Energies 2020, 13, x FOR PEER REVIEW 3 of 22

Energies 2020, 13, 3188 3 of 22

2. Marine Vessel DPS Description

2. Marine Vessel DPS Description
Energies 2020, 13, x FOR PEER REVIEW 3 of 22
2.1. Overview of DPS Applications
2.1. Overview
2. Marine
of DPS DPSApplications
For manyVessel Description
offshore activities, it is essential to keep the ship’s position and heading stable. The
For many offshore
DPS automatically activities,
controls it is essential
the position to keepofthe
and heading ship’s
a ship byposition and heading
using thrusters stable.
that are The DPS
persistently
2.1. Overview of DPS Applications
automatically
dynamic to face controls the position forces
the environmental and heading
inducedofby a ship by using
waves, currents,thrusters
and wind.that are persistently
Environmental
dynamicFor
disturbances many
to face
try to offshore
themove activities,
environmental
the itforces
vessel from is essential
theinduced to keep
desired by the ship’s
waves,
position position
currents,
while the DP and
and heading stable.
wind. Environmental
controller The
automatically
DPS
compensates automatically
disturbances for try to controls
move
those the
the vessel
forces position
and keeps and heading
from the desired of a
ship in positionship by using
a stable while
position thrusters
thebyDPusingthat are
controller persistently
automatically
its propellers and
dynamic
compensates
thrusters. to
forface
A vessel the forces
those
has environmental
6 degreesand of keeps forces
theinduced
autonomy ship inbya waves,
namely such currents,
stable position
as surging, and wind. its
by swaying,
using Environmental
propellers
yaw, heaving, and
disturbances
thrusters. try to move the vessel from the desired position while the DP controller automatically
rolling, andA vessel has
pitching, 6 degreesinofFigure
as presented autonomy namely such
1. Conversely, as surging,
the DPS is able toswaying,
control the yaw, heaving,
movement
compensates for those forces and keeps the ship in a stable position by using its propellers and
rolling,
of and pitching,
three horizontal assuch
axes presented in Figure
as surge, sway, 1. and Conversely, the DPS
yaw. To control is able
vessel to control
motion, the components,
the axis movement of
thrusters. A vessel has 6 degrees of autonomy namely such as surging, swaying, yaw, heaving,
three horizontal
including axesand
heave, roll, suchpitch,
as surge, sway, and
are ignored in theyaw. To control
estimator andvessel
controlmotion,
design the[49].axis components,
However, these
rolling, and pitching, as presented in Figure 1. Conversely, the DPS is able to control the movement
including
are measured
of threeheave, roll,movement
for the
horizontal and pitch,
axes such asofare ignored
the
surge, sway,in
navigation
and the estimator
and
yaw. antenna
To and
control control
system
vessel design
the[49].
purposes.
motion, However, these
axis components,
are measured
Figure 2 for
depicts the movement
a DPS of
consisting the ofnavigation
sub-systems and antenna
that are system
made purposes.
known
including heave, roll, and pitch, are ignored in the estimator and control design [49]. However, theseas follows:
are measured for the movement of the navigation and antenna system purposes.
Figure 2 depicts a DPS consisting
Z of sub-systems that are made known as follows:
Heave
Yaw
Z
Heave
Yaw

O
Pitch
O
Y Pitch

Y
Sway Roll
Sway Roll
Surge
Surge
X
ѱ X
ѱ

Figure 1. The ship movement components with six degrees of autonomy.

Figure 1. The ship movement components with six degrees of autonomy.

Figure 2 depictsFigure
a DPS 1. consisting
The ship movement components
of sub-systems withare
that six made
degreesknown
of autonomy.
as follows:

Sensors:
Sensors:
-Compass
-Compass (Heading)
(Heading)
-Wind
-Wind
-Vertical
-Vertical or Motion
or Motion
reference units
reference units
Wind Sensor Gyrocompass
Gyrocompass
-Draft
-Draft MRU
Wind Sensor
MRU
-SeaCurrent
-Sea Current

Satellite
Satellite

Operator Console:
Operator
(HumanConsole:
Machine Position Reference
(Human Machine
Interface) system:Position Reference
Interface)
-HMI Computer -GNSSsystem:
(Satellite system)
-Operator
-HMI panel
Computer -Hydro-GNSS
acoustic(Satellite system)
-Display screen
-Operator panel -Laser-Hydro acoustic
-Display screen -Microwave
-Laser
-Taut wire
-Microwave
-Taut wire

DP Controller:
-Process stations
DP Controller:
-I/O system
-Process stations
-I/O system

Main Propeller Power System:

-Power Generation
-Power distribution
Main Propeller Power System:
-Power management
-Power Generation
Backup Power (UPS)
-Power distribution
-Power management
Backup Power (UPS)
Azimuth Thruster Bow Thrusters

Azimuth Thruster Bow Thrusters

Figure
Figure 2. Componentsofofaageneric
2. Components generic dynamic
dynamic position
positionsystem
systemin in
marine vessels.
marine vessels.

Figure 2. Components of a generic dynamic position system in marine vessels.

Energies 2020, 13, 3188 4 of 22

2.1.1. Power Subsystem

The DP ship power subsystem consists of power plants, distribution and switchboards system,
transformers, electronic power units such as adjustable frequency drive (AFD), motor control system
(MCS), energy-storing devices (ESD) and observing and robotics system. They contain essential units
for providing efficient power for the DPS performance in various sea states.

2.1.2. Signal Processing Subsystem

The signal processing subsystem includes all the data sent by movement and weather sensors,
and satellite reference units. The measured signals pass systematically and analyze the signal processing
in a module separately to provide data for the station of the vessel in the operational condition. Then,
the DPS computes the essential thrusting force to compensate deviations of the ship movements in the
setpoint place.

2.1.3. Sensors Subsystem

The sensors subsystem contains:

• CPU Processor/joystick systems console

• Instruments sensor units, etc.
• Station reference unit consists of:

- Navigation system
- Acoustic systems
- Microwave and laser systems
- The interface control unit, which is used as an interface to read the thrusters, switchboard
feedback signals, as well as outputting command signals from the DP system to external units.
• Observing system and operator control panels.

2.1.4. Thruster Subsystem

In the main DPS components, the thruster subsystem consists of important mechanisms, such as
electronic drive units, principal propellers, bow, stern, and azimuth thrusters. They are controlled
by DPS to compute allocating thrust force and tracking the path. The research in this area proposes
that the high-level positioning controller can be applied to predict the required forces in the 6 degrees
of movement. Moreover, the thrust distribution system (TDS) which estimates the corresponding
thrust power to command the direction of each thrusting motors has proposed in [40–42]. Furthermore,
the effect of minimum control level of thrusting in calm to rough sea environments is emphasized
in [50] to prevent the corrosion of mechanical parts, avoid the shutdown, and harmonic distortion
in power delivery, respectively. The control framework of TDS and the overall methodology of DPS,
electronic units, and power are presented in Figure 3. Accordingly, the information of position and
speed movement of vessels measures by the sensors as we mentioned in Section 2.1.3. Consequently,
the measured data are transmitted to the DPS to compute the torque reference (τd) for low or high-level
of thrusting in different sea states, such as calm to the extreme, respectively. Then, the power plant
generates sufficient power and moment of force, where the vessel velocity reference is estimated by
TDS to carry the ship to the desired position. The TDS set of rules efforts to synchronize the thrusters
to match the computed torque τ with the reference. Hence, the TDS algorithm follows the reference
strictly with slight deflections from the setpoint to enhance the performance of power plant conditions.
For decreasing power consumption, TDS is essential to monitor the minimum power reference by
resultant in transient deviation from the vessel position. On the other hand, PMS usually needs
to confirm certain changes and inform the imminent fluctuations for the maximum existing power
and the present power consumption to TDS caused by the massive deviations from the heavy load
Energies 2020, 13, 3188 5 of 22

consumers such as winches, rotating equipment in drilling, and cable laying operation. Therefore,
the TDS controller computes the command of thruster power to rotational per minute (RPM) order to
drive speed regulation thrusters and feeds by generators. Furthermore, the level of thruster controller
Energies 2020, 13, x FOR PEER REVIEW 5 of 22
provides the prospect of predicting and restricting the deviances of signal errors in the vessel location
since
in thethe local
vessel thruster
location controllers
since the localdo not have
thruster any notice
controllers doabout the performance
not have of the
any notice about theother regional
performance
controllers. Therefore,
of the other regional they are not
controllers. able to calculate
Therefore, the able
they are not variations in thethe
to calculate resultant generalized
variations power.
in the resultant
Nevertheless, to counteract thrust load fluctuations, local thruster controllers have to be
generalized power. Nevertheless, to counteract thrust load fluctuations, local thruster controllers able to reduce
and
haveincrease
to be ablethetopower
reducedemand. Hence,
and increase the PMS
power is demand.
used to adjust
Hence, the available
PMS is usedpower signal
to adjust the in power
available
plants for the thruster controllers.
power signal in power plants for the thruster controllers.

Power management system

PESS
P-Gensets
Governor1 Governor2 Governor3 Governor4 Governor5 Governor6
Main
Generators
ESS fset1 fset2 fset3 fset4 fset5 fset6
8725 KVA
6980KW

AVR
G G G G
G G
Vset Vset Vset Vset
Vset AC Bus
CB
6.6 kv Bus

3600kVA
6.6kv/400v 3600kVA 6.6kv/400v

DB Service
loads Loads
1000 VDC Bus

Environment
disternabces
ƒ we
τd Wref Driver speed PWM
DP control system Thruster Allocation
regulation
I PWM Inverter v

PWM Inverter drive

Positions reference and sensors system

Figure 3.
Figure 3. The
The relationship
relationship schematic
schematic diagram
diagram between the PMS
between the PMS and
and DP
DP control
control system.
system.

2.1.5. Power Management Subsystem

The capability to execute and maneuvering of vessel positions relies on the sea state and the
capacity of the diesel generator power. Without doubt, inefficient power may lead to decrease DP
performance, failure of power, power, andand position,
position, which
which may
may lead
lead toto aa blackout.
blackout. The principal confusion
confusion
for the power system
system of of aa DP
DP ship
ship isisthat
thatthe
thepower
powerof ofthe
thethrusters
thrustersininrough
roughseaseastates
statesmay
maylead
leadtotoa
alarger
largerenergy
energyratio
ratiothan
thanthethepower
powerplantplantcapacity.
capacity. The
Theextreme
extreme DP DP demand
demand in in power units and
distribution in DP operational mode may potentially enhance the risk of a shutdown in all kind of
DP vessels. Consequently, the redundancy of power systems due to increasing the reliability of the
power
power system
systemby byusing
usingenergy
energystorage
storage systems
systems such as as
such thethe
battery, flywheel,
battery, ultra-capacitors,
flywheel, fuel fuel
ultra-capacitors, cell,
etc.,
cell, is costly
etc., as shown
is costly in Figure
as shown 3. Furthermore,
in Figure 3. Furthermore,cost-effective
cost-effectiveandandenergy efficiency
energy solutions
efficiency for
solutions
investing
for investingin tools, conservation,
in tools, harmful
conservation, gas emission,
harmful and fuel and
gas emission, consumption because of because
fuel consumption generatorsof
running hours are significantly demanded. For instance, the doubts in electrical
generators running hours are significantly demanded. For instance, the doubts in electrical power power demand of
drilling,
demand shuttle tankers,
of drilling, cable,tankers,
shuttle and pipecable,
layingandwhere
pipethelaying
importance
whereoftheDPSimportance
for transferring energy
of DPS for
from offshoreenergy
transferring wind farm
from by cable laying,
offshore wind farm oil resources by tankers
by cable laying, in operational
oil resources condition
by tankers are made
in operational
disturbances
condition arefor the PMS
made to provide
disturbances forthrusting
the PMSpower [36–41].
to provide In addition,
thrusting powerthe PMS beInable
[36–41]. to control
addition, the
PMS be able to control the optimum power flow be sure of DP demand in the future operational
condition. In this way, the DP load profile can be predicted by PMS based on effective optimization
algorithms [42–46]. As a result, the flexibility of power balance amongst power plant and the DP
command demand for producing sufficient energy resources by using advanced control techniques
for PMS is very crucial in the marine microgrids [47,48].
Energies 2020, 13, 3188 6 of 22

the optimum power flow be sure of DP demand in the future operational condition. In this way, the DP
load profile can be predicted by PMS based on effective optimization algorithms [42–46]. As a result,
the flexibility of power balance amongst power plant and the DP command demand for producing
sufficient energy resources by using advanced control techniques for PMS is very crucial in the marine
microgrids [47,48].

2.2. Dynamic Positioning Vessel Classification

On the authority of the international maritime organization (IMO) and global maritime standards
such as American Bureau of Shipping( ABS), Det Norske Veritas (DNV), Lloyds Register (LR), and so
on, DP ships are referred to classes (1–3) which are briefly described as follows [51–54]:

• DP class 1 is not redundant and can be positioned for a particular fault.

• DP class 2 is redundant to facilitate no particular fault in the operational condition lead to system
failure. Therefore, loss of vessel position does not occur due to a particular fault on the power
generations, distributions, and automatic valves, etc. However, it possibly failure will happen for
example in cables, pipes, manual valves as a static system.
• In vessels with DP class 3, flooding or firing occurs must also be removed in a non-system
enclosure. A loss of position should not result in a sudden defeat, including the entire distillation
section of the fire or the dewatering chamber.

Table 1 shows a summary of the IMO based on DP classification corresponds to DP

system components.

Table 1. Dynamic positioning vessel classifications.

Classification
Description IMO LR DNV ABS
No redundancy exists, the vessel is provided by
the manually stationary keeping and
Class 0 DP(M) DNV-T DPS-0
automatically heading control under the sea
disturbances.
No redundancy exists, only one computer system
for the DPS is equipped to automatically control DNV-AUT
Class 1 DP(A) DPS-1
the deviation of the vessel station and heading DNV-AUTS
displacement.
Two redundant computer systems are used to
automatically control the station and heading
movement under the sea disturbances. Hence, Class 2 DP(2A) DNV-AUTR DPS-2
the vessel position will not lose due to the failure
of the dynamic system.
Three redundant computer systems are
employed to control heading and environmental
disturbances, during the DPS failure containing Class 3 DP(3A) DNV-AUTRO DPS-3
the loss of the unit due to overflow or fire
situations.

2.3. Dynamic Positioning Mathematical Model

The mathematical calculation of DP vessels has a complexity class function. The first class is
described as the hydrodynamics typical procedure plant of vessels, which is a simplified mathematical
explanation and low-reliability model. This model would be set up as a part of the controller and is
signified as kinematic development that encloses the vital physical properties process individually.
The second mathematical model may have a general introduction of the actual construction, which is
expressed by high-reliability model formulations [55]. The key function of this simulation model is to
the dynamics of an actual vessel during the procedure of environmental disturbances, sensor outputs,
Energies 2020, 13, 3188 7 of 22

and control inputs. Moreover, the control models of roll and pitch are introduced in [56] for damping
the vessel motion of roll and pitch in small-water plane areas. However, the execution of control gains
for roll and pitch may be required to design an observer for approximating the roll and pitch angular
velocities. For this purpose, a mathematics reference model in DP vessels is used; this produces a
flexible set point to estimate acceleration, speed and position reference that are inputs to the positioning
controller. In [57], the authors described more details on reference models, guidance systems, and
some models of the local optimization reference points. Table A1 in Appendix A illustrates a summary
of DP mathematical models.

2.4. Dynamic Positioning Control Principles

Due to a lack of land resources, exploration in the oceans has been enhanced in recent years.
The ocean is doing a significant role in the development of economics and society. Therefore, the
exploration of ocean hydrocarbon resources has increased the demand for the DPS concept. DP systems
assist the vessels to be defined in a specific location or predetermined direction of platforms against
ocean environmental disturbances such as the wind, waves, and currents by using their propulsion
systems and engines. Compared to conventional stationary keeping equipment and anchor handling
operation, the DP mode has many advantages, containing positioning accuracy, a high degree of
flexibility, ability to operate at any depth of water, and avoiding marine pollution. Correspondingly,
DP systems have been widely used in different marine operations such as oil and gas exploration,
cable, and pipe laying. The first DP system, based on a conventional controller adapted to the PID,
was introduced in the 1960s [2].
The main achievements of PID based location development show that the PID controller can be
used to control the horizontal motions of vessels by using the thruster control system. This control
method was established in the 1970s, and the creation was named as DPS. The PID controller design
for DPS is considered an automatic feedback control problem that is a challenging problem to prevent
the forces of the first wave in the feedback loop. Some methods, such as the notch and low-pass
filtering, were tested for this purpose [6]. A ship is exposed to sea disturbances, forces and moments
produced by the ship’s thrusting system. Position reference systems, gyroscopic compasses, motion
reference unit, or vertical reference sensors measure the responses of the vessel to these deviations
in position, heading, and speed. Reading reference systems for scroll and ground are modified by
reading from vertical reference sensors and measured by wind sensors. The DPS is dependent on the
mathematical computational of the vessel, which comprises hydrodynamic properties such as current
tensile coefficients and robust mass data [5].

3. Review of Dynamic Positioning Controls

The DP controller includes a feed-forward and closed-loop controller. The closed control loop
suggested by Balchen has been encouraged and modified to the DPS concept. The calculation of the
feedback signals used in the controller as described in [4,5,7,55–59] based on Kalman’s theory and
survey. Complementary, further progressive control methods constructed on multi-adjustable control
optimization and Kalman’s filter concept are developed in the DPS control techniques as follows in the
following subsection:

3.1. Expanded Kalman Filter (EKF)

The (EKF) is introduced [55] to approximate the heading, position, and speed of the ship in each
of the three degrees of independence such as surging, swaying, and head deviations. It also includes
algorithms for computing the influences of the sea currents and waves. The EKF uses a mathematical
pattern of the ship to the precise demo of the real vessel motion behavior and constantly corrects the
model. The ship head and position deviations are evaluated by using of gyrocompasses and position
reference systems as input data to the DP system [56]. These estimates are compared with the predicted
Energies 2020, 13, 3188 8 of 22

measurement data generated by the mathematical model and the variances are then used to update
the model.
The position reference system computes a deviation for each position reference system in use
and places various weightings on their estimates according to each system’s individual quality in the
absence of position measurements [57]. That is to say that the system can perform positioning for some
time without position measurement updates from any position reference systems. In the EKF, the
mathematical ship model reliability and the noise level of the position measurement are the basis for
deciding how much to trust each estimate [58]. As time passes, the model uncertainty will be reduced
by learning from estimated ship responses. The method is tuned if, for instance, only one position
reference system is active, and it has a little update speed. In this situation, the model uncertainty
Energies 2020, 13, x FOR PEER REVIEW 8 of 22
may be increased in the period between measurements, and the ship model will be comprehensively
updated with each analysis [59]. Figure 4a,b demonstrates a simplified block scheme of the DPS
comprehensively updated with each analysis [59]. Figure 4a,b demonstrates a simplified block
controller and the expanded Kalman filter.
scheme of the DPS controller and the expanded Kalman filter.

Operational
Area

Environmental Forces Models

Wind , Wave, Current

Actual Forces to Thruster

Propeller
Allocation
Force and
Moment Demand
Desired Forces to Wind Wanted Position,
Propeller Measurement from
Controller
Heading and Speed
Reference Position System
and Sensors Position , Heading
Thruster Allocation and Speed Estimates

_ + Measurement from
Thruster Mathematical
Control Forces Reference Position
Feedback Vessel Model Position and System and Sensors
Set Point
Heading
Controller Estimates

Model Update
Filtered Motions

Measured Current
Motions
Extended Kalman Filter Estimate
Extended Kalman Filter

(a) (b)
Figure 4.
Figure 4. (a) A simplified block diagram
diagram of the DP control system
system and (b) the simplified
simplified block
block diagram
diagram
of the EKF in the DPS.
of the EKF in the DPS.

3.2.
3.2. Model
Model Predictive
Predictive Control
Control Model
Model
Over
Over the last three
the last three decades,
decades,the theconcept
conceptofofpredictive
predictive model
model control
control hashas
beenbeen widely
widely usedused in
in the
the industrial control system developments. In 1978, Richalet introduced his popular
industrial control system developments. In 1978, Richalet introduced his popular papers on the event, papers on the
event, application,
application, and implementation
and implementation of the
of the MPC MPC algorithm.
algorithm. The The theoryofofMPC
theory MPCutilizes
utilizes aa quadratic
quadratic
operational withdrawal optimization objectives and control response adjustment
operational withdrawal optimization objectives and control response adjustment strategies to strategies to dominate
the effect of
dominate thesignal
effecterror, controller
of signal error,parameters, and the environment
controller parameters, through thethrough
and the environment other undefined
the other
issues based on the predictive model [60,61].
undefined issues based on the predictive model [60,61].
As
As aa result
result of
of the
thecomputer
computer software
software and
and hardware
hardware technological
technological development,
development, itit hashas become
become
possible
possible to integrate the DPS and thruster controller into a combined algorithm through MPC.
to integrate the DPS and thruster controller into a combined algorithm through MPC.
Accordingly,
Accordingly, The MPC can mathematically perform highly accurate control of unknown plants by
The MPC can mathematically perform highly accurate control of unknown plants by
using
using linear
linear models
models of of ship
ship dynamic
dynamic in in the
the area
area of
ofthe
thedesired
desiredposition.
position. Moreover,
Moreover, the the advanced
advanced
method
method based
based on on feedback
feedback linearization
linearization isis evaluated
evaluated toto reduce
reduce thethe mathematical
mathematical model
model nonlinearity.
nonlinearity.
The
The MPC control technique simulates the future behaviors of ship replacement by using previous
MPC control technique simulates the future behaviors of ship replacement by using previous
control
controlinput
inputresults
resultsand,and,
consequently, predicts
consequently, suitablesuitable
predicts control output
controlresponses. It allows constraints
output responses. It allows
executed
constraints onexecuted
both input on and
bothoutput variables
input and output tovariables
predict disturbances’ impact on future
to predict disturbances’ impactprocedure
on future
evolution. Hence, the prediction model includes capabilities in the optimization
procedure evolution. Hence, the prediction model includes capabilities in the optimization method method to predict the
dynamic
to predictresponse of vessels
the dynamic overof
response a specified time
vessels over horizon. time horizon.
a specified
The MPC algorithm method is used for different restrictions of advanced control methodology
with complex controlled targets. In recent years, MPC has become an essential tool in the control
fields of industry. Due to the nonlinear behavior of DPS, the MPC was suggested for DPS as an
advantageous algorithm to overcome the conventional control method [30,31]. Based on an iterative
characteristic, the MPC algorithm is bounded perspective optimization of the dynamic model of a