IRJET- Optimal Shipboard Power Management by Classical and Differential Evolution Methods

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 09 | Sep 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 269
Optimal Shipboard Power Management by Classical and Differential
Evolution Methods
B.Manasa1, K.Vaisakh2
1B.Manasa, M.Tech Student, Department of Electrical Engineering, Andhra University College of Engineering
2K.Vaisakh, Professor, Department of Electrical Engineering, Andhra University College of Engineering,
Visakhapatnam, Andhra Pradesh, India
---------------------------------------------------------------------***----------------------------------------------------------------------
Abstract - From past few years, optimal electrification of
inaccessible offshore systems has become important and
received extensive attention from maritime industry. Total
electrification of the shipboard power systems known as all-
electric ships (AESs) is subjected to introduction of electric
propulsion has led to the need for more cost effective
solution. With the increasing nature of energy demand in
modern ships whether with the growing needs for good
energy conservations and environmental protection have
intended to pursue AES (All-Electric Ship) configurations.
AES is envisioned to become an interesting technology with
great potential for both emission and fuel reductions when it
is compared with conventional ship power systems. But such
on-board systems are inclined to sudden load variations due
to fluctuating mission profile as well as weather conditions,
thus they have need for effective PMSs (Power Management
Systems) to operate optimally under different working
environments. Here in this paper, coordinated optimal power
management at the supply side of a given All-Electric Ship is
studied. This paper proposes a Differential Evolution
Algorithm, for Shipboard Power Management. To show the
usefulness of the proposed Power Management Systems
(PMS), the results are compared with Classical method.
Key Words: All-electric ship; constrained optimization;
co-ordinated energy management; Power
management system.
1. INTRODUCTION
A ship board electrical system is small in size and has
fewer components than a typical commercial power
system. A classic combatant ship may have 3 or 4
generators with collective capacity of 80-100 MW. Utmost
of this capacity is utilized by propulsion motors, for which
a two shaft ship will be rated with the range of 35-40 MW
each. These loads are large with respect to the total
generating capacity has made the analysis of on-board
ship power systems more problematic than commercial
power systems. Most of the simplifying assumptions made
in the analysis of the commercial power systems are
invalid with that of present day ship power systems. This
complication requires a detailed model of entire systems
including the relevant dynamics of each component.
A reliable supply of electrical power is very essential in
these days. With the increasing needs for improved energy
conservation, the initiative to pursue an AES (all-electric
ship) configuration has emerged [1]. All Electric Ship
configuration is expected to change the current ways of
power generation, distribution and consumption for the
on board energy sub systems and to create a exemplar
shift in the processes of control, monitoring and
conserving energy through utilizing power for meeting the
demand that is propulsion and service loads. Moreover,
AES is envisioned to become an interesting technology
with a great potential for both emission and fuel
reductions in comparison with the typical conventional
on-board ship power systems. In an All Electric Ship, the
electrical motor driven systems can be substituted with
the main diesel propulsion while the required power is
provided by various sources such as steam or diesel
engines, energy storage systems (ESS), gas turbines (GTs),
fuel fells (FCs) and possibly renewable – based prime
movers such as PVs (photo voltaic systems), allowing a
high efficiency throughout the entire range of operation
with respect to vessel speed.
The main challenge with All Electric Ships (AES) is to
design and incorporate a PMS (power management
system) for optimal scheduling of the on-board isolated
ship power plants [3]. Well planned operations of a ship
board electrical systems at the supply side (in terms of
optimal generator loading), together with the efficient
scheduling to meet the loads, in particular electrical
propulsion demand, can affect the overall systems
efficiency and ensure economic, environmental benefits.
PMS can co-ordinate controllable power sources and the
loads in a way to meet systems dynamic requirements for
short-run intervals.
Power Management Systems can be built on basis of
economic dispatch and unit commitment traditional
economic load dispatch deals with minimizing power
generation cost while satisfying set of equality and
inequality constraints. On the other hand, some toxic
gasses are emitted polluting environment due to operation
of the fossil fuel plants. Thus conventional minimum
operation cost cannot be made on the mere basis for
generation dispatch, emission minimization to protect
environment must also be taken care of.
Many algorithms have been proposed to solve power
management problem in shipboard power system.

Classical methods to solve the proposed problem are
lambda iteration method, Merit order loading, gradient
methods for optimal dispatch and priority list method,
dynamic programming methods for optimal combination
of units [2]. Apart from these classical methods we have
different optimization techniques for the economic
operation of generators, which have fast convergence and
capability of finding global minimum regardless of initial
parameter values. The optimal power management in an
All Electric Ship with regard to different objectives and
related technical, environmental constraints can be
formulated as a mixed –integer nonlinear programming
model [4-9]. In this paper optimal management problem
can be solved by a heuristic approach using differential
evolution.
The DE algorithm is inspired by sociological and
biological motivations and can take care of optimality on
discontinuous, rough and multi-modal surfaces.
Differential Evolution is one of the simple yet powerful
population based stochastic optimizer for dealing with
variety of optimization problems including multi-modal,
constrained, nonlinear, non-differentiable and multi-
objective. DE mainly has three advantages, finding the true
global minimum regardless of the initial parameter values,
fast convergence, and using a few control parameters.
Renewable energy sources, such as photovoltaic
energy systems have been increasingly integrated into
shipboard power systems and the applications of
renewable energy sources has become a global trend. The
photovoltaic energy systems on shipboard power systems
are installed to produce electricity and will be used to
supplement the diesel generators and thus reduce the
power required from these units. The proposed PMS
performance is analyzed based on a RO-PAX ferry with
integrated full electric propulsion and realistic constraints.
These results are compared with the results obtained from
classical method.
The rest of the paper is structured as follows: Section 2
particularizes on the features of Shipboard power system
and Power management along with the technical and
environmental constraints, Solution of power
management problem in shipboard systems with classical
method in Section 3, Section 4 extends to obtain solution
of power management problem with differential evolution
optimization method, In Section 5, optimal power
management problem in shipboard systems including
solar PV generation system to meet load along with diesel
generating units is discussed , Appendix, Analysis of the
results from the proposed PMS applied to AES are
presented in Section 6. Finally Section 7 gives the
conclusion of the paper.
1.1 LITERATURE REVIEW
Before Power management in offshore systems is
very difficult. The main challenges with AES is to design
and incorporate a power management system (PMS) for
optimal scheduling of the on- board power plants and
electrical loads [3].PMS can co-ordinate controllable
power sources and loads in a way to meet the system
dynamic requirements for short-run intervals. Power
management mainly deals with Unit commitment and
economic dispatch. The unit commitment in a power
system constitutes a study of paramount importance. The
study reveals the optimal combination of operating units
in a time horizon by considering all the technical and
environmental constraints along with meeting the varying
demand. Further study and analysis of future expansion,
classical methods for unit commitment can be easily
analyzed in [1].In this paper the author explains the
different classical methods by considering different
strategies.
After obtaining the unit status (optimal
combination of operating units) we obtain optimal
schedule of the power outputs of the operating units
which are present at that particular time in order to meet
demand. This optimal schedule is called as Economic
dispatch. Thus conventional minimum operation cost
cannot be the only basis for generation dispatch, emission
minimization must also be taken care of. Further study of
different classical methods for economic dispatch along
with environmental constraints is discussed in [17,18]. In
this paper the author explains different methods for
economic dispatch considering the technical constraints
and environmental constraints. Environmental constraints
are very essential to be considered in present day as
optimization of total cost along with reduction of pollution
is more important [20-22].
Power management system using Particle swarm
optimization and dynamic programming was explained in
the paper [2]. Whereas the differential evolution algorithm
to solve power management problem as explained by
authors in [25-29]. In this paper we can see the complete
history of differential algorithm and how it is used to solve
the power management problem.
2. SHIPBOARD POWER SYSTEMS AND POWER
MANAGEMENT
A fully electrified shipboard power system is
considered, where generated electric power supplies
mainly electric propulsion motors and ship service loads.
Ship propulsion is provided by large electric motors driven
by power electronic converters that enable continuous
shaft variable speed operation in a wide speed range,
operational flexibility and fuel economy. Also, the need for
large shafts for the coupling of propellers and prime
movers and the use of mechanical gearboxes is eliminated
[10, 11].
The passage of ‘Clean Air Act Amendments of 1990’and
its acceptance by all the nations has forced many utilities to
modify their operating strategies in order to meet the

rigorous environmental standards set by legislation. Thus
the modern operational strategies used in the generating
plants now include the reduction of pollution level up to a
safe limit set by the environmental regulating authority, in
addition to the minimum fuel cost strategy [18-21].
Traditional ships as well as AES must employ a well-
designed Ship Energy Efficiency Management Plan
(SEEMP) [14]. In the coming future the major targets of
SEEMP will be operation cost minimization and gas
emission limitation. So far SEEMPs have focused on CO2 .
However, the formulation of the problem can be easily
generalized and other pollutants beyond CO2 can be
included in the future. The targets of operation cost
minimization and GHG emission limitation might conflict
with each other, making the optimal power management in
AES a very challenging problem. In this context, if
propulsion power is appropriately adjusted in a way to
meet AES operation constraints, it could greatly contribute
to the limitation of GHG emissions in real time.
In the considered shipboard power system there are five
generator units to meet the propulsion and ship service
loads [12]. The fuel consumption (FC) of onboard thermal
unit may accurately be approximated by a second –order
polynomial of its produced power Pi as
2..)(
i
Pa
i
Pbc
i
P
i
FC 
(1)
Pi is the power generated, i is ith generator
Objective function
The total variable cost of the power plant(ToCe) is
calculated by taking into account the fuel cost(FCi), the
maintenance cost per power unit(MCi) and the start-
up/shut-down cost(SCij) of the i-th generator, producing
active power during a time interval [25].
).).))(.(( 1,
1 1

 
  jiijij
T
j
N
i
ijijiijiije StStSCTPMCPSFCStToC
g
(2)
Where T is the total time period under study, SFCi is the
specific fuel consumption, Stij is equal to 1 if unit I is
operating, otherwise 0 and NE is the total number of
electric generators.
iiiii PPFCPSFC /)()( 
(3)
The main objective of the problem is to minimize the
total operation cost of AES. This cost minimization should
be done subject to several constraints [13].
The technical constraints considered in the cost
function minimization while solving power management
problem are
(1) Generator loading limits
max,min, iiji PPP 
(4)
Where, Pi, min , Pi, max are minimum and maximum
power generating limits of ith generator (MW)
(2) Power balance constraint


gN
i
jpropjijij PLPSt
1
,.
(5)
Subscripts i, j denote i-th generator and j-th time
interval, respectively.
(3) Minimum up/down time constraint
iONiONiOFF Ttt min,_,,  
--UP time
(6)
iOFFiOFFiON Ttt min,_,,  
-Downtime
(7)
are the time points that i-th generator
stops or starts operating. TON_min, i ,TOFF_min,i are minimum
allowable operation time and non-operation time of i-th
generator.
(4) Blackout prevention constraint
}max{. max,,max, ijpropji
i
ij PPLPSt 
(8)
max{Pi, max} : maximum power of the committed units
(5) Generator ramp rate constraint
max,
1,
i
j
jiij
Rc
T
PP


 
(9)
Where Rci, max is the maximum rate of change of the
power produced by the i-th generator.
Apart from the technical constraints considered in the
cost function minimization while solving power
management problem emission constraint is also
considered in order to reduce emissions.
(6) Emission constraint
In order to reduce CO2 emissions,
sea
j
N
i
ijiijiji
EEOI
VLF
PSFCPStc
g
max,
1
.
)(...


(10)
port
N
i
ijiijiji
EEOI
LF
PSFCPStc
g
max,
1
)(...


(11)
Where ci is conversion factor for gas emissions
estimation for the i-th generator (gCO2/gFuel), Pij is Power
produced by i-th thermal unit in the j-th time interval
(MW), SFCi is Specific fuel consumption of the i-th
generator (gFuel/kWh), Vj is Ship speed in the j-th time
interval (kn), LF is Ship loading factor (tns).

EEOImax, sea is upper limit of EEOI when the ship is
travelling and EEOImax, port is upper limit of EEOI when the
ship is at port.
EEOI
International Maritime Organization (IMO) has
promoted a method to assess ship operation efficiency
based on the calculation of Energy Efficiency Operation
Indicator (EEOI)
EEOI is defined as,
orktransportw
mCO
EEOI 2

(12)
Where mCO2 is the CO2 mass produced during ship
power system operation. EEOI is the ratio of CO2 mass
emitted and the transport work. It indirectly represents
ship operational efficiency, as according to the efficiency
definition the required consumed energy to produce the
relative transport work should be used.CO2 mass is at an
extent proportional to the consumed fuel(energy).Hence,
ship operational efficiency and CO2 emissions are both
represented in EEOI in a balanced way [22].
Ship energy efficiency operation indicators EEOIs and
EEOIp when ship is in the open sea or at port are defined as,
j
i
ijiiji
jj
js
VLF
PSFCPc
TVLF
mCO
EEOI
.
)(..
..
2
,




(13)
LF is Loading factor
LF
PSFCPc
TLF
mCO
EEOI i
ijiiji
j
jp




)(..
.
2
,
(14)
Ship loading factor LF depends on the type of the
examined ship, e.g., passenger ship, RO-PAX ferry, etc. Here,
LF is applied to a RO-PAX ferry and it is calculated as,
FLD
nn
nn
LF
vp
vp



1.0.
'1.0.'
(15)
Where, np is the maximum number of passengers, nv is
the maximum number of the carried vehicles, n’p is the
number of passengers ,n’v is the number of carried vehicles
and FLD is ship full load displacement (tns).
3. SHIPBOARD POWER MANAGEMENT USING
CLASSICAL METHOD 1
Power management problem in Shipboard power
systems can be solved by using Classical methods. In this
paper Unit commitment problem is solved by Priority list
Method by considering technical constraints. Unit
Commitment (UC) is an optimization problem used to
1
determine the operation schedule of the generating units at
every time interval with varying loads giving minimum
operational cost under different constraints and
environments.
Priority of each unit to commit or de-commit before
unit scheduling is determined on the basis of unit
characteristics. Cost of power produced per unit, of a unit
at its maximum output power usually is less than that at
any other output power levels.so, it is expected to run a
unit at its maximum output power. Priority list is prepared
based on fuel cost obtained from the average fuel cost of
each unit operating at its maximum output power. The full
load average production cost of a unit is defined as the cost
per unit of power (Rs/MW) when the unit is operating at
its full capacity. Fuel cost of unit is expressed as
max
maxmax
max
)(
iii
i
i
i
ii
i Pcb
P
a
P
Pf
FC 
(16)
The units are ranked by their FCi in ascending order.
Thus, the priority list of units will be formulated based on
the order of FCi, in which a unit with the lowest FCi will
have the highest priority to share the load to be dispatched.
Optimal scheduling of load among the generators is
solved by lambda iteration method. The purpose of
economic dispatch is to determine the optimal power
generation of the units participating in supplying the load.
The sum of the total power generation should be equal to
the load demand in that particular time. The economic
dispatch problems is a constrained optimization problem
and it can be mathematically expressed i.e., total
production cost


n
i
iiiiit PaPbcC
1
2
(17)
is minimum, subject to constraint.


gn
i
Di PP
1 (18)
A typical approach is to augment the constraints into
objective function by using the Lagrange multipliers
)(
1


gn
i
iDti PPCL 
(19)
The minimum of this unconstrained function is found at
the point where the partials of the function to its variables
are zero.
0


iP
L
(20)
From the above equation the condition for optimum
dispatch can be obtained as


i
i
dP
dC
(21)
i=1,2,………,ng, which results in
 iii Pab 2 (22)
Pi can be calculated as
i
i
i
a
b
P
2



(23)
After obtaining the optimal schedule of powers we
check for power balance constraint, generator loading
limits of units which are operating in that time interval and
generator ramp up/down rates for calculating total cost.
Methodology
The procedure for implementing Shipboard Power
Management using classical method is given below
Step 1: Specify the minimum and maximum generator
loading limits of each unit. Specify the fuel cost of each
unit, load even.
Step 2: Determine the full load average production cost
of each unit and arrange them in ascending order (as in
Priority list).
Step 3: Obtain optimal combination of units for the load
in that particular time interval. Optimal combination of
units (unit status) is obtained by following priority list
scheme.
Step 4: Check if any violations in blackout prevention
constraint, minimum up/down time constraint for the
optimal unit status generated.
Step 5: If there are any violations repair the system
until the constraints are satisfied and obtain a set of
optimal combination.
Step 6: Specify the load requirement in each time
interval.
Step 7: Distribute the Powers to be generated in order
to meet the load in that time interval, among the operating
units at that time (which we know from the unit status
obtained).
Step 8: Optimal combination of power output
(economic dispatch) of all generating units which are
operating in that time interval is obtained by using Lambda
Iteration Method.
Step 9: Check for the technical and environmental
constraints and repair if any violations.
Step 10: Calculate total cost from the objective function
which includes fuel cost, startup/shut-down cost and
maintenance cost.
4. SHIPBOARD POWER MANAGEMENT USING
DIFFERENTIAL EVOLUTION
The solution for power management in Shipboard power
systems aims to optimize a selected objective function with
subject to different technical and environmental
constraints [23]. Mathematically, the power management
problem can be formulated as mentioned in Section II.
Differential Evolution
Generally, most of the classical methods of optimization
apply sensitivity analysis and gradient based algorithms by
linearizing the objective function and the system
constraints around an operating point. Unfortunately, the
Optimal Power management problem is a nonlinear and is
formulated as mixed-integer nonlinear optimization
problem. Hence, classical optimization techniques are not
suitable for such problem. Moreover, there is no approach
to decide whether a local optimum is also the global
optimum. Therefore, conventional optimization methods
that make use of derivatives and gradients may not be able
to identify the global optimum [26-28].
Recently, evolutionary optimization techniques have
been used to solve Power management problem to
overcome the limitations of classical optimization
techniques. A wide variety of heuristic optimization
techniques have been applied such as genetic algorithm
(GA), simulated annealing (SA), tabu search, and particle
swarm optimization(PSO). The results reported in the
literature were promising and encouraging for further
research in this direction.
In 1995, K.Price and R.Storn proposed a floating
point encoded evolutionary algorithm for global
optimization and named it differential evolution (DE)
algorithm owing to a special kind of differential operator,
which they invoked to create new off-spring from parent
chromosomes instead of classical crossover or mutation.
Similar to GA, DE (differential evolution) algorithm is a
population based algorithm that uses crossover, mutation
and selection operators. The key differences between GA
and DE algorithm are selection process and mutation
scheme that makes DE self-adaptive.
Differential Evolution is one of the simple
yet powerful population based stochastic optimizer for
dealing with variety of optimization problems including
multi-modal, constrained, nonlinear, non-differentiable and
multi-objective. The standard DE algorithm has been
competitive in solving global optimization problems over
continuous spaces. Due to its simplicity, robustness and
effectiveness, it has been successfully applied in various
scientific and engineering fields.
Moreover, there are only few control parameters
used to update the population of DE, thus it is easy for
implementation and parameter tuning. The mutation
operator, the crossover operator and the selection
operator are the three main evolutionary operators,

commonly applied in DE to update the population. First
two operators (the mutation operator and the cross-over
operator) are used to generate the trial vectors, and the
third operator (the selection operator) determines the
better one between the target vector and its trial vector for
the next generation based on their fitness values. The
standard DE algorithm and most of its improved variants
have operated on real values. However, the UCP in power
systems is dealing with 0-1, binary variables which
represents on/off schedule of the thermal generating units
in each time-period of planning horizon [24].
The main features of the DE algorithm can be stated as
follows:
1) Like any other evolutionary algorithm, DE starts
with a population size of NP individuals where the
individuals are D-dimensional variable vectors.
2) The subsequent generations will be represented by
discrete time steps like t=0,1,2,………,t,t+1,etc
3) Since the vectors are likely to be changed over
different generations, the following notation may be
adopted for representing the ith vector of the population
at the current generation (i.e. at time t):
⃗⃗⃗ ( ) ( ) ( ) ( )
This vector is referred to as 'genome', 'individual' or
'chromosome'.
4) Several optimization parameters must also be tuned.
These parameters are joined together under the common
name of control parameters. As a matter of fact, there are
only three real control parameters in the algorithm, which
are:
(a) Differentiation (or mutation) constant F,
(b) Crossover constant CR, and
(c) Population size NP.
(d) Dimension of problem D that scales the difficulty of
the optimization task;
(e) Maximal number of generations (or iterations) GEN,
which may serve as a stopping condition;
(f) Low and high boundary constraints of variables that
limit the feasible area.
DE Algorithm flow
Generate the initial population of size NP
Do while
For each individual, j, in the population
Generate three random integers
( )
Generate a random integer ( )
For each parameter i
( )
{
( )
End for
Replace with the child if is better
End for
Until the termination condition is achieved
Methodology
Management by using Differential Evolution is given below
Step 1: Specify the minimum and maximum generator
loading limits of each unit.
Step 2: Obtain the Unit status randomly. Units are
randomly committed by considering a variable h, whose
value is assigned by rand () function. Range of the rand ()
function to be (0,1). If the value assigned for h is more than
0.5 then we consider the unit to be 1(ON). If the value
assigned for h is less than 0.5 then we consider the unit to
be 0(OFF).
Step 3: Check for the Blackout prevention constraint
and minimum up/down time constraint by following the
repair algorithms.
Step 4: Unit status is updated if any constraints are
violated by using repair strategies employed respectively.
Step 5: For this updated unit status, obtain the optimal
schedule of power outputs (economic dispatch) in order to
get optimum cost.
Step 6: Check for the power balance constraint,
generator power limits and ramp rate limits of generating
units.
Step 7: If there are any violations in the constraints go
to repair strategy to satisfy them.
Step 8: Calculate total cost and fitness from the
objective function which includes fuel cost, startup/shut-
down cost and maintenance cost.
Step 9: Create a new population by using differential
evolution and calculate cost which satisfies all the technical
and environmental constraints.
Step 10: By following the DE cycle of mutation, cross-
over and selection obtain the best set of schedule.
Step 11: Obtain the optimal cost by using this optimal
schedule of power outputs from DE (as the control variable
is power).
Step 12: Check whether all the constraints are satisfied
and thus obtained cost is the best cost.

5. SHIPBOARD POWER MANAGEMENT WITH PV
USING DE
Apart from solving Power management problem in Ship
power system and obtaining the Optimum cost by using the
proposed Differential evolution algorithm, we are
incorporating PV to reduce the use of fossil fuels and to
reduce emissions. As the present day world is moving on to
renewables for generation of electricity, we have
considered solar as one of the reliable source of energy. A
PV system or solar power generation system is a power
system designed to supply solar power which is usable by
means of photovoltaics. Designing reliable and effective PV
systems requires understanding both the art and science of
photovoltaics and applying the strategies, skills and
techniques necessary to meet specific goals and objectives.
It consists of several components, including solar panels to
absorb and convert sunlight into electricity, a solar inverter
to convert output from direct to alternating current [29,
30].
Here we are installing three PV generating units of
100KW capacity each. We consider half of the PV
generation (0.15MW) in one case and perform the optimal
power scheduling with DE to solve power management
problem. We consider full generation capacity from PV
(0.3MW) in another case and perform the optimal power
scheduling with DE to solve power management problem.
Methodology
Management with PV using Differential Evolution is given
below
Step 1: The power generation from PV (solar) is
considered to be some value.
Step 2: At each time interval load can be considered as
the difference of the total load and the power generated
from solar in that particular time horizon.
Step 3: Specify thus obtained load for the Optimal
combination of units and Optimal schedule of power
generation among the operating units.
Step 4: Check for any violations in technical and
environmental constraints and repair them.
Step 5: Rest of the procedure for obtaining optimal
combination and optimal schedule which results in
optimum cost and fitness follows the steps as in DE
6. ANALYSIS OF RESULTS
The proposed optimization method is applied on RO-
PAX ferry with five generators supplying two electric
propulsion motors. The power management problem in
ships is solved by using Classical method and Differential
evolution method. Optimal power scheduling obtained
from classical method is presented in Fig. 1.
Fig. 1 Power generation schedule by classical method
Optimal power generation scheduling by DE method is
presented in Fig. 2
Fig. 2 Power generation schedule by DE method
Convergence characteristics of the cost function by DE
is shown in Fig.3
Fig.3 Convergence of cost function using DE
Operation cost during all time intervals obtained by
Classical and DE are presented in Fig. 4
Blue: Classical, Red: DE
Fig. 4 Operating cost by both Classical and DE method
0
10
20
30
40
1 2 3 4 5 6 7 8 9 10111213141516171819
Power(MW)
Time
P5
P4
P3
P2
P1
0
10
20
30
40
1 2 3 4 5 6 7 8 9 10111213141516171819
Power(MW)
Time
P5
P4
P3
P2
P1
0
1000
2000
3000
4000
1 2 3 4 5 6 7 8 9 10111213141516171819
Operationcost(m.u.)
Time

CASE 1
Optimal power generation scheduling with PV
(0.15MW) using DE is shown in Fig. 5 and Convergence
characteristics of cost function with PV (0.15MW) using DE
is shown in Fig. 6. Operation cost during all the intervals
obtained by DE and DE with PV (0.15MW) is presented in
Fig. 7.
Fig. 5 Power generation schedule with PV (0.15MW)
using DE
Fig. 6 Convergence of cost function with PV (0.15MW)
using DE
Fig. 7 Operation cost by DE and DE with PV (0.15MW)
CASE 2
Optimal power generation scheduling with PV (0.3MW)
using DE is shown in Fig. 8 and Convergence characteristics
of cost function with PV (0.3MW) using DE is shown in Fig.
9. Operation cost during all the intervals obtained by DE
and DE with PV (0.3MW) is presented in Fig. 10.
Fig. 8 Power generation schedule with PV (0.3 MW) using
DE
Fig. 9 Convergence of cost function with PV (0.3 MW)
using DE
Fig. 10 Operation cost by DE and DE with PV (0.3 MW)
0
10
20
30
40
1 3 5 7 9 11 13 15 17 19
Power(MW)
Time (hours)
P5
P4
P3
P2
P1
0
1000
2000
3000
4000
1 3 5 7 9 11 13 15 17 19
Operationcost
Time
0
10
20
30
40
1 2 3 4 5 6 7 8 9 10111213141516171819
Power(MW)
Time
P5
P4
P3
P2
P1
0
1
2
3
4
5
Category 1Category 2Category 3Category 4
Operationcost
Time

Fig. 11 Operation cost by Different Methods
Appendix2
The technical parameters of the ship and the onboard
power system are presented here. Ship parameters
depends on the type of ship,e.g., passenger ship, RO-PAX
ferry. Here we have considered RO-PAX ferry comprising
two large electric propulsion motors supplied by a set of
five electrical generators.
Table 1 Cost coefficients of generating units
Generat
or number
a B c
1 5.40 61.5 390
2 5.40 63 400
3 5.60 65 420
4 13.1 12 430
5 13.5 10 450
Table 2 Ship parameters
PARAMETER SPECIFICATION
TYPE RO-PAX ferry
Nominal speed (kn) 24
Maximum no of
passengers
2500
No of vehicles (nv) 700
Full Load Displacement
(tns)
70,000
EEOImaxs (gCO2/tn.kn) 27.5
EEOImaxp (gCO2/tn.kn) 165
Table 3 Data for Ship pay load
Part of the
Examined
Route
Number
of
Passengers,
nP1
Numbe
r of
vehicles,
nV1
Ship
Loading
Factor, LF
(tns)
Departure- 1955 600 58,616
2
Intermediate
port
Intermedia
te port-Final
destination
1720 500 49515
Table 4 Parameters of units
PARAMETERS GEN
1
GEN
2
GEN
3
GEN
4
GEN
5
Minimum UP
time (Hours)
1 1 1 1 1
Minimum
DOWN time
(Hours)
1 1 1 1 1
Start-up/Shut-
down cost
(m.u.)
0 0 0 0 0
CO2 emissions
(gCO2/g fuel)
Ramp up rate
Ramp down
rate
3.20
8
7
3.20
8
7
3.20
2
2
2.50
1
1
2.50
1
0
Table 5 Power Generation limits of units
PARAMETERS GEN
1
GEN
2
GEN
3
GEN
4
GEN
5
Technical
maximum
power (MW)
15 15 15 9 9
Technical
minimum
power(MW)
3 3 3 2 2
Nominal
power (MW)
15 15 15 9 9
Ship power systems are prone to sudden load
variations due to the changing weather conditions as well
as mission profile. Ship load (MW) during all the time
intervals and Ship speed during the entire route travelled
according to time interval during the complete route
travelled is in Table 6.
Table 6 Ship Load data and speed
Time interval Load (MW) Speed (kn)
1 19.9 17
2 27 19
39,500.00
40,000.00
40,500.00
41,000.00
41,500.00
42,000.00
CM DE DE PV 1 DE PV 2
Cost
Cost

3 33 20.5
4 35 21.2
5 36 21
6 37 22.5
7 34 22
8 32 20.5
9 29 19.8
10 6 0
11 22 17.5
12 28 18.5.
13 29 19.8
14 30 20.2
15 30 20
16 29 20.2
17 27.5 18
18 21.5 17
19 6 0
7. CONCLUSIONS
From the results it can be concluded that Power
management problem in ship power systems comprises of
UCP and optimal scheduling. Unit commitment problem
and optimal power scheduling (economic dispatch)
problem is solved by using Differential evolution method
and Classical method. Operating cost is calculated by
satisfying all the technical constraints. The total operation
cost obtained by classical method is 41,770.4986, whereas
the total operation cost obtained by differential evolution
optimization method is 40,759.4049.
Table 7 Data for Installation of PV (100 KW capacity)
Capacity of Power Plant 100KW
Generation per year 2,70,000
Cost of Electricity per unit Rs.1.8
Investment Cost per MW 68 Lakhs
Operation and Maintenance cost
per year
60,000
Payback period 25 years
Thus we can observe the optimum cost obtained by DE
method is smaller when compared with classical method
by 2.42%. The load sharing of generating units is analyzed
and presented under Case study and discussion section.
Operation cost with respect to time horizon is calculated
and is presented in section VI. Emission constraint is also
been satisfied. In order to obtain optimal power
management, power is generated by incorporating PV
system. The results are been analyzed by considering the
PV generation of 0.15MW in one case and 0.3MW in
another case. The total operation cost has been reduced by
0.39% in case of 0.15MW PV generation using DE and the
cost is 40,600.0917 in comparison with DE method. The
total operation cost has been reduced by 0.91% in case of
0.3MW PV generation using DE and the cost is 40,386.5387
in comparison with DE method. By this analysis we can
conclude integration of PV generation system in ship
power system is advantageous and reduces the emissions
apart from cost optimization.
REFERENCES
[1] F. Kanellos, A. Anvari-Moghaddam, and J.
Guerrero, “Smart shipboard power system
operation and management,” Inventions, vol. 1,
no. 4, 2016.
[2] A.J. Wood, B. F. Wollenberg, Power generation
operation and control, 2nd ed., New York:
John Wiley & Sons, Inc., 1996.
[3] Anvari-Moghaddam, A.; Dragicevic, T.; Meng, L.;
Sun, B.; Guerrero, J.M. “Optimal planning and
operation management of a ship electrical
power system with energy storage system”. In
proceedings of the 42nd Annual Conference of
the IEEE Industrial Electronics Society ,
Florence, Italy, 23-26 October 2016.
[4] Kanellos, F.D.; Tsekouras,G.J.; Hatziargyriou,N.D.
“Optimal demand side management and power
generation scheduling in All Electric Ship”. IEEE
Trans. Sustain. Energy 2014, 5, 1166-1175.
[5] McCoy, T.J. “Trends in ship electric propulsion
”.In proceedings of the IEEE Power Engineering
Society meeting, Chicago, IL, USA, 21-25 July
2002; pp. 343-346.
[6] Seenumani, G.; Sun,J.; Peng, H. “Real-time power
management of integrated power systems in all
electric ships leveraging multi time scale
property”. IEEE Trans. Control syst. Technol.
2012, 20, 232-240.
[7] Sudhoff, S.D. Currents of change. IEEE Power
Energy Mag. 2011, 1, 30-37.
[8] Gertsos, S.A.; Prousalidis, J.M.; Frangopoulos,
C.A. Electric propulsion: From infancy to
adolescence. In proceedings of the 8th
International Marine Design Conference (IMDC),
Athens, Greece, 5-8 May 2003.
[9] McCoy, T.J.; Amy, J.V. The state-of-the-art of
integrated electric power and propulsion
systems and technologies on ships. In
proceedings of the IEEE Electric Ship
Technologies Symposium (ESTS), Baltimore,
MD, USA, 20-22 April 2009.
[10] Faruque, M.O.; Dinavahi, V.; Sioderbeck, M.
Geographically distributed thermo-electric co-
simulation of all-electric ship. In proceedings of
the IEEE Electric Ship Technologies Symposium
(ESTS), Baltimore, MD, USA, 20-22 April 2009;
pp. 36-43.
[11] Prousalidis, J.M.; Tsekouras, G.J.; Kanellos, F.
New challenges emerged from the development
of more efficient electric energy generation
units. In proceedings of the IEEE Electric Ship
Technologies Symposium (ESTS), Alexandria,
VA, USA, 10-13 April 2011; pp. 374-381.
[12] Wu, W.; Wang, D.; Arapostathis, A.; Davey, K.

Optimal Power Generation Scheduling of a
Shipboard Power System. In proceedings of the
IEEE Electric Ship Technologies Symposium
(ESTS), Arlington, VA, USA, 21-23 May 2007;
pp.519-522
[13] Park, H.; Sun,J.; Pekarek, S.; Stone, P.; Opila, D.;
Meyer,R.; Kolmanovsky, I.; DeCarlo, R. Real-time
model predictive control for shipboard power
management using the IPA-SQP approach. IEEE
Trans. CONTROL syst.TECHNOL. 2015, 23, 2129-
2143.
[14] Mashayekh, S.; Butter-purry, K.L. An Integrated
security-constrained model-based dynamic
power management approach for isolated
microgrids in All-Electric Ships. IEEE Trans.
Power Syst. 2014, 29, 330-339.
[15] Tafanidis, K.K.; Taxeidis, K.D.; Tsekouras, G.J.;
Kanellos, F.D. Optimal operation of war-ship
electric power system equipped with energy
storage system.J. Comput. Model. 2013, 3, 41-60.
[16] Zahedi, B.; Norum, L.E.; Ludvigsen, K.B.
Optimized efficiency of All-Electric Ships by DC
hybrid power systems. J. Power sources 2014,
255, 341-354.
[17] Hadi Saadat, “Power System Analysis”, Tata
McGraw-Hill Education, 2nd Edition, 2002.
[18] Kanellos, F.D.; Prousalidis, J.M.; Skouras, G.J.
Control system for fuel consumption
minimization-gas emission limitation of full
electric propulsion ship power systems. J. Eng.
Marit. Environ. 2014, 228, 17-28.
[19] International Maritime Organization (IMO).
Prevention of Air Pollution from Ships; MARPOL
73/78 Annex VI; Technical Report; IMO:
London, UK, 2004.
[20] International Maritime Organization (IMO).
RESOLUTION MEPC.213(63): 2012 Guidelines
for the Development of a Ship Energy Efficiency
Management Plan (SEEMP). 2012.
[21] Kanellos, F.D. Optimal power management with
GHG Emissions limitation in All-Electric Ship
power systems comprising energy storage
system. IEEE Trans. Power Syst.2014, 29, 330-
339.
[22] Larsson, L.;Raven, H.C. The Principles of Naval
Architecture Series: Ship Resistance and Flow;
Society of Naval Architects and Marine
Engineers: Jersey City, NJ, USA, 2010.
[23] Seenumani, G.; Sun, J.; Peng, H. A hierarchical
optimal control strategy for power management
of hybrid power systems in all-electric ships
applications. In proceedings of the 49th IEEE
Conference Decision & Control (CDC), Atlanta,
GA, USA, 15-17 December 2010;pp.3972-3977.
[24] Jatinder Singh Dhalwal, Jaspreet Singh Dhillon.;
“Modified Binary Differential Evolution
Alogithm to Solve Unit Commitment Problem”,
Electric Power Components and Systems, 1-19,
2018
[25] Carrion M., and Arroyo J. M., “A Computational
efficient mixed-integer linear formulation for
the thermal unit commitment problem”, IEEE
Trans.on Power Syst, Vol. 21, No. 3, pp.1371-
1378, 2006.
[26] Patra S., Goswami S.K., and Goswami B.,
“Differential evolution algorithm for solving unit
commitment with ramp constrints”, Electr.
Power Compon. Syst, Vol.36, No. 8, pp.771-787,
2008.
[27] Uyar A. S., Turkay B., and Keles A., “A novel
Differential evolution application to short term
electrical power generation scheduling”, Electr.
Power Syst, Vol. 33, No.6, pp. 1236-1242, 2011.
[28] Yen S.J., Chen W.L., Chen J.C., Hsu S. C., and Wu
I.C., “An improved differential evolution
algorithm with priority calculation for unit
commitment problem”, In proceedings. IEEE Int.
Conference on Granular computing, pp. 352-
357, 2014.
[29] Tsekouras, G.J.; Kanelos, F.D. Optimal operation
of ship electrical power system with energy
storage system and photovoltaics: Analysis and
application. WSEAS Trans. Power Syst. 2013, 8,
145-155.
[30] Banaei, M.R.; Alizadeh, R.Simulation-based
modeling and power management of all-electric
ships based on renewable energy generation
using model predictive control strategy. IEEE
Intell. Transp. Syst. Mag. 2016, 8, 90-103.

IRJET- Optimal Shipboard Power Management by Classical and Differential Evolution Methods

More Related Content

What's hot (20)

Similar to IRJET- Optimal Shipboard Power Management by Classical and Differential Evolution Methods (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET- Optimal Shipboard Power Management by Classical and Differential Evolution Methods