1
A Computational study on the performance and emissions
parameters mapping of a ship propulsion system
Gerasimos Theotokatos
1
and Vasileios Tzelepis
1
corresponding author
email: gerasimos.theotokatos@strath.ac.uk
Department of Naval Architecture and Marine Engineering
University of Strathclyde, Glasgow, UK
Abstract
In the present paper, the mapping of the performance and emission parameters of a merchant
vessel propulsion system over the ship operating envelope was carried out by using a model capable of
representing the ship propulsion system behaviour. The model was developed based on a modular
approach and was implemented in the MATLAB/Simulink environment. The various parts of the
propulsion engine as well as the shafting system, the propeller and ship hull were represented by
separate submodels having the appropriate interface for exchanging the required variables to each
other. The output of the model includes the performance and emission parameters of the engine as well
as the operating parameters of the propeller and ship. Initially, the propulsion engine operation under
steady state conditions was simulated and the predicted engine performance parameters results were
validated. Then, simulations of the ship propulsion system operating points at various resistance curves
were performed. Based on the derived results, the mapping of the ship propulsion system performance
and emissions parameters was presented and their variation throughout the ship operating envelope was
discussed. Finally, an example of using the derived results in order to minimise the fuel consumption
and CO
2
emissions for a typical ship route is presented and discussed.
Keywords: Propulsion system modelling, mean value engine model, performance and emissions
parameters mapping, two-stroke marine Diesel engines
2
Introduction
As the worldwide shipping activities have been continuously growing, more pressure is put forth
towards the greener shipping in order to limit the shipping industry environmental impact. This was
initiated by the regulatory framework imposed by the International Maritime Organisation (IMO) and
the national authorities for the limitation of the non-greenhouse emissions, which include nitrogen
oxides (NOx) and sulphur oxides (SOx), as well as the greenhouse gaseous emissions; mainly the
carbon dioxide (CO
2
). The recent amendments of the international legislation [1] and the introduction
of the Energy Efficiency Design Index (EEDI) [2] as well as the Ship Energy Efficiency Management
Plan (SEEMP), which can be based on the Energy Efficiency Operational Indicator (EEOI) [3], focus
on the reduction of both CO
2
emissions and fuel consumption throughout the ship lifetime. Thus, apart
from the environmental benefits, the ship operational cost can be confined, which positively affect the
competitiveness of the shipping companies.
A number of measures can be used for complying with the existing legislation [4]. The reduction
of ship sailing speed, known as slow steaming [5], can reduce the consumed fuel, CO
2
emissions and
ship operational cost but it cannot be considered as an acceptable measure for increasing the ship
propulsion plant efficiency and achieving a value for the energy efficiency design index below the
imposed baseline value [1-2]. Measures that positively affect the ship energy efficiency [4] include the
improved hull and ship structure designs, which result in ship resistance decrease, as well as more
efficient propulsors designs that can inrease the ship propulsive efficiency. On the other hand, for
obtaining an environmentally cleaner operation of the ship propulsion plant, the engine manufactures
have been introduced new engine designs with higher stroke to bore ratio that can be combined with a
large diameter propeller [6-7], electronically controlled engine versions [7-8] in which the engine
settings can be adjusted over the engine operating envelope to optimise the engine performance, and
waste heat recovery systems [9-11]. In addition, retrofitting packages for engine cylinders cut off,
turbocharger isolation and turbochargers with variable geometry turbines have been presented [12-13],
so that the engine operation, especially at slow steaming, becomes more efficient,. However, for
complying with the more stringent regulations for NOx and SOx emissions [13-14], techniques such as
exhaust gas recirculation (EGR), selective catalytic reaction (SCR) and exhaust gas scrubbers must be
used [16-17], which however deteriorate the engine efficiency and increase the CO
2
gaseous emissions.
Alternatively, “cleaner” fuels such as liquefied natural gas (LNG) can be used for the ship propulsion
3
and auxiliary engines, thus significantly reducing CO
2
emissions and almost eliminating NOx and SOx
emissions [18].
Apart from the improved designs of the ship propulsion system, the efficient operation of the ship
is also crucial for constraining the fuel consumption and gaseous emissions over the ship lifetime. In
that respect, engine monitoring systems [19-20], which include recently developed sensors (e.g. sensors
for continuously measuring the engine cylinder pressure [21]), have been evolved and used onboard
ships for adjusting the engine settings in order to achieve the highest engine efficiency. In addition,
fleet management systems as well as optimum routing software packages [22] are used for reducing the
fleet consumed fuel, and thus the fleet operating cost. In these systems, the propulsion system initial
performance parameters must be provided, so that the propulsion system operation is evaluated by
comparing the performance parameters recorded values with their initial values, so that corrective
actions are taken in order to obtain the desirable operation.
In that respect, the engine shop and ship trials measurements can be used to provide the ship
propulsion baseline data. However, since the ship propulsion system will operate in a variety of
conditions throughout the ship lifetime depending on the ship resistance variation and the engine
degrading, the enrichment of the initial set of data is desirable. The quantification of the ship
propulsion system behaviour over the ship operating envelope can be addressed by mapping the ship
propulsion system performance and emissions parameters. These maps can be then used for optimising
the propulsion system throughout its lifetime, for adjusting the engine settings where significant
deviations are observed, for taking decisions for the engine and ship maintenance schedule, as well as
for obtaining better understanding of the ship propulsion system operation and the interaction between
the system components. The propulsion system mapping can be accomplished by using appropriate
ship propulsion system modelling techniques.
Various types of models have been used in the past for the simulation of marine engines and the
ship propulsion system under steady state and transient conditions. The more commonly used types are
the zero dimensional models [23-28] and the cycle mean value engine models [29-33]. The former are
more complex, require a large number of input data and their execution time lasts longer but can very
accurately represent the engine processes. The latter are simpler and therefore less laborious, require
limited amount of input data and their execution time is quite reasonable, whereas they can predict the
engine performance parameters with adequate accuracy. The cycle mean value models consider the
4
engine cylinders flow process taking place continuously, thus neglecting the intermittent nature of the
engine cylinders processes, and as a result, they can provide the engine cycle averaged temporal
evolution of the engine operating parameters, whereas their in-cycle variation (e.g. per degree of crank
angle) cannot be calculated. More detailed description of the mean value engine models is given in
[33]. For enhancing the models user friendliness, the concept of modular model development has been
previously used for the modelling marine diesel engine and the ship propulsion system components
[33-35]. According to that approach, the model is built using separate blocks for each one of the system
components, whereas the required variables exchanging between the blocks is accomplished using
appropriate connections. The advantage of this approach is that each submodel can be easily replaced
by a more detailed or a simpler one without interacting with the other model elements. The mapping of
engine performance and emissions parameters is a technique that has been used in previous research
studies. In [36], the marine engine and its turbocharger parameters maps were used for the development
of the engine control system. In [37] and [38], maps of engine cylinders parameters created based on a
zero dimensional model and then used for the control schemes development and testing of the
propulsion system of an ice-class ship operating at ice breaking conditions.
In the present paper, the propulsion system performance and emission parameters of a handymax
size vessel were mapped using the results derived by a modular approach built model, which was
implemented in the MATLAB/Simulink environment. The mean value engine model presented in a
previous study of the first author [33] was modified to incorporate appropriate blocks for the propeller
and ship longitudinal movement modelling. Separate blocks were used for representing the parts of the
ship propulsion system including the engine components, the shafting system, the propeller and the
ship hull, whereas the appropriate interface and connections were used for exchanging the required
variables between the model blocks. The engine cylinders were modelled using a mean value
modelling approach, the engine scavenging air and exhaust receivers were considered open
thermodynamic systems, whereas the turbocharger compressor and turbine were represented by their
steady state performance maps. The engine exhaust gas carbon dioxide (CO
2
) and sulphur dioxide
(SO
2
) emissions were calculated using the approach of perfect combustion in excess air, whereas the
nitrogen oxides (NOx) emissions were estimated based on typical values of NOx composition of the
exhaust gas. The propeller is modelled using polynomial equations for the non-dimensional torque and
thrust coefficients, whereas the ship surge dynamics is taken into consideration for calculating the ship
5
longitudinal velocity and position. First, the ship propulsion engine was simulated under steady state
conditions in order to validate the derived engine performance parameters results. Then, simulation
runs for a number of ship propulsion system operating points in various values of ship resistance were
performed and the derived results were used for mapping the propulsion system performance and
emissions parameters. The variation of the propulsion system performance and emission parameters
throughout the ship operating envelope is discussed. Finally, an example of minimising the consumed
fuel and emitted CO
2
gaseous emissions during the ship operation in a typical route based on the
mapping of the propulsion system performance and emission parameters is presented and discussed.
Ship propulsion plant modelling
The propulsion plant installation of a typical merchant vessel consists of the main engine, the
shafting system and the propeller. Depending on the vessel type and size, the ship main engine can be
two-stroke or four-stroke turbocharged marine Diesel engine. The shafting system comprises the
connecting shafts and the bearings and additionally for the four-stroke type engines, a gear box
installed between the engine crankshaft and the propeller shaft. In high power installations, a shaft
generator is often installed in order to produce the required electric power during ship voyages, where
the engine operates at relatively high load. In the case of merchant vessels propulsion plant installation,
the propeller is usually of fixed pitch type, although during the last years, designs with controllable
pitch propellers have also been used.
In this work, the ship propulsion system modelling was implemented in MATLAB/Simulink
environment following a modular approach, as it is depicted in Figure 1. Each part of the engine is
modelled using a separate block, which exchanges variables with the adjacent blocks of the model
through the appropriate connections. The marine Diesel engine is modelled using flow receivers
(control volumes) interconnected between flow elements. Fixed fluid elements having with constant
pressure and temperature are used for modelling the engine boundaries. Shaft elements are used for
calculating the engine crankshaft and turbocharger shaft rotational speeds. The engine governor
element, which is used to adjust the engine fuel rack position, is considered to be of the proportional-
integral (PI) type and incorporates the appropriate fuel rack limiters. The propeller and ship elements
are used for calculating the propeller and ship parameters, respectively.
