONS READS
38 1,342

6 authors, including:

Espen Skjong Rune Volden

Blue Ctrl AS 4 PUBLICATIONS   49 CITATIONS   
19 PUBLICATIONS   166 CITATIONS   
SEE PROFILE
SEE PROFILE

Egil Rødskar Marta Molinas

Ulstein Design & Solutions Norwegian University of Science and Technology
3 PUBLICATIONS   75 CITATIONS    391 PUBLICATIONS   4,810 CITATIONS   

SEE PROFILE SEE PROFILE

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

Operation and control of micro-grids in the Colombian power system and the non-interconnected zones View project

Assistive Robot Control based on the Fusion of EEG and EMG Signals View project

All content following this page was uploaded by Espen Skjong on 24 October 2017.

The user has requested enhancement of the downloaded file.

IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION 2016 1

Past, Present and Future Challenges of the Marine

Vessel’s Electrical Power System
Espen Skjong, Rune Volden, Egil Rødskar, Marta Molinas, Member, IEEE, Tor Arne Johansen, Senior Member,
IEEE, and Joseph Cunningham

Abstract—The evolution of the use of electricity in marine outbreak of WWII stimulated new developments that brought
vessels is presented and discussed in this article in an historical the T2-tanker with turbo-electric propulsion into the picture.
perspective. The historical account starts with its first commercial Nuclear powered vessels emerged in the late 1950s and the
use in the form of light bulbs on the SS Columbia in 1880 for
illumination, going forward through use in hybrid propulsion first passenger liner to use alternating current was inaugurated
systems with steam turbines and diesel engines and then transi- in 1960 (SS Canberra), 70 years after the invention of the
tioning to the present with the first fully electric marine vessel alternating current motor. In the period 1956-1985, the power
based entirely on the use of batteries in 2015. Electricity use electronics revolution triggered by the innovative solid-state
is discussed not only in the light of its many benefits but also technology marked the beginning of a new era for marine
of the challenges introduced after the emergence of the marine
vessel electrical power system. The impact of new conversion vessels; the era of the all-electric vessel. As a result of that,
technologies like power electronics, battery energy storage, and Queen Elizabeth II was inaugurated in 1987 with the first
the DC power system on overall energy efficiency, power quality, diesel-electric integrated propulsion system. And in the last
and emission level is discussed thoroughly. The article guides two decades, the marine vessel community has witnessed the
the reader through this development, the present and future development of the first vessels having LNG as fuel. In January
challenges by calling attention to the future research needs and
the need to revisit standards that relate to power quality, safety, 2015, marking the start of the era of the all-electric vessel,
integrity, and stability of the marine vessel power system, which the world’s first purely battery-driven car and passenger ferry
are strongly impacted by the way electricity is used in the marine Ampere was placed in use and is being regularly operated in
vessel. Norway. Fig. 1 guides the reader through the milestones in
Index Terms—Marine vessel electrical power system, diesel- the evolution of the the marine vessel electrical power system
electric propulsion, steam turbine, power electronics, battery from 1830 to 2015.
energy storage, power quality, harmonics This new era of electric marine vessels does not come with-
out challenges, however. In what follows, the paper highlights
I. I NTRODUCTION the different stages in the evolution of the marine vessel’s
development and the impact of electricity use in this evolution.
TARTING with the earliest records of a commercially
S available shipboard electrical system which date back to
the 1880s with the onboard dc system of the SS Columbia; the
Following the historical account, the paper moves towards
modern electric ship propulsion discussing the new challenges
of moving towards hybrid AC/DC and pure DC power systems,
invention of the AC induction motor, the transformer, and the
the challenge of electrical stability, harmonic pollution, and
diesel engine triggered new research and development toward
power quality in stand-alone microgrids like the marine vessel,
the end of the 19th century and the beginning of the 20th.
the role of battery energy storage systems, and the move
In this period, the initial steps were made in research related
towards emission free operation among others. Along with
to submarines, batteries, steam turbines, and diesel engines.
these challenges, potential solutions and possible roads to
The two most important developments before WWI were the
follow are presented.
first diesel-electric vessel (Vandal) in 1903 and the first naval
vessel with electric propulsion in 1912 (USS Jupiter). During II. E ARLY S TEPS OF THE M ARINE V ESSEL
the period of rising tension that preceded WWI the first cargo E LECTRIFICATION
vessels with turbo-electric propulsion were conceived and
developed in the United States and the United Kingdom. The The first recorded effort to apply electric power on a marine
vessel occurred in the late 1830s after Moritz Hermann Jacobi
E. Skjong, R. Volden and E. Rødskar are with Ulstein Power & Control of Germany invented a simple battery powered direct current
AS, 6018 Ålesund, Norway (dc) motor which was installed experimentally on small boats
E. Skjong, M. Molinas and T. A. Johansen are with the Department of
Engineering Cybernetics, Norwegian University of Science and Technology, [1]. It suffered from numerous imperfections and there was no
7034 Trondheim, Norway immediate adoption of electric propulsion for ships. The first
E. Skjong and T. A. Johansen are with the Centre for Autonomous Marine successful application of electric power on ships was that of
Operations and Systems (AMOS), Norwegian University of Science and
Technology, 7052 Trondheim, Norway gun firing circuits in the 1870s. The development of arc lamps
J. Cunningham is an adjunct lecturer, engineering development and history, for illumination of streets and public spaces was followed by
at various institutions and associations. arc lamp searchlights on ships to illuminate attacking ships
Email: [email protected], [email protected],
[email protected], [email protected], and blind enemy gunners. Luxury liners were equipped with
[email protected], [email protected] call bells for the convenience of passengers [2].
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION 2016 2

;ϭϵϭϲͿh^^EĞǁDĞdžŝĐŽĂŶĚ
DũƂůŶĞƌ;ĐĂƌŐŽͿ͗
ƋƵŝƉƉĞĚǁŝƚŚƚƵƌďŽͲĞůĞĐƚƌŝĐ
ƉƌŽƉƵůƐŝŽŶ;Ϳ

;ϭϵϭϴͿtƵůƐƚLJĂƐƚůĞ͗ ;ϭϵϯϱͲϯϲͿ ;ϮϬϭϱͿD&ŵƉĞƌĞ͗

:ĂĐŽďŝ͛Ɛ ;ϭϴϴϬͿ^^ ttϮ DŽĚĞƌŶ^ŽůŝĚͲ^ƚĂƚĞ
&ŝƌƐƚǀĞƐƐĞůǁŝƚŚƚƵƌďŽͲĞůĞĐƚƌŝĐ ;ϭϵϮϭͲϮϮͿ ^ĞĐŽŶĚ ;ϭϵϲϬͿ^^ dŚĞǁŽƌůĚ͛ƐĨŝƌƐƚĨƵůůLJ
ĞdžƉĞƌŝŵĞŶƚ͗ ^ŚŝƉďŽĂƌĚ ŽůƵŵďŝĂ͗ ^ƵďŵĂƌŝŶĞƐ͗ WŽǁĞƌůĞĐƚƌŽŶŝĐƐ
ƉƌŽƉƵůƐŝŽŶŝŶ'ƌĞĂƚƌŝƚĂŝŶ tĂƐŚŝŶŐƚŽŶ >ŽŶĚŽŶEĂǀĂů ĂŶďĞƌƌĂ͗ ĞůĞĐƚƌŝĐďĂƚƚĞƌLJƉŽǁĞƌĞĚ
^ŵĂůůĞůĞĐƚƌŝĐ ĂƌĐůŝŐŚƚƐ &ŝƌƐƚŝŶƐƚĂůůĞĚ ZĞƐĞĂƌĐŚĨŽƌ ZĞǀŽůƵƚŝŽŶ͗
dƌĞĂƚLJ͗ dƌĞĂƚLJ͗ &ŝƌƐƚƌŝƚŝƐŚ ƉĂƐƐĞŶŐĞƌĂŶĚĐĂƌĨĞƌƌLJ
ƉƌŽƉĞůůĞĚ ;ŵŝĚͬůĂƚĞ ƐŚŝƉďŽĂƌĚ /W;ŝƌ dŚLJƌŝƐƚŽƌ;ϭϵϱϲͿ͕
;ϭϵϭϵͿƵďĂ͗ ǀŽŝĚŝŶŐŶĂǀĂů >ŝŵŝƚŝŶŐǀĞƐƐĞů ƉĂƐƐĞŶŐĞƌ
ďŽĂƚƐ;ůĂƚĞ ϭϴϳϬƐͿ ĞůĞĐƚƌŝĐĂů /ŶĚĞƉĞŶĚĞŶƚ WŽǁĞƌͲDK^&dƐ
&ŝƌƐƚƉĂƐƐĞŶŐĞƌǀĞƐƐĞůǁŝƚŚ ĂƌŵƐƌĂĐĞ ƐŝnjĞĂŶĚŐƵŶ ůŝŶĞƌƚŽƵƐĞ ;ϮϬϭϱͿtŝƌĞůĞƐƐWŽǁĞƌ
ϭϴϯϬƐͿ ƐLJƐƚĞŵ;Ϳ WƌŽƉƵůƐŝŽŶͿ ;ϭϵϳϬͿ͕/'d;ϭϵϴϱͿ
ƚƵƌďŽͲĞůĞĐƚƌŝĐƉƌŽƉƵůƐŝŽŶ ĐĂůŝďƌĞ dƌĂŶƐĨĞƌ

ϭϴϰϬ ϭϴϴϬ ϭϵϬϬ ϭϵϮϬ ϭϵϰϬ ϭϵϲϬ ϭϵϴϬ ϮϬϬϬ ϮϬϭϱ

ttϭ ttϮ

;ϭϴϵϬƐͲϭϵϭϬƐͿ/ŶŝƚŝĂů ;ϭϵϬϴͿ:ŽƐĞƉŚDĞĚŝůů͗&ŝƌĞͲďŽĂƚ
/ŶĚƵĐƚŝŽŶDŽƚŽƌ͗ ;ϭϵϭϴͿ ;ϭϵϰϮͿdϮͲ EƵĐůĞĂƌWŽǁĞƌĞĚ >E'ĂƐ&ƵĞů͗
^ƵďŵĂƌŝŶĞZĞƐĞĂƌĐŚ͗ ǁŝƚŚƚƵƌďŽͲĞůĞĐƚƌŝĐƉƌŽƉƵůƐŝŽŶ ;ϭϵϯϬͿ ;ϭϵϱϬƐͿ,D^ ;ϭϵϴϲͬϴϳͿ
ŽŶĐĞƉƚĚĞƐĐƌŝƉƚŝŽŶďLJ tƵůƐƚLJ ƚĂŶŬĞƌ͗ sĞƐƐĞůƐ͗ ϮϬϬϬ͗D&'ůƵƚƌĂ
/ŶĐůƵĚŝŶŐďĂƚƚĞƌLJ͕ >ŽŶĚŽŶ džƉůŽƌĞƌĂŶĚ YƵĞĞŶ
&ĞƌƌĂƌŝƐĂŶĚdĞƐůĂ;ĞĂƌůLJͬ ĂƐƚůĞ͗ ^ƚĂƌƚĞĚ ^ƵďŵĂƌŝŶĞh^^ ;ĨŝƌƐƚĨĞƌƌLJͿ
ƐƚĞĂŵͲƚƵƌďŝŶĞƐĂŶĚ ;ϭϵϭϭͿůĞĐƚƌŝĐƌĐ͗dƵƌďŽͲĞůĞĐƚƌŝĐ EĂǀĂů ,D^ ůŝnjĂďĞƚŚ//͗
ŵŝĚϭϴϴϬƐͿ͕WĂƚĞŶƚĞĚďLJ &ŝƌƐƚǀĞƐƐĞů ďƵŝůĚŝŶŐ EĂƵƚŝůƵƐ;ϭϵϱϱͿ͕
ĚŝĞƐĞůͲĞŶŐŝŶĞƐ ;ĂĐͿ dƌĞĂƚLJ͗ džĐĂůŝďƵƌ͗ &ŝƚƚĞĚǁŝƚŚ
dĞƐůĂ;ϭϴϴϳͿ ǁŝƚŚƚƵƌďŽͲ ͨŶĂǀLJ ϮϬϬϯ͗sŝŬŝŶŐ
džƚĞŶĚŝŶŐ ^ƵďŵĂƌŝŶĞƐ ĚŝĞƐĞůͲ
ĞůĞĐƚƌŝĐ ŽŝůĞƌƐͩǁŝƚŚ ^ƵƌĨĂĐĞEĂǀLJsĞƐƐĞů ŶĞƌŐLJ͕^ƚƌŝůWŝŽŶĞƌ
;ϭϵϬϯͿsĂŶĚĂů͗ ;ϭϵϭϮͿh^^:ƵƉŝƚĞƌ͗ tĂƐŚŝŶŐƚŽŶ ǁŝƚŚ/W ĞůĞĐƚƌŝĐ
;ϭϴϴϲͿůĞŬƚƌĂ͗&ŝƌƐƚ ƉƌŽƉƵůƐŝŽŶ ƚƵƌďŽͲ h^^>ŽŶŐĞĂĐŚ ;ĨŝƌƐƚW^sƐͿ
&ŝƌƐƚǀĞƐƐĞůǁŝƚŚĚŝĞƐĞů &ŝƌƐƚŶĂǀĂůǀĞƐƐĞůǁŝƚŚĞůĞĐƚƌŝĐ dƌĞĂƚLJ͕ ;ŚLJĚƌŽŐĞŶ ;ŝŶƚĞŐƌĂƚĞĚͿ
ĞůĞĐƚƌŝĐĂůůLJƉƌŽƉĞůůĞĚƐŚŝƉ ŝŶ'ƌĞĂƚ ĞůĞĐƚƌŝĐ ;ϭϵϱϵͿ͕
ĞŶŐŝŶĞƚĞĐŚŶŽůŽŐLJ ƉƌŽƉƵůƐŝŽŶŝŶh^ ůŝŵŝƚŝŶŐ ƉĞƌŽdžŝĚĞͿ ƉƌŽƉƵůƐŝŽŶ
ƌŝƚĂŝŶ ƉƌŽƉƵůƐŝŽŶ ϮϬϭϱ͗/ƐůĂĞůůĂ
;ĚŝĞƐĞůͲĞůĞĐƚƌŝĐͿ ƐƵďŵĂƌŝŶĞ
;ϭϴϴϲͿ&ŝƌƐƚůĞĐƚƌŝĐĂůůLJ DĞƌĐŚĂŶƚsĞƐƐĞůE^ ;ĨŝƌƐƚĐŽŶƚĂŝŶĞƌ
;ϭϵϭϯͿdLJŶĞŵŽƵŶƚ͗ŝĞƐĞůͲ ǁĂƌĨĂƌĞ
ƉƌŽƉĞůůĞĚƐƵďŵĂƌŝŶĞ ^ĂǀĂŶŶĂŚ;ϭϵϱϵͿ ƐŚŝƉͿ
ĞůĞĐĐƚƌŝĐ;ĂĐͿ

Fig. 1. Historical highlights of the development of the Marine Vessel’s Power System from 1830 to 2015 [2].

The development of incandescent lighting by Thomas Edi- was the Hungarian team of Kàroly Zipernowsky, Ottò Blàthy,
son and others was followed by an installation on the passen- and Miksa Déri, (ZBD) whose closed core transformer of
ger and freight vessel SS Columbia in 1880. That consisted high efficiency made practical ac power distribution and was
of 120 lamps powered by a set of dynamos; the system was adopted by Westinghouse [5], [6].
crude with lead wires functioning as fuses and lamp intensity Still, the issue of ac power factor (the useful power deliv-
was regulated only by the engine room crew’s adjustment of ered after losses due to inductive and capacitative reactance)
the generators according to the appearance of the lamps [3]. constrained the adoption of ac for commercial power, railways,
Nonetheless, it led the US Bureau of Navigation to mandate and on ships. However, dc systems were heavier and larger;
additional installation of electric lights. Soon after, electric thus in an effort to reduce weight ac systems were designed
motors were installed in ventilation and gun firing circuits. for frequencies up to 400 Hz but the mechanical frequency
In 1896, the USS Brooklyn was fitted with an 80 volt dc converters of the time were cumbersome. Practical ac propul-
electrical system to operate winches, deck machinery, and gun sion was demonstrated in 1908, though without modern power
mounts [2]. Most installed shipboard power systems were dc electronics control was complex; effected by voltage and
as alternating current (ac) motors were not yet perfected. The frequency changes and variations in pole connections. It is
first successful electrically powered vessel was the Elektra, said that the complexity of ac systems led the British Navy to
a passenger ferry with a capacity of 30 persons, built by the retain dc systems, even though Germany followed the lead of
German firm Siemens & Halske in 1885. Measuring 11 meters the US which had adopted ac systems in 1932 [2].
long by 2 meters wide, it was powered by a 4.5 kW motor
supplied by batteries [4]. B. Turbo-electric Powered Vessels
In the early 1900s, Britain favored the development of
A. Alternating Current Motor and Transformer steam turbine drive systems with reduction gears while the
The development of ac motors based on the inductive US focused on electric drive with the first turbo-electric drive
effect of phase displaced conductors; primarily by Nikola installed in 1908. Rated at 400 shaft horsepower, it was
Tesla (US), Galileo Ferraris (Italy), and Michael Osipowitch installed on the Joseph Medill, a fireboat [1], [7]. Four years
Dolvio-Dobrowolsky (Germany), made possible an alternative; later, the collier USS Jupiter became the first naval vessel
however, the reliable dc motors developed by Frank Sprague to adopt turbo-electric propulsion. That was an experiment,
(US) and others tended to favor the use of direct current the ship also included a diesel engine and a direct coupled
motors. Regardless, ac research continued; George Westing- steam turbine. The 3,500 hp turbo electric system supplied by
house took the lead in the United States. Often overlooked General Electric was deemed a success and the Navy decided
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION 2016 3

to convert all front line battle ships to electric power. The to continue limits imposed on navies until Japan withdrew
Jupiter went on to become the first aircraft carrier the USS from the agreements in 1934. An arms race followed, and the
Langley [8], [9]. United States commenced construction of battleships though
The first battleship to adopt turbo-electric propulsion was electric propulsion was not adopted due to concern for elec-
the USS New Mexico. Launched in 1914, it was equipped trical system vulnerability to damage during battles and also
with a pair of 11.5MW 3,000/4,242 volt dual voltage, variable a concern for the additional weight which could be better
frequency generators that powered four 7,500 hp 24/36 pole utilized for weapons or armor [2], [13].
induction motors and was capable of a speed of 21 knots. Electric propulsion was adopted by the United States for
The shaft alley was shorter and thus less of a target, and the navy oiler, a tanker that supplied oil to ships at sea. With
fuel economy was improved substantially. All that came at the a maximum power of 7,240 hp and a speed of 15 knots it
expense of weight the electric motors and controls were heavy had a range of 12,600 miles. 481 were constructed during
though reversal was accomplished easily by the switching of the war [2]. Diesel-electric submarines of various types were
circuits with no need to change steam systems [2], [8]. constructed in large numbers during World War II.
The first passenger vessel to incorporate the new system
was the Yorktown. Built as the Cuba in 1894, after a 1916
III. T OWARD M ODERN E LECTRIC S HIP P ROPULSION
wreck it was reconfigured as a turbo electric system in 1919
[2], [10]. Electric drive was not limited to the US. In Sweden The development of the mercury pool rectifier for power
the shipbuilder Rederiaktiebolaget Svea constructed in 1916 a conversion in the early 1900s produced a practical alternative
pair of cargo ships. One of them, the Mimer, was constructed to mechanical power conversion; both as a rectifier and also as
with steam power; the other, the Mjölner, was equipped an inverter. Solid state power electronics emerged in the 1960s
with electric drive; radial flow reaction turbines powered ac and 70s to enable significant advances in the power systems of
induction motors coupled to a single shaft through reduction ships. The first British passenger liner with alternating current
gears [11]. Two years later the cargo ship SS Wulsty Castle propulsion was the SS Canberra in 1960. Equipped with three
was constructed in Britain with a similar drive system. 6,000 volt synchronous motors that produced 85,000 hp, the
While the steam engine was practical for land based power most powerful ever installed on a ship, they enabled it to cruise
generation, the low efficiency of fuel consumption led to a at 27.5 knots. Separate generators supplied non-propulsion
search for a better method. Rudolf Diesel, a German inventor, loads [14], [15].
patented the diesel engine in 1892 and licensed production in The use of power electronics to maximize fuel efficiency
Sweden and Russia. In 1903, constant speed diesel engines became a trend in the 1980s. In 1984, the Cunard Line re-
coupled to an electric transmission were installed in the Van- equipped the Queen Elizabeth II with nine German MAN
dal, a river barge sailing on the Volga River that transported diesel engines coupled to an electric transmission. The system
coal to St. Petersburg and also to Finland [1], [2]. was designed such that only seven engine sets were required to
maintain the design speed of 28.5 knots, which thus effected
C. Submarines a fuel savings of 35% [2].
Power electronics found extensive application in off-shore
The availability of electric power for illumination, commu-
vessels such as Platform Supply Vessels (PSV) and other
nication, and propulsion had fostered the concept of an all-
service ships. Dynamic Positioning (DP) systems required
electric ship. It was therefore logical to extend that concept to
sophisticated control systems to maintain position in special-
submarines for which a practical power source had remained
ized operations. Diesel electric propulsion was the standard
an elusive goal. There had been much experimentation with
method, though LNG was also adopted in the early 2000s.
the concept of underwater crafts during the 19th century;
Nuclear reactors for steam turbine systems were developed
propulsion varied from manual to stored compressed air,
initially for submarines the USS Nautilus of 1954 being
even pressure from chemical reactants. In 1885, the French
the first such vessel, the USS Long Beach the first nuclear
designer Claude Goubet had introduced electric propulsion
powered surface vessel followed in 1959. That same year the
with a pair of experimental submarines. By 1900, France,
first passenger and cargo ship, the NS Savannah was launched
the United States, and Britain were exploring the submarine
[16], [17].
concept, the latter two nations expanded on the work of John
In the constant drive for greater fuel economy, hybrid drive
Philip Holland. Most of those schemes focused on the internal
ships have been developed; the propulsion supplied by gas
combustion engine for surface operation and the charging of
turbine direct drive or electric motors supplied by diesel
batteries for use when submerged [2], [12].
engine-generator sets, the system configured as needed to
maximize fuel efficiency.
D. Effect of World Wars I and II Fuel cells and Battery Energy Storage Systems (BESS)
Germany made extensive use of submarines to attack ships to adapt ships to renewable energy sources have emerged
during WW I; subsequently the United States, Britain, and recently. In January 2015, the world’s first fully electric pas-
Japan engaged in an arms race. That was stopped by treaties in senger and car ferry, the MF Ampere was launched. Capable
the 1920s which limited or forbade entirely the construction of of accommodating 120 cars and 360 passengers, it makes a
new, or the reconstruction of, existing vessels thus effectively 30 minute crossing between Oppedal and Lavik near Bergen,
halting technical development. Subsequent agreements sought Norway. One MW of battery capacity supplies the ferry, the
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION 2016 4

battery sets charged when the vessels arrive at the docks. became the main trend in shipboard power system designs.
Operated by Norled AS, the ferry is a product of the Fjellstrand One of the reasons for this was that the early dc systems
shipyard and Siemens AS [2]. (without power electronics) needed rotating devices to trans-
form the power from one voltage level to another [23]. The
IV. P ROPERTIES AND C HALLENGES OF THE M ARINE ac power system has been the most used power system in
V ESSEL’ S P OWER S YSTEM marine vessels, but now, with modern power electronics and
Electrical power systems for marine vessels have existed other technological advantages, the discussion of using dc or
for more than 100 years, and history has shown a high level ac distribution in shipboard power systems has been brought
of research and innovation, to bring the early applivations of to the table and some of the key points whether to use ac
shipboard electricity of the 1880s to modern power systems. or medium-voltage dc (MVdc) are (adopted from [20], [24]–
Vessels today often consist of an ever increasing electrical [26]):
load: The majority of the propulsion systems and auxiliary
loads, such as weapon systems in naval vessels, hotel and • Impedance: MVdc power systems are capable of pro-
service loads in a cruise vessel, and station-keeping (DP) viding greater energy dynamics than the classical ac
systems for subsea operations, are of an electrical type. The power systems due to elimination of many components
power is, in general, generated from prime movers using e.g. for power conversion and optimizing the use of the cables
diesel and/or gas, or from nuclear plants with a turbo-electric (only ohmic resistance). The dc distribution doesn’t expe-
configuration. In many modes of operation the power systems rience skin effect in the power transmission, as is the case
need to be reliable and exercise a high level of survivability. in ac transmission. Also, due to the lack of a fundamental
The Naval Sea Systems Command states the design philosophy frequency, the dc system does not have a power factor,
for naval power systems very well [18]–[20]: and depending on the voltage levels, the weight of cables
The primary aim of the electric power system design may decrease for a given power level. Unlike the dc
will be for survivability and continuity of the elec- system, the ac system has reactive currents that increase
trical power supply. To insure continuity of service, the losses, which thus reduce the energy transportation
consideration shall be given to the number, size and capability. Cable impedance in an ac system causes a
location of generators, switchboard, and to the type current-dependent voltage drop along the cable, however
of electrical distribution systems to be installed and the impedance of the cable automatically limits the short-
the suitability for segregating or isolating damaged circuit currents. In dc systems only the (very low) ohmic
sections of the system. resistance of the cables limit the short-circuit currents,
This design philosophy does not only apply to naval ships. thus all parts of the power system are equally effected by
Vessels that exercise dangerous operations, such as DP op- a short-circuit at an arbitrary position. This effect, and
erations near offshore structures, or operations in which in the missing natural zero-crossing of the ac current makes
general, any failure could have a high economical or environ- it hard to break a connection (bus-tie/circuit breaker) or
mental consequence, need power systems with high levels of even limit the dc current, which may endanger power
reliability and survivability and electrical stability. converting devices that contain power electronics.
On the commercial side the vessels should be fuel efficient, • Prime mover speeds: In dc systems the speeds of
thus keeping the emissions (air pollution) to a minimmum the prime movers can be altered, as the prime mover
and the fuel costs low. One of the most critical issues facing speeds are largely decoupled from the power quality of
ship owners and builders today is stricter regulations for the grid. Since frequency control is not a concern, the
emissions, such as the International Maritime Organization’s prime movers can run at optimized speeds (relative power
(IMO) MARPOL air pollution regulations [8], [21]. Due to demand with the objective of increased fuel efficiency)
these stringent exhaust emission regulations, a lot of focus connected to generators with an arbitrary number of
has been devoted toward technology such as exhaust catalysts, poles.
electronically injected common rail diesels, and waste-energy • Connection of parallelled power sources: In ac systems
recovery, such as heat-exchange systems. Also alternative fuel, parallelled power sources must be both voltage and phase
such as LNG, has also found its way to a broader market. matched before being connected to the power system. In
Properties (and requirements) such as reliability, survivabil- a dc system the phase matching is not needed, resulting
ity, and continuity of electrical power supply, sustainability, in a faster power generation response time.
and efficiency can all be related to the power system’s design, • Power Electronics Conversion System: In dc systems,
electrical stability, and manner of operation. In the following medium or high frequency transformers (dc-ac-dc elec-
some of the aspects of the shipboard power system’s ongoing tronic transformers) can be used resulting in a smaller
design trends, properties, and challenges will be discussed. For footprint. On the other hand, in ac systems the transform-
a thorough introduction to the most common shipboard power ers make an easy and reliable adaption of the voltage
system designs it is referred to [22]. levels, however the conversion system often includes a
dc-link stage. Hence, using a cable connection instead of
A. AC vs DC the internal direct connection of the dc-links between the
The early shipboard power systems were of a dc type, but source- and load-side of the converter leads to a dc grid.
with the introduction of the ac motor this changed and ac Linking the dc-links from the converters directly will
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION 2016 5

demand a sufficiently high dc-link voltage in the order B. Marine Vessel Power Systems and Microgrids
of 10kV. Using back-to-back converters with internal dc-
links, which are state of the art, this dc-link voltage can Microgrids are electrically and geographically small power
be reduced by adapting to the high ac-side voltage by systems capable of operating connected to, or islanded from,
a transformer, at the cost of increased weight and space a national grid [27]. In islanded mode, the microgrid has
and reduced efficiency. strict requirements imposed such as energy independence and
• Fault currents and circuit breaker technology: In dc service quality for an extended period. Marine vessel’s power
systems the fault currents can be controlled to levels systems are indeed microgrids; they are isolated (and islanded)
considerably lower than in ac systems. This is because while at sea) and part of a terrestrial grid while docking and
power electronics can be used instead of conventional connected to shore power. Shipboard power systems have a
circuit breakers. Lower fault currents will also reduce lot in common with terrestrial stand-alone microgrids; many
damage during faults. On the other hand, the ac systems of the methods and a lot of equipment and components are
can use much simpler circuit breaking technology than the same [28]. In addition many control strategies and design
dc systems as electrical arcs clear at zero-crossing of the principles used in microgrids may be applicable for shipboard
current. power systems, and also the other way around. Examples
• Acoustic signature: The dc system does not have a of such control strategies and design principles are voltage
significant acoustic signature, as is the case with ac and frequency control schemes, power quality improvement
systems due to a common fundamental frequency. This strategies, power sharing methods for multiple distributed
can be an important feature for naval vessels. However, generators, and energy management systems [29]–[33]. A
the constant magnetic field created by dc current can thorough overview of technical cross-fertilization between ter-
leave a residual magnetic field in ferrous materials, which restrial microgrids and ship power systems is presented in [27].
contributes to the overall ship magnetic signature. This Some of the main differences between a shipboard micrigrid
tends to be, among other things, a disadvantage with and larger terrestrial (commercial) grids are summarized in the
regards to mines and sensor/equipment interference. following [20], [27]:
• Weight and space: In dc systems, high-speed gas tur- • Frequency: The shipboard power system’s fundamental
bines can be used in conjunction with high-speed gener- frequency cannot be assumed constant. Due to limited
ators, without the need for reduction gears for frequency rotational inertia of the prime movers and the generators,
control, which is often the case in ac systems. A mated rapid load changes can cause fast acceleration and de-
high-speed gas turbine and generator enables a shorter celeration of the motor shafts, which causes frequency
generator with reduced footprint. This is desirable due fluctuations. Such fluctuation may last for a couple of
to space and weight savings. For constant power, the dc seconds until the speed of the shafts reach a steady state
system needs only two conductors compared to the ac that coincides with the reference frequency.
system, which needs three. Removing one conductor is • System analysis: In analysis of a commercial grid all
beneficial due to weight savings. the system’s time constants are quantified and used to
analyze the problem by time-scale separation. However,
such analysis is not easy to conduct in a shipboard power
system due to the principal time constants for motor
New technological advances, such as the modular multilevel dynamics, electrical dynamics, and controls which all lie
converters (MMC) can, in special configurations, solve many in the same time range of milliseconds to seconds.
of the issues and challenges of dc power grids, thus making • Planning of power generation: In a commercial grid
the dc system a more interesting solution in shipboard power the power delivered by each generating unit is scheduled.
systems than before. Even though the MVdc solution may The difference between consumed and produced power is
lead to reduced weight, increased efficiency, and offer high- regulated through equipment acting as swing generators.
energy transport capability at low losses, challenges such This is not the case in a shipboard power system as
as short-circuit currents, dc-breaker technology, and system all the generators share the active and reactive power
standardization must be solved [24]. The different power through fast exchange of load-sharing information, which
system solutions, whether it is pure ac, a hybrid between ac amplifies the parallelled generators’ dynamics. Hence,
and dc or pure dc, have different properties, advantages and instead of generator scheduling the shipboard power
disadvantages. The choice of power system (pure ac, ac/dc generation exhibits load sharing, which is often realized
or pure dc) will be strongly dependent on, among others, by generator droop control.
available technology and different components from different • Electrical distances and load flow: In the commercial
manufacturers, developer and customer preferences, most eco- power sector it is important to model the electrical
nomical solution, type of equipment connected to or powered distances (transmission lines) in the power distribution to
by the power system, possibilities for energy storage, space achieve the right dynamics and proper voltage regulation.
and weight requirements, the level of redundancy and rules and This is not the case in a marine vessel’s power system
regulations from classification entities. These aspects, along as the electrical distances are short, thus trivializing the
with an economical point of view, will influence in shaping load-flow problem. The short electrical distances result in
the power system solution. low impedance which increases the coupling between the
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION 2016 6

^t and the auxiliary loads are separately powered by dedicated

ƵdžŝůŝĂƌLJ
generators [22]. Fig. 2 illustrates the main structural difference
&ƵĞů WƌŝŵĞ 'ĞŶĞƌĂ WŽǁĞƌ between the conventional (segregated) power system and the
DŽǀĞƌ;ƐͿ ƚŽƌ;ƐͿ ;ƐĞƌǀŝĐĞ IPS.
ůŽĂĚƐͿ
The propulsion system in a conventional power system
was originally a mechanical-drive system with reduction gears
&ƵĞů WƌŝŵĞ ^ŚĂĨƚ connecting the prime movers to propeller shafts. Many vessels
'ĞŶĞƌĂ WŽǁĞƌ WƌŽƉ͘
DŽǀĞƌ;ƐͿ ƚŽƌ;ƐͿ ŽŶǀĞƌƐŝŽŶ DŽƚŽƌ were converted to electric propulsion to gain faster response,
which resulted in the separated conventional electric-drive
power system. Even today there exist numerous vessels with
this kind of power system. As Fig. 2a indicates, the conven-
(a) Conventional electric-drive power system: Separated, or segregated,
power generation for propulsion and auxiliary loads.
tional power system consists of two separated subsystems; one
^t
for propulsion and one for auxiliary loads. Due to the separa-
ƵdžŝůŝĂƌLJ tion between the subsystems, the engines of each subsystem
WŽǁĞƌ are only connected to their respective systems and can only
;ƐĞƌǀŝĐĞ
&ƵĞů WƌŝŵĞ 'ĞŶĞƌĂ ůŽĂĚƐͿ be used within that subsystem. This configuration has been
DŽǀĞƌ;ƐͿ ƚŽƌ;ƐͿ the leading design for ensuring maneuverability; almost 90%
WŽǁĞƌ
WƌŽƉ͘ ^ŚĂĨƚ of the vessel’s generated power is locked into the propulsion
ŽŶǀĞƌƐŝŽŶ
;ƌŝǀĞͿ DŽƚŽƌ system [34]. However, this separation, where the majority of
the vessel’s power supply are limited to the propulsion system,
can be a disadvantage as the propulsion power is not available
(b) Integrated power system (IPS): Integrated power generation for propul-
for other mission specific systems.
sion and auxiliary loads. To tackle the disadvantage with the conventional power
Fig. 2. Simplified drawing illustrating the main structural difference between system, the IPS was introduced as a solution. Instead of
conventional and integrated power systems. separating power generating units into stand-alone subsystems,
the IPS shares all generated power from all the generators on
an integrated power grid, which distributes the power to all
different parts of the power system. Hence, to strengthen individual consumer systems located throughout the grid in a
coupling between devices and subsystems the assurance utility fashion. The IPS’s ability to share the generated power
of stability needs proper attention. between all (online) consumers is also an important property
• System’s size and extent: Due to the shipboard power for easing aftermarket installations of electric equipment, as
system’s limited extent, a higher level of centralized new equipment is simply connected to the distribution grid.
control can be applied than in commercial systems. The The property of power sharing is the main advantage of IPS,
shorter electrical distances also facilitate easier synchro- and improves power flexibility (operational flexibility) and
nization of data and measurement retrieval than in a availability. At low- and medium-speed ranges, the IPS can
commercial grid. generate the same amount of power as a conventional power
• Load profile: In a shipboard power system the load system with fewer running prime movers. This is preferable
profile is often rapidly changing due to the power de- both from an economical and an environmental point of view,
mand from the propulsion system and other high-rated as fewer running generator sets (gensets) will enhance the
systems and equipment. Hence, the power (both active fuel efficiency and reduce exhaust emissions. By starting and
and reactive) is changing more rapidly in a shipboard stopping gensets relative to the vessel’s power demand, the
power system than in commercial distribution systems. IPS provides a stepped power generation, and by equipping
• Single line faults: A shipboard power system is designed a vessel with gensets of different power ratings, the power
to continue operation with a single phase (line) to ground. production could be optimized to avoid low non-ideal loading
For safety reasons such medium voltage systems always conditions of the prime movers. However, this is seldom the
include high impedance grounding systems. case since all or multiple gensets in a vessel are often of same
• Environmental effects: A shipboard power system must size to make maintenance and access to spare parts easier. In
be able to operate in a tough environment, which is addition, if the IPS operates with open bus-ties (see Fig. 3),
characterized by vibrations, shock and motion dynamics, both sides should have the same power generating capacity.
and should survive salinity and moisture. The future shipboard power system may have an elegant
solution to the optimal prime mover loading problem involving
Energy Storage Systems (ESS) [35], [36], that can store excess
C. Integrated Power System (IPS) and Grid Design power to achieve ideal prime mover loading conditions, which,
In an Integrated Power System (IPS), or integrated-electric among other scenarios, can be used to give a green approach
ship, all the required power, for the vessel’s propulsion and to harbors without emissions.
auxiliary (service) loads, is generated and distributed by the 1) Electrical Stability: Reliability, dependability, and sur-
same main generators. In comparison, in a conventional (seg- vivability are important properties for many shipboard power
regated) electric-drive vessel power system, the propulsion systems. A naval vessel must be able to survive an attack
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION 2016 7

^ƚĂƌďŽĂƌĚůĞĐƚƌŝĐWƌŽƉƵůƐŝŽŶ^LJƐƚĞŵ

WƌŽƉ͘
DŽƚŽƌ
&ƵĞů WƌŝŵĞ 'ĞŶĞƌĂ
DŽǀĞƌϭ ƚŽƌ

EŽŶͲǀŝƚĂů

D^ϭ
^ĞƌǀŝĐĞ>ŽĂĚƐ
^ĞƌǀŝĐĞ
&ƵĞů WƌŝŵĞ 'ĞŶĞƌĂ dƌĂŶƐĨŽƌŵĞƌ

ĞŶƚĞƌϭ
DŽǀĞƌϯ

>ŽĂĚ
ƚŽƌ ϭ

ƵƐdŝĞ
sŝƚĂů
ƵƐdŝĞ

ƵƐdŝĞ
WŽǁĞƌ'ĞŶĞƌĂƚŝŽŶ d ^ĞƌǀŝĐĞ>ŽĂĚƐ
^LJƐƚĞŵ

ĞŶƚĞƌϮ
>ŽĂĚ
^ĞƌǀŝĐĞ
&ƵĞů WƌŝŵĞ 'ĞŶĞƌĂ dƌĂŶƐĨŽƌŵĞƌ
DŽǀĞƌϮ ƚŽƌ Ϯ
EŽŶͲǀŝƚĂů ^ĞƌǀŝĐĞ>ŽĂĚ
^LJƐƚĞŵ
D^Ϯ
^ĞƌǀŝĐĞ>ŽĂĚƐ

&ƵĞů WƌŝŵĞ 'ĞŶĞƌĂ

DŽǀĞƌϰ ƚŽƌ
WƌŽƉ͘
DŽƚŽƌ

WŽƌƚůĞĐƚƌŝĐWƌŽƉƵůƐŝŽŶ^LJƐƚĞŵ

Fig. 3. Example of a typical redundant IPS for PSVs and small-medium naval ship. Redundancy for bus-tie breakers connecting Main SwitchBoard (MSB).
Redundant power supply for vital loads using Automatic Bus Transfer (ABT) [22], [34].

where parts of the power system are down, but still be able to to a switchboard (MSB 1 and 2), and the switchboards are con-
bring the ship away from the situation and have the power nected through redundant bus-ties. Each switchboard supplies
needed for initiating defense measures. An offshore vessel one propulsion system, and both switchboards are serving the
conducting a station-keeping operation (DP) near offshore service loads. The load center is split in two switchboards.
structures needs to survive single faults and have the power The vital service loads have redundant power supply from
needed to bring the vessel to a safe position away from the both switchboards using an Automatic Bus Transfer (ABT)
structures. In the same way, a deep-sea drilling vessel must unit, while the non-vital loads are served by one of the
have a reliable power system that survives faults and maintains switchboards, one on each side of the vessel. Depending on
station-keeping to avoid critical situations that can harm both the vessel type and class regulation from classification entities,
equipment and crew. the IPS may include an emergency generator supplying vital
• Reliability is often explained as a fail-safe operation [34], loads, and in some cases part of the propulsion loads. The
and the term system reliability is a standard measure for bus-tie between the load center switchboards has the ability to
the effect of component failures and internal errors and is connect the switchboards if, for instance, one of the service
calculated using component mean time to failure (MTTF) transformers fails. The IPS is equipped with many breakers,
statistics and static dependency analysis [37]. which may be used to isolate faults from propagating through
• Dependability is given as the system’s ability to continue the grid and causing a complete blackout. Hence, this property,
operation despite component failures, internal errors and reconfigurability, is important for achieving the needed system
exogenous disruptions. reliability, and is closely related to the IPS’s practical design
• Survivability, on the other hand, is mostly used for naval and installation, as well as fast and reliable fault detection
vessels and military applications and deal with continuity systems that are able to invoke protection schemes isolating
of vital services during major disruptions associated with the faults.
battle and damage control operations.
2) Radial and Zonal Grid Designs: Traditionally, the
In many settings the terms are mixed together, and reli- practical solution to provide redundant power distribution
ability often comprises both dependability and survivability. was to install alternate power routes between components
To achieve a reliable IPS, which cultivates both dependability using longitudinal cables connecting vital loads to multiple
and survivability, the most used design principle is redundancy, switchboards. This solution, a radial distribution, was shown
however, spatial separation and manual backup systems have to be a bulky and heavy solution with the ever-increasing
also been used to a great extent. number of vital electrical loads. As a solution, the zonal
An often used redundant two-split IPS design for small and distribution grid was introduced in the 1990s, where the
medium size vessels is shown in Fig. 3. As can be seen, the redundant power supply was realized by providing vital loads
power generating units are split in pairs, each pair connected with alternate power routes using shorter transverse feeder
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION 2016 8

Generator
Switchboard
Bus Transfer
Load Center
Bus Tie Point
Profile View Profile View
SWBD Load
Center

Plan View Plan View

Radial Distribution System Zonal Distribution System

Fig. 4. Comparison of radial and zonal power distribution systems in a marine vessel [20].

cables from port and starboard switchboards [34], [37]. This go to the propulsion systems [34], flowing through power
may be seen as stretching the switchboards along the vessel’s electronics devices. Power electronics is at the heart of power
longitudinal axis, one switchboard for starboard side and one conversion, and, because of this, the IPS includes numerous
for port side. Bus-ties are used to isolate faults, or segregate different power electronics devices to be able to supply the
parts of the switchboards. With this solution, the long feeder right form and level of power to the connected systems and
cables in a radial system are removed, with the effect of equipment. The shipboard electrical power demand continues
reduced cost and weight - which again leads to lower fuel to increase, from tens of MW and in some cases even greater
consumptions and emissions. The zonal distribution topology than 100MW [38]. However, such high power ratings lead
is usually adopted in the IPS design philosophy, enabling eas- to power electronic devices that are both heavy and have a
ier aftermarket installations of equipment and more flexibility large footprint. This is a real handicap for serving high power
regarding installation of redundant solutions for achieving a demands. In general, in the given order of priority, size, losses,
design with the needed level of reliability and survivability at cost, and weight are interrelated factors that limit acceptable
relatively low cost. An illustration showcasing the differences applications of power electronics.
between radial and zonal grids is given in Fig. 4. An important power electronic device is the con-
It is expected that tomorrow’s power system design so- verter/inverter, which is able to convert the electric power
lutions will be completely different from today’s solutions. from one form to another, i.e. ac/dc, ac/ac, dc/ac, dc/dc. The
Future shipboard power systems should aim for a higher necessary power for each load, or group of loads with the same
quality of service (QOS), increased reliability and efficiency power requirements, are in an IPS converted at point-of-use.
as key requirements, which may be achieved by, among In fact, almost all power sources and loads need a converter.
other means, a completely new design strategy, advanced Pulse-Width-Modulation (PWM)1 has been widely used for
monitoring of system health and state as part of new sensor modulating small- and medium size converters. A switch-mode
technology, and advanced and efficient stability and power power electronics converter, which consists of switches that are
quality improvement methods and devices. either on or of, uses PWM to control the time the switches
are on and off, and by this, the converter (which in fact is
D. Power Electronics and Harmonic Pollution an array of switches) can be programmed to produce voltage
Electricity enables a more flexible way to utilize energy and current waveforms, different power factors, and obtain a
than any other energy source. Technology such as infor- desired frequency from a range of different input waveforms.
mation systems, radar and sonar systems, advanced motion From this point of view, there is little difference between
compensation, and military precision weaponry would not be motor drives, power supplies, and active power filters, and the
possible without electricity. Future predictions show that more composition of power electronics in such devices can be gen-
and more equipment is of an electrical type, and the marine eralized to form a Power Electronics Building Block (PEBB)
vessel is asymptotically converting towards an All-Electric [39], [40]. These building blocks are intended to minimize
Ship (AES), where all installed equipment and systems are the number of different power electronics devices in a power
of an electrical type [22], [38]. The broad variety of electrical system and can be mass-produced due to their generality. The
equipment and systems connected to the power system require general design will also allow the power electronics to be
different power conversions. Some of the equipment and tightly packed, which will reduce weight and footprint. The
systems are powered by ac, while others are powered by dc. blocks may be controlled by different algorithms and software
In addition, the needed (and rated) voltage levels may span solutions, through a general interface (communication proto-
from a few volts to thousands of volts, and different systems col), and can be changed in the field, depending on operational
and equipment may require different frequency levels. Almost
90% of the vessel’s generated power may at some points 1 Often realized with a hysteresis control scheme.
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION 2016 9

status or mission type. The blocks can easily be installed (plug However, the VSC introduces harmonics dependent on the
and play) with an interface which allows information sharing modulation frequency, which may be 1kHz or higher. LCL
between the components. Depending on the way the blocks are filters are often used to suppress the harmonics generated
connected to each other, different algorithms may be deployed by the VSC, but LCL filters are passive devices and tuned
as part of a configuration scheme; and, depending on the for a given modulation frequency. If, for some reasons, the
power system’s status and classification requirements, different VSC changes its modulation frequency the LCL filters have
algorithms may be enabled to perform functions such as power to be re-tuned. In addition, the harmonics from a VSC may
conversion, harmonic mitigation (as an active filter), active or cause harmonic resonances due to interaction with passive
reactive power control, or inherit a simple breaker’s properties filters [43]. Hence, harmonic pollution can, to some extent,
to isolate faults. Due to its generality, a wide range of different be suppressed by design, but the ever-increasing number of
modelling and simulation tools may be developed around this electrical devices, which are directly or indirectly dependent
block, which will ease power system design and realization on power electronics, will introduce even more non-linear
dramatically, thus ensuring stability, reliability and efficiency. elements into the power system, making harmonic mitigation
An important part of developing PEBB is the continuation and power conditioning devices a necessity.
of improving power electronics in the sense of minimizing Harmonic pollution is defined as any waveform with fre-
weight, size, and losses, to achieve components that can handle quencies that are multiples of the fundamental frequency, and
more heat and have faster dynamic response with increased is measured as Total Harmonic Distortion (THD), which is a
power ratings. The PEBB is seen as tomorrow’s solution for normalized quantity describing the relation between the am-
advanced power systems. Even though a lot of research and plitudes of the harmonic frequencies and the amplitude of the
development has been devoted to realizing such a standardized fundamental frequency. Most shipboard power systems today
building block, a general solution has not yet become available are affected by harmonic pollution in some or another way
on the market. [43]. Harmonic pollution, which impairs the power system’s
A lot of research has also been conducted towards power power quality, leads to higher fuel consumption and emission.
semiconductor devices, which consist of a variety of diodes, Harmonics are closely connected to reactive power, and high
transistors, and thyristors. New designs have produced compo- levels of harmonics may lead to equipment and system break
nents with better performance and lower losses, but few of the down, and even cause catastrophic events like explosion and
designs have reached the market. Also silicon carbide (SiC) fire [43]. Theoretically, this can, in the worst case, cause a
has been devoted attention due to the material’s properties complete blackout as a result of voltage collapse. A complete
which leads to lower switching losses, high voltage and high blackout may occur due to high levels of harmonic pollution,
temperature capabilities. SiC devices are expensive, but have but is usually caused by operational mistakes. The term voltage
a huge impact on converter size, losses, weight, cooling dip ride through capability is often used to describe the
requirements and potential for high PWM frequencies [20], consumers’ ability to cope with faults and malfunctions where
[38], [41]. in worst case it must be assumed that the voltage becomes
The composition and use of different power electronics zero until the faults are fixed or isolated. Examples of such
to make a general PEBB will affect the shipboard power malfunctions and failures may be short circuits and high inrush
system in many ways. The transition from early solutions using currents while starting large motors. The allowed voltage drop
Line Commutated Converters (LCC) and Cyclo-converters to is dependent on the vessel and its operations and is set by
today’s PWM Voltage Source Converters (VSC) had many classification entities [44].
advantages, including lower harmonic pollution, four-quadrant In DP-operations (e.g. DP2 [45]) with closed bus-ties,
operation and converter reversibility [42]. It is also expected assessments regarding voltage dip ride through capability must
that the introduction of the PEBB will lead to an increased be conducted as part of FMEA to assure continued opera-
power quality: The PEBB can be designed and controlled to tion after faults or malfunctions occur. Many DP-operations
achieve redundant and reliable solutions, with fewer building (station-keeping operations) are performed with open bus-
blocks, which minimize losses and keep the power quality tie, splitting the power system in two, thus minimizing the
higher than what is achieved in today’s solutions. However, chances for a complete blackout. This is not an economical
power electronics in general are non-linear elements, with non- nor an efficient solution, as splitting the power system in
linear behavior, and are in most cases sources of harmonic two requires an increased number of online prime movers
pollution. In thyristor-based devices (which is often the case for power generation, and also requires multiple separated
in motor drives) the harmonic spectrum is not dependent power management systems (PMS). Harmonic mitigation is
on impedance, thus introduces characteristic harmonic pollu- therefore not only important for the power system’s efficiency,
tions relative to the devices’ different designs. In a 6-pulse but also for its stability and reliability. Harmonic mitigation
converter, the characteristic harmonics are of 5th, 7th, 11th, and power conditioning is a active research topic, and many
13th, etc. order, and in a 12-pulse converter, the characteristic active and passive filter solutions have been proposed [46].
harmonics are of 11th, 13th, 23rd, 25th, etc. order. In a Passive filters do not have the ability to change their tuned
voltage source converter (VSC), which is not based on thyris- frequency, and due to changes in power system configura-
tors, these characteristic harmonics do not occur, and motor tions (and changes in load profiles) as a result of different
drives consisting of VSCs instead of thyristor-based drives operational requirements and mission types, passive filters
may solve the problems with the characteristic harmonics. are not always a good solution for harmonic mitigation as
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION 2016 10

a change in the harmonic spectrum requires a re-tuning of them to the power system. Because of speed variations and al-
the filters. An active filter, on the other hand, has the ability lowed frequency fluctuations within a designed deadband, the
to mitigate any frequency spectrum, the only limitation being frequency in shipboard ac power systems cannot be assumed
the bandwidth of its controller, thus increasing flexibility for constant.
changes in the power system’s harmonic frequency spectrum. In addition to frequency fluctuations, the speed variations
Active filters have also a smaller footprint than passive filters, on the motor shafts will also increase wear and tear leading
which is a desired property in marine vessels. Active filters to higher maintenance costs. Controlling the prime movers to
are expensive devices, thus location of installation in a power track a constant speed greatly affects the power generation
system is important for maximum utilization (and mitiga- as an optimal increase or decrease in power generation is
tion) of the filter’s power rating. A conceptual method using related to starting and stopping prime movers in a stepwise
optimization (Model Predictive Control) to perform system (ac) power generation [51]. As the load demand must be met
level harmonic mitigation has also been proposed [47]–[50]. at all times this means that prime movers running at low non-
Active filters come in many forms, and can be part of e.g. a optimal loading conditions is often the case in shipboard ac
propulsion system’s motor drive, realized as controlled Active power systems. To increase the fuel efficiency related to the
Front End (AFE) converters or simply stand-alone devices. power demand, the prime mover loading could be increased
Harmonic mitigation (and power conditioning) is, as earlier and power stored to be used in situations where the power
mentioned, important for achieving an efficient and reliable demand surpasses the power generation. An example would
power system, and the harmonic pollution problem is also be to provide the difference between consumed and generated
expected to be an issue in future power systems, consisting of power while additional prime movers are being started and
even more non-linear components. As of today, there are no connected to the power system to meet an increasing power
classification entities that require real-time THD surveillance, demand.
which would be an important measure for detecting potential In dc power systems, where the power distribution is
stability issues as well as performing fuel efficient operations. conducted on dc grids, the prime movers may run at vary-
THD requirements are checked by classification entities during ing speeds to meet the power demand. As in ac systems,
the vessel’s commissioning and certification using handheld the voltage level is maintained by controlling the generators
measuring devices. The future power system, where relia- excitation fields. Due to the flexibility of being able to change
bility and efficiency are cultivated, may require real-time the prime movers’ speeds, the power generation will adopt a
THD surveillance and power conditioning devices (possible more stepless behaviour than in ac systems. However, prime
consisting of PEBBs), which may be backed on optimization movers running outside their optimal speed ranges are prone
for system level harmonic mitigation, to comply with stringent to wear and tear, and especially at low speeds the combustion
air pollution regulations, as well as achieving higher reliability is not optimal and will increase sooting of the prime movers,
in terms of blackout-prevention due to increased power quality. thus increasing maintenance costs. At high speeds the fuel
consumption is not in line with the produced power (non-linear
relationship between fuel consumption and produced power),
E. Energy Management Systems (EMS) and Energy Storage thus reducing the fuel efficiency which leads to increased fuel
Systems (ESS) costs and emissions. As with ac power system, the dc power
Planning power generation, energy management, is impor- system could also benefit from a ESS that facilitates optimal
tant for achieving an economical and efficient power genera- operation of the prime movers.
tion with optimal prime mover loading conditions, thus keep- 1) ESS Applications: Many suitable ESS technologies
ing the fuel consumption at a minimum. In ac power systems, that facilitate a more economical and redundant operation
the prime movers are speed-controlled, mostly connected to in a marine vessel are available on the market today. The
fixed speed generators, to maintain a desired (and designed) choice of ESS technology is related to area of application,
frequency within allowable variations (deadband). As the energy density, size, weight, and cost, expected lifetime,
prime movers’ speeds are more or less fixed due to frequency charge/discharge rates, and other functional requirements.
control, the loading of each prime mover determines the fuel Examples of ESS technologies are Battery Energy Storage
efficiency in terms of amount of fuel per delivered amount System (BESS), Compressed Air Energy Storage (CAES), fly-
of useful energy - Specific Fuel Oil Consumption (SFOC) wheels, Superconducting Magnetic Energy Storage (SMES),
g
kW h . The prime movers often experience speed deviations capacitors (including ultra-capacitors) and Pumped Hydro
as an effect of dynamically changing load profiles (active and Storage (PHS) [52]. Depending on the power system (ac or
reactive power demand), in which affects the inertia on the dc) most ESS technologies need power conversion devices that
shafts between the prime movers and the generators. If such convert the power from and to the power system for charging
prime mover speed variations result in frequency fluctuations and discharging purposes. An obvious application of a ESS
exceeding the allowed deadband, the prime mover needs to be would be to serve as a backup power source similar to an
isolated and shut down. Large negative frequency fluctuations Uninterruptible Power Supply (UPS), which sets strict require-
can also be an indication of the running prime movers are ments to the ESS technology’s energy density and rate of
unable to meet the load demand, thus additional supervisory discharge. This type of application can be beneficial for many
steps should be taken to either shed non-essential loads or marine operations. An example would be an offshore vessel
spin up idle prime movers, and after synchronization connect conducting a DP-operation alongside an offshore structure that
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION 2016 11

experiences faults that cause power losses which may lead to

DŽƚŽƌ WƌŽƉ͘
a blackout. The ESS may in this case be crucial for powering ƌŝǀĞ DŽƚŽƌ
&ƵĞů WƌŝŵĞ 'ĞŶĞƌĂ
the propulsion system for a short period of time to be able to DŽǀĞƌϭ ƚŽƌ
reposition the vessel at a safe distance away from the structure Ͳ
to get the time-window needed for isolating the faults and to ͘
͘ ^^
re-power the vessel. ͘
н
Many power consuming systems and equipment on a vessel &
&ƵĞů WƌŝŵĞ 'ĞŶĞƌĂ
do not have a flat load profile. Propulsion systems, while DŽǀĞƌŶ ƚŽƌ
conducting station-keeping, have a load profile which corre- Ƶdž͘ůŽĂĚƐ
lates with waves and ocean currents. Weapons systems aboard
a naval vessel may give a pulsed load profile at irregular DŽƚŽƌ WƌŽƉ͘
ƌŝǀĞ DŽƚŽƌ
time instants, which would be more or less impossible to
predict. Due to the vessel’s dynamic load profile, the energy (a) Battery Energy Storage System (BESS) connected to the main bus
management is not an easy task, and, as earlier mentioned, (switchboard) in an IPS configuration. An Active Front End (AFE) is
often more prime movers are running than are actually needed installed alongside the BESS as a solution for supervision and BESS
control purposes.
to be sure of serving the load demands. One application of
the ESS, which is a feature that is sought for in a shipboard
power system, is load shaving or more precisely peak shaving WƌŽƉ͘
W
[53]. By using the ESS to flatten the vessel’s total load DŽƚŽƌ
profile, energy management, in terms of starting and stopping & /ŶǀĞƌƚĞƌ
generators, would be easier, and fewer prime movers have to ^^
be on line to meet a potential high and instant power demand. н Ͳ
Under low non-ideal loading conditions the ESS charges,
(b) Battery Energy Storage System (BESS) as part of a motor drive for
and while the load demand exceeds the power generation propulsion systems [34]. Point of Common Coupling (PCC) refers to
capabilities the ESS discharges. Whether the power system e.g. the vessel’s main switchboard. An Active Front End (AFE) is part
is dc or ac, the prime movers can be run at optimal speed for of the depicted motor drive and supervises charging and discharging of
the BESS.
maximum fuel efficiency. This feature, peak shaving, may be
seen as one of the strongest arguments for installing a suitable Fig. 5. Simplified illustrations of different installations of Battery Energy
Storage Systems.
ESS in a shipboard power system, as peak shaving may result
in a lower fuel consumption (and emission) due to the need
for fewer running gensets.
Another interesting application of ESS, dependent on ESS power conversion devices, reducing weight, footprint, and
technology employed, is harmonic mitigation [34], [54]. De- costs [34]. For BESS the available battery technology also
pending on the ESS’ speed of discharge, it may be used to introduces challenges, as the battery packs are heavy (relative
suppress harmonic pollution. The ESS may also be used to to power capacity) and in many cases have a large footprint.
charge a dc capacitor in an active filter, which strengthens the Despite weight and volume, the BESS may allow removal
filter’s capabilities, and thus enables the filter to use active of one prime mover from a vessel, which justifies the use
power in harmonic mitigation. Also frequency control by use of BESS. Another issue is the battery packs’ lifetime. Rapid
of ESS has been proposed [55]. As an example, an ESS such as charging and discharging of battery generates a lot of heat,
BESS may be installed alongside an Active Front End (AFE) which can be seen as losses, and may have critical effect on
(Fig. 5), which is a realistic scenario if for instance the ESS is the battery’s life. Thus a possible realistic outcome is that
part of a motor drive. In this case, when the ESS is installed the battery pack dies before the BESS manages to pay back
alongside an AFE, the ESS could attain application flexibility, the installation costs by reduced fuel consumption. In some
thus being able to do harmonic mitigation, peak shaving, and applications ultracapacitors or fuel cells can switch places with
even act as a reactive power source or consumer to increase the the battery pack, giving the energy storage system different
power system’s voltage stability margins. Fig. 5 showcases two properties such as increased lifetime, charge/discharge speed,
different locations in the grid for installing a BESS. The BESS energy density relative to footprint and weight, etc. Also
may be installed alongside an AFE or other power electronic hybrid energy storage systems, including different energy
devices which supervise the BESS state of charge (SOC) and storage devices, may also be interesting possibilities, thus
state of health (SOH), and control charging and discharging increasing applications and system flexibility [56], [57].
dependent on the load demand and the BESS’ SOC. When moving towards all electric-battery powered vessels, a
Even though the advantages of ESS in shipboard power new emerging technology -the inductive charging technology-
systems are many, it doesn’t come without challenges. Many has attracted the attention of the marine vessel community
of the available and suitable ESS technologies are expensive and the old concept of Inductive Power Transfer has re-
solutions, and are dependent on power conversion devices emerged for contactless battery charging of marine vessels
relative to ac or dc power systems. An effective solution, which [58]–[60]. Significant progress toward the development of
was illustrated in Fig. 5b, is to install an ESS, such as BESS, commercial solutions for wireless charging is already on its
as part of motor drives, thus eliminating the need for additional way for high power wireless transfer in the MW range [61].
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION 2016 12

This technology will greatly benefit coastal vessels operating close the bus-ties in all operational scenarios would be a
with a tight schedule as it will significantly reduce charging necessity for future power systems with increased efficiency
time and improve reliability. This will bring unavoidable and stringent emission requirements. To achieve this, the
challenges to the local power grid from which high power power systems must be equipped with stability-improving
will be tapped in a short time to charge the vessel’s battery systems and devices that, in a safe way, handle faults without
packs. This impending impact on the local electrical grid will harming the rest of the power system. Such systems may
require grid reinforcements and new solutions that will require involve harmonic mitigation, reactive power control, voltage
collaborative efforts between the utility and marine vessel and frequency control, peak (load) shaving, UPS systems
sectors. and advanced power system segregation and fault-isolation
2) Standards and Guidelines: Many classification entities systems. In order to take advantage of new technological
and interest groups impose strict regulations and set forth developments to increase operational flexibility without in-
guidelines for redundancy for many types of marine vessels creasing risk, DNV-GL recently introduced the DYNPOS-ER
to avoid total loss of maneuverability. This is mostly the case (Enhanced Reliability) notation for DP class 2 and 3 vessels.
for offshore vessels, like PSVs, but the requirements can also 3) Emission Free Operation: In tomorrow’s shipboard
be found for passenger vessels and cargo vessels transporting power systems the BESS (or another suitable ESS) may be
hazardous materials. The International Marine Contractors essential to cultivate reliable and efficient power systems (both
Association (IMCA) [62] states (for an offshore vessel) that if ac and dc), and applications such as harmonic mitigation, peak
there is a realistic chance of the bus-ties not opening or not shaving, reactive power control (voltage stability), voltage and
opening fast enough then the switchboard should be split for frequency control, and backup power can simply be different
the work (two-split in Fig. 3), and if so the power system algorithms deployed to a PEBB-based ESS. It is also expected
must include an independent power system (Power/Energy that in the near future harbors may require an emission-free
Management System - PMS/EMS) for each individual split approach for vessels to load and unload, thus an ESS may
[63]. Furthermore IMCA states that for a diesel-electric vessel be part of a larger green system keeping the air pollution
a task appropriate mode could mean operating with closed (emission) in harbors at a minimum. In addition, the EMS must
bus-ties, whereas a critical activity mode of operation may be intuitive and easy to understand, and provide supportive
require open bus-tie configuration [64]. These guidelines are and advisory actions which are trusted by the operators. Many
based on risk assessments (Failure Modes and Effect Analysis EMS systems today are hard to understand, as a result they
- FMEA) and fault tolerance (isolation of faults) dependent on are disregarded by the operators and kept out of the control
classification and control system redundancy [65]. DNV-GL loop with the effect being an inefficient power system. A
(earlier DNV) [66] describes that the traditional interpretation lot of work remains to map the operators’ behaviors and
of the DP-3 requirements has been to run the power system as interaction with the system to make an optimal, reliable, and
separated (segregated) subsystems with open bus-tie breakers. trustworthy interaction for efficient and economical control of
This is backed on IMO [67] MSC/Circ.645 guidelines for the shipboard power systems.
vessels with dynamic positioning systems, which states that
for equipment of class 3 the power system should be divisible
into two or more systems such that in the event of failure of F. Increasing Need for Measurements, Big-Data, and Software
one system, at least one other system will remain in operation Complexity
[68]. However, closed bus-tie DP operations have economical, To achieve a reliable and efficient shipboard power system,
technical, operational, and environmental benefits, thus some many different measurements are needed. Active power mea-
DP operators run the power system with closed bus-ties for surements (voltage and current measurements) are important
as large periods of operation as possible [45], [69]. ABS for the EMS to be able to meet the load demand, and an ESS
[70] also refers to the IMO MSC/Circ.645 guidelines, and needs power measurements for conducting peak shaving. In
states that these guidelines should be followed in the design ac distribution systems reactive power measurements (voltage
of DPS-2 (DPS - Dynamic Positioning System) and DPS-3 and current measurements) are important for voltage stability
systems where loss of position is not allowed to occur in assessments, and give a measure of the system’s efficiency.
the event of a single fault [71]. For ships normally operating Frequency measurements are needed in ac distribution systems
in transit, such as tankers and cargo ships, the equivalent as feedback to the prime movers’ speed controllers. Voltage
concept is redundant propulsion as described in e.g. DNV- measurements are needed for controlling the generators’ exci-
GL’s class notation RP. In short, all these regulations and tation fields, which are done by Automatic Voltage Regulators
guidelines state that, dependent on the vessel’s classification, (AVR), and also for transformers and power converters con-
one should not loose maneuverability, and due to the fact necting equipment and subsystems (including energy storage
that it has been difficult to both engineer completely fail-safe systems) to the power system need voltage measurements.
power systems and prove that there is no chance for power When starting a prime mover and connecting it to the grid in a
losses impairing the maneuverability, the trend has been to synchronization process both phase and voltage measurements
operate the power systems with open bus-ties (a split power are needed. Voltage measurements with high sampling fre-
system). This type of operation increases the number of needed quency are needed for harmonic mitigation, to assure voltage
online prime movers, which results in lower efficiency (higher quality within boundaries set by classification entities. These
fuel consumption) and increased emissions. To be able to are only a few examples of needed measurements.
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION 2016 13

Many parts of the power system have high real-time de- following:
mands (high sampling frequency demands) for measurements.
• External cyber-attacks can be classified as cyber-attacks
Harmonic mitigation using Active Power Filters (APF) and
originating remotely from the marine vessels. There are
power converters such as Active Front Ends (AFE) are ex-
different strategies for protecting the vessel from such
amples of systems that require (internal or external)a high
attacks. A vessel’s access point to the rest of the world
rate of sampling measurements. In addition, fast hardware
and potential remote systems, which normally is a 3 layer
and software is required to process the measurements in
switch, has authentication and VPN capabilities which
real-time to be able to utilize the information for control
provide basic security. The switch can also limit input and
purposes. Redundancy in measurement devices (sensors) is
output network ports, which restrict the communication
also a requirement for achieving a reliable system. If one
channel. By enabling only output ports, the vessel data
measurement device goes down another has to take over to
can be encrypted and exported to e.g. onshore fleet
keep the needed information to the system flowing. Redun-
management systems without allowing any input traffic
dancy in measurement devices comes in many forms, and a
from a potential cyber-attack. A practical approach is
common approach in systems that relies on correct information
described by DNV-GL [72], where the main access point
is to have a minimum of three measurement devices and use
to remote systems is to be powered on only when allowed
voting algorithms to assure the correctness of the measurement
by the vessel’s crew. Another form of attack is related
information.
to connection to other equipment or systems that are
Some measurements may be contaminated by noise, and
infected. An example of such a case might be the vessel’s
communication delays between taking the measurement and
shore power connection while docking, where the shore
sending it to the subscribing system may make the information
power is altered to harm the vessel’s power system and
no longer valid. Thus the use of filtering techniques for
put the vessel out of operation. Another example could be
removing noise, and estimators for estimating biases and
infection of onshore fleet management systems, or other
transport delays may in some cases be a necessary requirement
vessels within the same fleet that have dedicated ship-to-
for optimal control, giving the subscribing system correct and
ship communication equipment.
valid information. Advanced signal processing methods may
• Internal cyber-attacks can be classified as attacks orig-
also be used to detect and solve measurement drop-outs as
inating within the marine vessel. This could either be a
part of a solution to redundancy requirements for improving
passenger or trusted insider (crew) that gains control over,
system reliability.
or infects, one of the vessel’s distributed control system
With increasing system integrity that cultivates both effi-
nodes. The cyber-attacks could be based on maleware
ciency and reliability of the shipboard power system, there is
delivery by a USB stick or different internal access inter-
also an increasing need for measurements. The present trend
faces such as an Ethernet that connects the vessel’s office
shows that more and more devices and subsystems are given
network to the control system network. These types of
an IP-address and system information and measurements are
attacks are more difficult to handle, however procedures
broadcast on a local network in the vessel in a cloud-based
such as disabling unused potential access points (such as
architecture - The Industrial Internet of Things (IIoT). As a
USB connections) and limiting input and output ports
consequence the future system integrity may involve consumer
on the router that connects the office network to the
systems planning their power consumption, which is available
control system network can reduce the risk of internal
information for the EMS for use in power generation planning.
cyber-attacks. If one of the distributed control nodes gets
With the expected enormous amount of data as a conse-
infected it is important to isolate that controller from
quence of an increase in measurement devices and broadcast-
the rest of the system. However, to quickly realize and
ing of system information to get more efficient and reliable
identify the attack before any harm is done might be
control, problems such as limited network throughput and data
a challenge, which puts stringent requirements on the
processing resources may appear. Maybe the most frightening
vessel’s distributed control system’s middleware to limit
issue is that when all the vessel’s systems ”come online”,
potential attacks [73]. Such requirements can be based
the vessel is vulnerable to cyber-attacks. Even though an
on each control node’s accessibility and level of security
increase in available system information, measurement data,
clearance to distribute control actions to the rest of the
and distributed control may be beneficial for controlling the
vessel’s control nodes. If for instance the middleware
vessel’s power system in an optimal, reliable, and efficient
detects that one of the control nodes tries to control parts
way, the development of the future shipboard power systems
of the system outside the controller’s security clearance,
have to address the Big-Data challenge in the design of its
e.g. the vessel’s prime movers or propulsors, it might be
architecture and assure cyber-security. There exists a range
considered as an attack, which should trigger isolation
of different types of cyber-attacks, some of which are based
procedures and alert the crew. In addition, it is essential
on gaining access to data and information, and others that
to keep operation systems and firmware up to date to be
are disruptive and intended to take over or break down a
more resistant to cyber-attacks.
system. The latter may have catastrophic consequences if they
enable the attacks that gain control over the vessel’s power There is a drive towards increased fuel optimality, reduction
and propulsion system. A small selection of potential external of emissions, increased safety, and performance and oper-
and internal cyber-attacks will be treated separately in the ational flexibility. The technologies that are supporting this
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION 2016 14

development tend to increase system complexity, which has for the future by stimulating the discussion and identifying
consequences for ship designers, ship builders, ship owners, synergies between the modern power grid and the electrical
crew and other stakeholders such as classification societies and grid of the marine vessels today.
authorities. Like the automotive and aerospace industries, the
electric power plant is a highly computer controlled system ACKNOWLEDGMENT
with advanced functionality offering endless user configura-
tions and options embedded in software. The control of the This work has been carried out at the Centre for Au-
power plant itself is also integrated with the control of power tonomous Marine Operations and Systems (AMOS) whose
consumers, e.g. [74]. This leads to more complex processes main sponsor is The Research Council of Norway. The work
with new tasks, skills and training required by the crew. was supported by Ulstein Power & Control AS and The
Due to the safety-critical nature of the ship’s power plant Research Council of Norway, Project number 241205.
and electric system, the maritime industry is looking to learn
from the automotive, aerospace, and defense industries that R EFERENCES
have experienced the paradigm shift due the huge impact of [1] L. Horne, Electric propulsion of ships. North East Coast Institution of
information and communication technology. This has led to Engineers and Shipbuilders, 1939.
new standards, certification, and classification schemes related [2] E. Skjong, E. Rødskar, M. Molinas, T. Johansen, and J. Cunningham,
“The Marine Vessel’s Electrical Power System: From its Birth to Present
to integrated systems development and more extensive use of Day,” Proceedings of the IEEE, vol. 103, no. 12, pp. 2410–2424, Dec
simulator-based training and verification technologies, [75], 2015.
[76]. Future visions for unmanned and autonomous shipping, [3] C. Sulzberger, “First Edison Lights at Sea: The SS Columbia Story,
1880-1907 [history],” Power and Energy Magazine, IEEE, vol. 13, no. 1,
[77], [78], are indicators of the opportunities and challenges pp. 92–101, Jan 2015.
that are emerging. [4] Siemens, “Electrical Propulsion Systems,” http://w3.siemens.no/home/
no/no/sector/industry/marine/pages/electrical propulsion systems3.
aspx, accessed: 2015-12-22.
V. C ONCLUSION [5] A. A. Halacsy and G. H. von Fuchs, “Transformer invented 75 years
ago,” Electrical Engineering, vol. 80, no. 6, pp. 404–407, June 1961.
Past, present, and future challenges in the electrification [6] J. J. Cunningham, “Manhattan Electric Power Distribution, 1881-1901,”
Proceedings of the IEEE, vol. 103, no. 5, pp. 850–858, 2015.
of the marine vessel have been discussed in this paper. The [7] H. F. Harvey Jr and W. Thau, “Electric propulsion of ships,” Trans-
milestones in the evolution of the development of marine actions of the American Institute of Electrical Engineers, vol. 44, pp.
vessels, from the earliest introduction of electricity in com- 497–522, 1925.
[8] T. McCoy, “Electric Ships Past, Present, and Future [Technology Lead-
mercial vessels with the SS Columbia in 1880 to the new ers],” IEEE Electrification Magazine, vol. 3, no. 2, pp. 4–11, June 2015.
era of the all electric vessels marked by the Ampere ferry, [9] W. McBride, Technological Change and the United States Navy, 1865–
have been highlighted in the historical part of the paper. The 1945, ser. Johns Hopkins Studies in the History of Technology. Johns
Hopkins University Press, 2000.
use of electricity in marine vessels which started far from [10] D. Dyal, B. Carpenter, and M. Thomas, Historical Dictionary of the
the idea of an electric power system on board has, however, Spanish American War, ser. Gale virtual reference library. Greenwood
spurred the development of electric propulsion systems, and Press, 1996.
[11] E. Smith, A short history of naval and marine engineering. Cambridge
also the concept of the integrated power system. The electrical University Press; Reissue edition (October 31, 2013), 1938.
system of today’s marine vessels can be compared to a land- [12] P. Fontenoy, Submarines: An Illustrated History of Their Impact, ser.
based stand-alone microgrid system, with which the marine Weapons and warfare series. ABC-CLIO, 2007.
[13] S. Sandler, World War II in the Pacific: An Encyclopedia, ser. Garland
vessel power system shares many common features. Present Military History of the United States. Garland Pub., 2001.
and future challenges include issues such as harmonics, power [14] P. S. Dawson, British Superliners of the Sixties: A Design Appreciation
quality, fault handling, and stability. These issues will be as of the Oriana, Canberra and QE2. London: Conway Martime Press,
1990.
relevant during normal operation of the marine vessel as they
[15] ——, Canberra: In the Wake of a Legend. Conway Maritime Press,
are at commissioning today. Many of the features required 1997.
today to handle the modern land-based electrical system (smart [16] N. Polmar and K. Moore, Cold War Submarines: The Design and
grid) will be a necessity in marine vessels as the use of elec- Construction of U.S. and Soviet Submarines. Brassey’s, 2004.
[17] W. Donnelly, Nuclear Power and Merchant Shipping, ser. Nuclear Power
tricity becomes more intensive. Characterization of the marine and Merchant Shipping. U.S. Atomic Energy Commission, Division
vessel electrical grid through real-time measurements, and the of Technical Information, 1964, no. v. 7.
monitoring of fundamental parameters such as impedance in [18] “NAVSEA Design Practices and Criteria Manual, Electrical Systems for
Surface Ships, Chapter 300,” Naval Sea Systems Command (NAVSEA),
addition to fundamental and harmonic currents and voltages, NAVSEA T9300-AF-PRO-020.
will be essential to ensure the safety, integrity, and stability of [19] N. H. Doerry and D. H. Clayton, “Shipboard electrical power quality of
the marine vessel power system. Lately, re-emerging wireless service,” in Electric Ship Technologies Symposium, 2005 IEEE. IEEE,
2005, pp. 274–279.
power transfer for battery pack charging in vessels will make [20] N. Doerry, “Naval Power Systems: Integrated power systems for the con-
the link between the land-based power grid and the marine tinuity of the electrical power supply.” IEEE Electrification Magazine,
vessel power grid even tighter and will create a new form of vol. 3, no. 2, pp. 12–21, June 2015.
[21] “International Maritime Organization: Prevention of Air Pollution
interaction. Ultimately, as the use of all electric ships becomes from Ships,” http://www.imo.org/en/OurWork/Environment/
widespread, the electric vessel will become a part of the PollutionPrevention/AirPollution/Pages/Air-Pollution.aspx, accessed:
land-based power grid as a high impact electric load, thus 2015-08-21.
[22] M. Patel, Shipboard Electrical Power Systems, ser. Shipboard Electrical
bringing new challenges. This paper aims at anticipating the Power Systems. Taylor & Francis, 2011.
potential new challenges and the associated research needs [23] M. Hanna, “Voltage transformation.” May 14 1912, US Patent 1,026,391.
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION 2016 15

[24] V. Staudt, R. Bartelt, and C. Heising, “Fault scenarios in dc ship grids: optimization,” in 2015 IEEE Transportation Electrification Conference
The advantages and disadvantages of modular multilevel converters.” and Expo (ITEC). IEEE, 2015, pp. 1–8.
IEEE Electrification Magazine, vol. 3, no. 2, pp. 40–48, 2015. [48] E. Skjong, M. Molinas, and T. A. Johansen, “Optimized current ref-
[25] M. Baran and N. Mahajan, “Dc distribution for industrial systems: erence generation for system-level harmonic mitigation in a diesel-
opportunities and challenges,” IEEE Trans. Ind. Appl., vol. 39, no. 6, electric ship using non-linear model predictive control,” in 2015 IEEE
pp. 1596–1601, Nov 2003. International Conference on Industrial Technology (ICIT). IEEE
[26] G. Reed, B. Grainger, A. Sparacino, and Z.-H. Mao, “Ship to grid: Conference Publications, 2015, pp. 2314–2321.
Medium-voltage dc concepts in theory and practice,” IEEE Power and [49] E. Skjong, M. Molinas, T. A. Johansen, and R. Volden, “Shaping the
Energy Magazine, vol. 10, no. 6, pp. 70–79, Nov 2012. current waveform of an active filter for optimized system level harmonic
[27] R. E. Hebner, F. M. Uriarte, A. Kwasinski, A. L. Gattozzi, H. B. conditioning,” in Proceedings of the 1st International Conference on
Estes, A. Anwar, P. Cairoli, R. A. Dougal, X. Feng, H.-M. Chou, L. J. Vehicle Technology and Intelligent Transport Systems, 2015, pp. 98–
Thomas, M. Pipattanasomporn, S. Rahman, F. Katiraei, M. Steurer, 106.
M. O. Faruque, M. A. Rios, G. A. Ramos, M. J. Mousavi, and T. J. [50] E. Skjong, J. A. Suul, A. Rygg, M. Molinas, and T. A. Johansen,
McCoy, “Technical cross-fertilization between terrestrial microgrids and “System-wide harmonic mitigation in a diesel electric ship by model
ship power systems,” Journal of Modern Power Systems and Clean predictive control,” IEEE Trans. Ind. Electron., 2016, manuscript ac-
Energy, pp. 1–19, 2015. cepted for publication.
[28] A. Vicenzutti, D. Bosich, G. Giadrossi, and G. Sulligoi, “The role of [51] K. Davey, “Ship component in hull optimization,” Marine Technology
voltage controls in modern all-electric ships: Toward the all electric Society Journal, vol. 39, no. 2, pp. 39–46, 2005.
ship.” IEEE Electrification Magazine, vol. 3, no. 2, pp. 49–65, 2015. [52] Z. Zhou, M. Benbouzid, J. F. Charpentier, F. Scuiller, and T. Tang,
[29] S.-J. Ahn, J.-W. Park, I.-Y. Chung, S.-I. Moon, S.-H. Kang, and S.- “A review of energy storage technologies for marine current energy
R. Nam, “Power-sharing method of multiple distributed generators systems,” Renewable and Sustainable Energy Reviews, vol. 18, pp. 390–
considering control modes and configurations of a microgrid,” IEEE 400, 2013.
Trans. Power Del., vol. 25, no. 3, pp. 2007–2016, 2010. [53] A. Oudalov, R. Cherkaoui, and A. Beguin, “Sizing and optimal operation
[30] R. Lasseter, “Microgrids,” in Power Engineering Society Winter Meeting, of battery energy storage system for peak shaving application,” in IEEE
2002. IEEE, vol. 1, 2002, pp. 305–308 vol.1. Lausanne Power Tech, 2007, July 2007, pp. 621–625.
[31] T. Vandoorn, B. Renders, L. Degroote, B. Meersman, and L. Vandevelde, [54] K. Divya and J. stergaard, “Battery energy storage technology for power
“Active load control in islanded microgrids based on the grid voltage,” systemsan overview,” Electric Power Systems Research, vol. 79, no. 4,
IEEE Trans. Smart Grid, vol. 2, no. 1, pp. 139–151, March 2011. pp. 511 – 520, 2009.
[32] J. Lopes, C. Moreira, and A. Madureira, “Defining control strategies for [55] A. Oudalov, D. Chartouni, and C. Ohler, “Optimizing a battery energy
microgrids islanded operation,” IEEE Trans. Power Syst., vol. 21, no. 2, storage system for primary frequency control,” IEEE Trans. Power Syst.,
pp. 916–924, 2006. vol. 22, no. 3, pp. 1259–1266, Aug 2007.
[33] B. Kroposki, R. Lasseter, T. Ise, S. Morozumi, S. Papatlianassiou, and [56] A. Khaligh and Z. Li, “Battery, ultracapacitor, fuel cell, and hybrid
N. Hatziargyriou, “Making microgrids work,” IEEE Power and Energy energy storage systems for electric, hybrid electric, fuel cell, and plug-
Magazine, vol. 6, no. 3, pp. 40–53, 2008. in hybrid electric vehicles: State of the art,” IEEE Trans. Veh. Technol.,
[34] S. Kim, S. Choe, S. Ko, and S. Sul, “A naval integrated power system vol. 59, no. 6, pp. 2806–2814, July 2010.
with a battery energy storage system: Fuel efficiency, reliability, and [57] P. Ribeiro, B. Johnson, M. Crow, A. Arsoy, and Y. Liu, “Energy storage
quality of power.” IEEE Electrification Magazine, vol. 3, no. 2, pp. 22– systems for advanced power applications,” Proceedings of the IEEE,
33, June 2015. vol. 89, no. 12, pp. 1744–1756, Dec 2001.
[58] G. Maggetto and P. Van den Bossche, “Inductive Automatic Charging–
[35] H. Ibrahim, A. Ilinca, and J. Perron, “Energy storage systems charac-
The Way to Safe, Efficient and User-Friendly Electric Vehicle Infras-
teristics and comparisons,” Renewable and sustainable energy reviews,
tructure,” in Electric Vehicle Symposium EVS-18, 2001, pp. 20–24.
vol. 12, no. 5, pp. 1221–1250, 2008.
[59] G. Covic, J. T. Boys et al., “Modern trends in inductive power transfer
[36] R. Hebner, K. Davey, J. Herbst, D. Hall, J. Hahne, D. Surls, and
for transportation applications,” IEEE Journal of Emerging and Selected
A. Ouroua, “Dynamic load and storage integration,” Proceedings of the
Topics in Power Electronics, vol. 1, no. 1, pp. 28–41, 2013.
IEEE, vol. 103, no. 12, pp. 2344–2354, Dec 2015.
[60] A. Brecher and D. Arthur, “Review and Evaluation of Wireless Power
[37] E. Zivi, “Design of robust shipboard power automation systems,” Annual Transfer (WPT) for Electric Transit Applications,” Tech. Rep., 2014.
Reviews in Control, vol. 29, no. 2, pp. 261 – 272, 2005. [61] G. Guidi and J. A. Suul, “Minimization of converter ratings for mw-scale
[38] T. Ericsen, N. Hingorani, and Y. Khersonsky, “Power electronics and inductive charger operated under widely variable coupling conditions,” in
future marine electrical systems,” in Petroleum and Chemical Industry 2015 IEEE PELS Workshop on Emerging Technologies: Wireless Power
Technical Conference, 2004. Fifty-First Annual Conference 2004, Sept (WoW). IEEE, 2015, pp. 1–7.
2004, pp. 163–171. [62] IMCA, “http://www.imca-int.com/.”
[39] T. Ericsen and A. Tucker, “Power electronics building blocks and [63] ——, “Guidelines for The Design and Operation of Dynamically Po-
potential power modulator applications,” in Conference Record of the sitioned Vessels,” http://www.imca-int.com, December 2007, IMCA M
1998 Twenty-Third International Power Modulator Symposium, 1998, 103 Rev. 1.
Jun 1998, pp. 12–15. [64] ——, “International Guidelines for The Safe Operation of Dynamically
[40] T. Ericsen, “Power electronic building blocks - a systematic approach Positioned Offshore Supply Vessels,” http://www.imca-int.com, April
to power electronics,” in IEEE Power Engineering Society Summer 2015, IMCA 182 MSF Rev. 2.
Meeting, 2000, vol. 2, 2000, pp. 1216–1218 vol. 2. [65] ——, “A Guide to DP Electrical Power and Control Systems,” http:
[41] ——, “Future navy application of wide bandgap power semiconductor //www.imca-int.com, November 2010, IMCA M 206.
devices,” Proceedings of the IEEE, vol. 90, no. 6, pp. 1077–1082, Jun [66] DNV-GL, “https://www.dnvgl.com/.”
2002. [67] IMO, “http://www.imo.org/.”
[42] M. Lehti, P. Hyvarinen, and T. Tissari, “The new generation of propul- [68] ——, “Guidelines for Vessels with Dynamic Positioning Systems,” http:
sion drive systems,” in All Electric Ship 2003 Conference, 2003, pp. //www.imo.org, June 1994, MSC/Circ.645.
13–14. [69] DNV, “Failure Mode and Effect Analysis (FMEA) of Redundant Sys-
[43] T. Hoevenaars, I. Evans, and A. Lawson, “New marine harmonic tems,” http://www.dnv.com, January 2012, DNV-RP-D102.
standards,” vol. 16, no. 1, January 2010, pp. 16–25. [70] ABS, “http://ww2.eagle.org/.”
[44] M. Bollen, M. Stephens, S. Djokic, K. Stockman, B. Brumsickle, [71] ——, “Guide for Dynamic Positioning Systems,” http://ww2.eagle.org/,
J. Milanovic, J. R. Gordón, R. Neumann, G. Ethier, F. Corcoles et al., November 2013.
“Voltage dip immunity of equipment and installations,” Prepared by the [72] DNV-GL, “Recommended Practice: Safety, operation and performance
members of CIGRE/CIRED/UIE Joint Working Group C, vol. 4, 2010. of grid-connected energy storage systems,” https://www.dnvgl.com/,
[45] DNV, “Dynamic Positioning Systems,” http://www.dnv.com, July 2014, December 2015, DNVGL-RP-0043.
Rules for Classification of Ships, Part 6, Chapter 7. [73] C. Esposito and M. Ciampi, “On security in publish/subscribe services:
[46] H. Akagi, E. Watanabe, and M. Aredes, Instantaneous Power Theory and A survey,” Communications Surveys & Tutorials, IEEE, vol. 17, no. 2,
Applications to Power Conditioning, ser. IEEE Press Series on Power pp. 966–997, 2015.
Engineering. Wiley, 2007. [74] E. Mathiesen, B. Realfsen, and M. Breivik, “Methods for reducing
[47] E. Skjong, M. Ochoa-Gimenez, M. Molinas, and T. A. Johansen, frequency and voltage variations on DP vessels,” in MTS Dynamic
“Management of harmonic propagation in a marine vessel by use of Positioning Conference, Houston, TX, 2012.
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION 2016 16

[75] T. A. Johansen and A. J. Sørensen, “Experiences with HIL simulator Marta Molinas (M’94) received the Master of
testing of power management systems,” in MTS Dynamic Positioning Engineering degree from Ryukyu University, Japan,
Conference, Houston, TX, 2009. in 1997, and the Doctor of Engineering degree from
[76] T. I. Bø, A. R. Dahl, T. A. Johansen, E. Mathiesen, M. R. Miyazaki, the Tokyo Institute of Technology, Tokyo, Japan, in
E. Pedersen, B. Rokseth, R. Skjetne, A. J. Sørensen, L. Thorat, I. B. 2000. She was a Guest Researcher with the Uni-
Utne, and K. K. Yum, “Real-time marine vessel and power plant versity of Padova, Padova, Italy, during 1998. From
simulator,” in ASME 34th International Conference on Ocean, Offshore 2008-2014 she has been professor at the Department
and Arctic Engineering (OMAE2015) in St. John’s, Newfoundland, of Electric Power Engineering at the Norwegian Uni-
Canada, 2015. versity of Science and Technology (NTNU). From
[77] Rolls-Royce-Marine, “Rolls-royce drone ships challenge $375 billion 2008 to 2009, she was a Japan Society for the
industry: Freight,” http://www.bloomberg.com/news/articles/2014- Promotion of Science (JSPS) Research Fellow with
02-25/rolls-royce-drone-ships-challenge-375-billion-industry-freight, the Energy Technology Research Institute, National Institute of Advanced
2014, accessed: 2015-09-26. Industrial Science and Technology, Tsukuba, Japan. In 2014, she was Visiting
[78] DNV-GL, “The revolt – a new inspirational ship concept,” https: Professor at Columbia University and Invited Fellow by the Kingdom of
//www.dnvgl.com/technology-innovation/revolt/index.html, 2015, ac- Bhutan working with renewable energy microgrids for developing regions.
cessed: 2015-09-26. She is currently Professor at the Department of Engineering Cybernetics,
NTNU. Her research interests include stability of power electronics systems,
harmonics, oscillatory phenomena, and non-stationary signals from the human
and the machine. Dr. Molinas has been an AdCom Member of the IEEE
Power Electronics Society. She is Associate Editor and Reviewer for IEEE
Transactions on Power Electronics and PELS Letters.

Espen Skjong received his MSc degree in Engi-

neering Cybernetics at NTNU in 2014, specializing
in model predictive control (MPC) for autonomous Tor Arne Johansen (M’98, SM’01) received the
control of UAVs. He is currently employed in Ulstein MSc degree in 1989 and the PhD degree in 1994,
Power & Control AS as an industrial PhD candidate. both in electrical and computer engineering, from
His research topic is optimization within marine the Norwegian University of Science and Technol-
vessel power systems. His industrial PhD fellowship ogy (NTNU), Trondheim, Norway. From 1995 to
is with the Center of Excellence on Autonomous 1997, he worked at SINTEF ICT as a researcher
Marine Operations and Systems (AMOS) at NTNU. before he was appointed Associated Professor at
NTNU in Trondheim in 1997 and Professor in 2001.
He has published several hundred articles in the
areas of control, estimation and optimization with
applications in the marine, automotive, biomedical
and process industries. In 2002 Johansen co-founded the company Marine
Cybernetics AS where he was Vice President until 2008. Prof. Johansen
received the 2006 Arch T. Colwell Merit Award of the SAE, and is currently a
principal researcher within the Center of Excellence on Autonomous Marine
Operations and Systems (AMOS) and director of the Unmanned Aerial
Rune Volden received his BSc Hons. deg. at UMIST Vehicle Laboratory at NTNU.
(University of Manchester) in 1989. He attended
Junior Officer School at Royal Norwegian Marine
1991 (Submarine). He finished his PhD at Norwe-
gian University of Science and Technology in 1995.
He was Associate Professor at Østfold Engineering Joseph Cunningham received the B.S. degree
College in 1994 - 1995. He has 20 years of industrial in physics from St. Francis College of Brooklyn
research and development experience in automatic Heights, Brooklyn, NY, USA. His interest in electric
control within production machines and maritime power systems dates to his youth when his high
applications, e.g. dynamic positioning, winch control school science project, ”The Theory and Operation
and automation. He was chairman and member of of Alternating Current,” was awarded a first-place
the board for Norwegian Society of Automatic Control (NMO of IFAC) 2009- gold medal. This success led to a scholarship and
2014. He is currently R&D manager at Ulstein Power & Control AS and a degree in physics from St. Francis College of
Associate Professor at NTNU Campus Ålesund. Brooklyn Heights, Brooklyn, NY, USA. He has
researched and authored numerous booklets, articles,
and books on topics such as industrial electrification,
electric utility power systems, and electric rail transportation. His latest book,
New York Power, was published in 2013 by the IEEE History Center Press.
In 1976, he coauthored a definitive three-volume History of the New York
City Subway System. He has also lectured and taught widely on the history
of electrotechnology and has consulted on numerous history projects and
Egil Rødskar received his BSc degree in Elec- television productions.
trical Power Systems from Møre og Romsdal In-
geniør Høgskole (MRIH) in 1987. He is currently
a Technical Manager within Systems with Ulstein
Power & Control AS, Ulsteinvik, Norway. He has
25 years of experience in marine electrical system
engineering and project management for installation
and commissioning of marine electrical systems.
His experience also covers failure mode and effect
analysis (FMEA), failure mode, effect and criticality
analysis (FMECA) for marine ship systems and
knowledge of ship operations.

View publication stats