1-s2.0-s0196890408002999-main.pdf

Energy Conversion and Management 49 (2008) 3468–3475

Contents lists available at ScienceDirect

Energy Conversion and Management journal homepage: www.elsevier.com/locate/enconman

A review on the use of gas and steam turbine combined cycles as prime movers for large ships. Part II: Previous work and implications Fredrik Haglind * Technical University of Denmark, Department of Mechanical Engineering, Building 402, DK-2800 Kgs. Lyngby, Denmark

a r t i c l e

i n f o

Article history: Received 1 January 2008 Accepted 5 August 2008 Available online 14 September 2008 Keywords: Combined cycle Large ship Prime mover weight Prime mover space requirement Start-up time

a b s t r a c t The aim of the present paper is to review the prospects of using combined cycles as prime movers for large ships, like, container ships, tankers and bulk carriers. The paper is divided into three parts of which this paper constitutes Part II. In Part I, the environmental and human health concerns of international shipping were outlined. The regulatory framework relevant for shipping and the design of combined cycles were discussed. Here, previous work and experience are reviewed, and an overview of the implications of introducing combined cycles as prime movers is included. In Part III, marine fuels are discussed and the pollutant emissions of gas turbines are compared with those of two-stroke, slow-speed diesel engines. In the past, combined cycles of COGAS and COGES configurations have been considered for ship propulsion. Another application where gas turbine-based systems have been considered as prime movers is for LNG ships, which are traditionally powered by steam turbines with gas-burning boilers. Previous experience for a cruiser shows that combined cycles weigh less and require a smaller volume than diesel engines, resulting in increased passenger capacity and/or voyage performance. Diesel engines respond faster to a load change than combined cycles. Provided parts within the steam cycle are warm, the start-up times of combined cycles and diesel engines are similar. Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction Today the great majority of prime movers and auxiliary plants of ocean-going ships are diesel engines. In terms of maximum installed engine power of all civilian ships above 100 gross tons, 96% is produced by diesel engines [1]. As for large ships, almost all are powered by slow-speed, two-stroke diesel engines. The primary reasons for this dominance are their high efficiency (also at part-load), and that they can run on heavy fuel oil (residual fuel oil manufactured at the ‘‘bottom end” of the oil refining process). Heavy fuel oil (HFO) is cheap fuel containing large quantities of sulphur and other impurities. The ship emissions contribute to acidification, eutrophication and the formation of ground-level ozone, as well as impact on the climate. Acidification is caused primarily by airborne deposition of sulphur, and in some coastal areas, shipping is the major contributor. The emissions of nitrogen oxides, carbon monoxide and volatile organic compounds give rise to the formation of ozone, which damages vegetation and affects human health. The groundlevel ozone is the dominant component in photochemical smog, and it has numerous health effects. Related to other power sources and engine types, the particulate emissions per kW h of slow-speed * Tel.: +45 45 25 41 13; fax: +45 45 93 52 15. E-mail address: [email protected]. 0196-8904/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.enconman.2008.08.004

diesel engines running on heavy fuel oil are large. Particulate matter poses human health concerns, especially that composed of the smallest particles which are suspected to be the most harmful. Following the increasing awareness of the environmental and human health concerns of shipping, legislative actions have been taken on global and national levels. Internationally, ship emissions are restricted by the International Maritime Organization (IMO). Current regulations are covered in MARPOL Annex VI which was put into force in May 2005 [2]. Sulphur oxides are restricted by a global cap of 4.5%mass on the sulphur content of fuel oil, and the IMO is to monitor the worldwide average sulphur content of marine fuels. Limitations of oxides of nitrogen are covered in Annex VI (regulation 13) in a mandatory NOx technical code. Within the EC current regulations on the fuel sulphur content are more stringent [3,4]. Even stricter regulations are imposed in Sweden, where environmentally differentiated fairway and port dues have been applied since 1998. As a consequence of the limitations on the fuel sulphur content, the price of HFO is very likely to rise. In order to reduce the sulphur content of HFO there are different methods, and the price premium will be dependent on the quantity required; the more required, the higher the price. A low quantity of low-sulphuric fuel would be produced by re-blending distillate fuels; but a large quantity of low-sulphuric fuel would require refinery investments in residue desulphurization/conversion facilities.

F. Haglind / Energy Conversion and Management 49 (2008) 3468–3475

3469

Nomenclature CEC CODLAG COGAS COGES DWT EC EV HFO HRSG

Commission of European Community COmbined Diesel-eLectric And Gas turbine COmbined Gas turbine And Steam turbine COmbined Gas turbine Electric and Steam deadweight tonnage economiser European commission evaporator heavy fuel oil heat recovery steam generator

A 2002 study suggests that if the sulphur content of all marine HFO produced worldwide would need to be reduced to 1.5%, the price of HFO would reach a level similar to MGO [5]. However, this conclusion was based on the average European fuel bunker price over the period 1997–2002. Since then the prices of both HFO and MGO have increased several times, implying that the results of the 2002 study cannot be applied to today’s situation without substantial revisions. Besides, future legislation will impose much harder restrictions on the fuel sulphur content (see Part I), raising the price premium of HFO even more. Whether future restrictions on the sulphur content of HFO will make distillate fuels competitive is unknown. What is clear, however, is that such regulations will change the prevailing economical situation, reducing the price difference between HFO and distillate fuels. In order to lower the environmental and human health impacts of shipping, and conform to current and future regulations regarding fuel quality and pollutant emissions, finding novel technologies that offer improvements in these aspects is essential. A combined cycle is a plant in which a higher thermodynamic cycle produces power, but all or part of its heat rejection goes to supply heat to a lower cycle [6]. The objective of combined cycles is to achieve a greater work output for a given fuel supply. This is attained by increasing the mean temperature of heat supply and/or decreasing the mean temperature of heat rejection (compared to those of single cycles), thus approaching more closely the best possible power plant, based on the Carnot cycle. Combined cycles featuring one or several gas turbines (higher cycle) and a steam cycle (lower cycle) is a power plant option commonly used for power production that offers high efficiency and inherently low levels of pollutant emissions, see Fig. 1. Combined cycles have rarely been used in the past for marine applications, but following the expected legislative actions, increasing environmental awareness and the increasing price of heavy fuel oil, it might be a viable option for the future. The aim of the present paper is to review the prospects of using combined cycles as prime movers for large ships, like, container ships, tankers and bulk carriers. When discussing previous work, however, the review is not limited to applications for those ship types, but all applications are included. The primary reason for focusing on large ships is that combined cycles need to be fairly large in order to attain a high efficiency (see also Section 3.1). The paper includes descriptions of environmental and human health concerns of shipping, and the regulatory framework, leading to the interest in unconventional prime movers for large ships. Previous work in this field is reviewed and some of the implications of using combined cycles as prime movers are discussed. Marine fuels are discussed, in particular in the context of the requirements of gas turbines. Since one of the drivers for considering combined cycles for powering large ships is the desire to reduce pollutant emissions, the mechanisms for emissions formation in gas turbines and diesel engines are addressed. Considering the prospects of introducing novel prime mover concepts, one crucial question is whether these plants can achieve

IFO IMO LNG MGO NOx SH SO2 TEU

intermediate fuel oil International Maritime Organization liquid natural gas marine gas oil oxides of nitrogen (NO and NO2) superheater sulphur dioxide twenty-foot equivalent unit

efficiencies over the load range comparable to those of a slowspeed, two-stroke diesel engine, which is the conventional prime mover for large ships today. In order to address that question properly, detailed performance calculations considering the whole load range would be required. Moreover, the propeller arrangement (governing the transmission efficiency from main engine to propeller) needs to be considered. Such performance computations are beyond the scope of the current paper. A qualitative discussion about the efficiency of combined cycles is included in Section 3, but detailed performance calculations are the subject of further work. The paper is divided into three parts of which this paper constitutes Part II. In Part I, the environmental and human health concerns of international shipping were outlined. The regulatory framework relevant for shipping and the design of combined cycles were discussed. Here, previous work and experience are reviewed in Section 2, and an overview of the implications of introducing combined cycles as prime movers is included in Section 3. Finally, Section 4 contains a summary and outline of conclusions. In Part III, marine fuels are discussed and the pollutant emissions of gas turbines are compared with those of two-stroke, slow-speed diesel engines. 2. Previous work and experience A review of existing literature reveals that combined cycles as prime movers for ships have been studied in the past for various reasons and applications, and a few ships using such plants have been built. In this section a brief overview of previous work is outlined, and the ships, including their propulsion systems, powered by combined cycle plants are described. 2.1. Previous work Due to concerns of depletion of world petroleum reserves from time to time and the will to render the vessels more effective, the combined cycle of COGAS configuration has been considered as an option for naval and commercial applications in the past [7–13]. The naval ships are in most cases powered by gas turbines, and by introducing heat recovery of the exhaust and a steam cycle, the thermal efficiency of the plant is improved (see also Section 1). By using a more efficient propulsion system, the ship cruising range can be extended, or the required fuel capacity (for the same range) can be reduced. Mills [12] concluded that the use of a combined cycle power plant would provide large savings in fuel, ship size and construction, maintenance, and operating costs compared with gas turbine propulsion. The study by Brady [10] suggested that it was possible to reduce the fuel consumption of naval gas turbine-power ships by about 33% by introducing a COGAS plant, at the price of additional space and weight requirement of the system. Merz and Pakula [11] presented a COGAS marine system that

3470

F. Haglind / Energy Conversion and Management 49 (2008) 3468–3475

Fig. 1. Combined cycle power plant featuring two aero-derivative gas turbines (consisting of a gas generator and power turbine) and a single-pressure steam cycle. The heat recovery steam generators (HRSGs) are of drum-type (see Part I).

attained a thermal efficiency of 41%, and claimed that the plant offers low fuel rate, flexibility and simplicity, and minimum environmental impact. The control and dynamic behaviour of COGAS plants were studied by Abott et al. [13]. More recently, a number of different studies on the use of combined cycle power plants of COGES configuration as prime movers for various ship types have been published. Ahlqvist [14] compared five different propulsion systems, namely, diesel-electric, COGES based on aero-derivative gas turbines, COGES based on industrial gas turbines, COmbined Diesel-eLectric And Gas turbine (CODLAG) based on aero-derivative gas turbines, and diesel-mechanical, for application on a 2500 passenger cruise ship. In particular, the possibility of increasing the availability of the power plants through introduction of redundancy (in order to increase the safety of the ship) was studied. It was concluded that the diesel-electric system is the most promising, particularly if focusing on the costs (investment and operating costs) and the redundancy of the propulsion system. Moreover, investments for all systems with electric transmission were estimated to be approximately equal. In Shipping World and Shipbuilder [15] a number of different gas turbine plants and combined cycles plants powering a fast ferry and cruise ship, respectively, were compared from a systems point of view. It was concluded that the best way to combine the components into a system is application specific. For fast ferries where high power, and weight and space constraints are paramount, gas turbines with mechanical drive is preferred. The COGES configuration works well for cruise ships with frequent part-power operation and large steam demand for auxiliaries. By using the electric drive CODLAG plant, consisting of one gas turbine and two diesel engines, it is possible to increase the number of passenger cabins and gain a significant amount of public space. The COGES system has also been suggested as prime mover for merchant ships in order to cope with future environmental regula-

tions in the Baltic Sea area [16]. In this study a diesel plant, gas turbine system and COGES plant were compared in terms of performance, costs and environmental impact. When applied to a fast Ro-Ro ship of 10,000 DWT loading capacity, a plant efficiency of 46.8% for the COGES system was attained, while the corresponding figure for the diesel system was 43.6%. According to Domachowski and Dzida [16], gas turbine systems have many technical advantages over the diesel system; however, slow-speed diesel engines are superior in terms of specific fuel oil consumption cost. For the future, however, they believe that the situation may change. Domachowski and Dzida [16] concluded that provided that regulations prohibit the use of fuels with high sulphur content, gas turbine-based systems will become comparable in terms of specific oil consumption cost. Another application where gas turbine-based systems have been considered as prime movers is for ships transporting liquid natural gas (LNG) – a fleet type which is expanding rapidly. LNG is transported in un-pressurised, insulated tanks, and the temperature of the cargo is kept constant by free vaporisation [17]. A requirement on the propulsion system is that it must be capable of utilising the boil-off gas as fuel. Thus, traditionally these ships have been powered by steam turbines with gas-burning boilers. The natural boil-off represents about one-third of the fuel consumption at sea; the rest is vaporised using heat exchangers. The primary disadvantage with the steam cycle is the poor efficiency, and hence alternatives like diesel engines or systems based on gas turbines have been considered. The diesel engine would have to be adapted for a gaseous fuel; otherwise expensive liquefaction plants would be required, making it a less attractive option. Because of advantages in terms of economics (considering both first costs and operating costs) and redundancy, the COGES plant is believed to become a viable competitor to the steam plant for LNG ships [17].

F. Haglind / Energy Conversion and Management 49 (2008) 3468–3475

As prime mover for LNG ships, a COGES plant consisting of the aero-derived gas turbine MT30 rated up to 36 MW and a steam turbine rated up to 10 MW has been suggested [18]. It is claimed to achieve a net plant efficiency of 49%. Another possibility which has been proposed for LNG ship propulsion is a COGES plant with a gas turbine capable of operating on gas and/or liquid fuels simultaneously [19]. This plant features the SGT 500 gas turbine, and its fuel system is designed to ensure that all boil-off gas is used, with liquid fuel delivered only when it is needed. 2.2. Practical experience To the author’s knowledge there are only two ship types powered solely by gas and steam turbine combined cycles that have been built in the past. This section describes these two ship types. 2.2.1. Millennium ship In 2000, the world’s first cruise ship powered by a gas turbinebased power plant, Celebrity Cruises’ GTS Millennium, began commercial operation, see Fig. 2 [20]. Cruise ships are usually powered by diesel-electric engines. The Millennium ship and others of her class use a combined cycle power plant with a turbo-electric transmission, i.e. a COGES plant (see Part I), consisting of two LM2500+ aero-derivative gas turbines, delivering 25 MW apiece, and one steam turbine, delivering 9 MW. Each of the gas turbines is coupled to one Brush alternating current generator. In addition to providing power to the electric motors/propeller pods, it provides all onboard power requirements. Each gas turbine measures 15 m in length and weighs about 100 tons [20]. The operating speed of the ship is 24 knots [21]. At steam turbine inlet, the steam pressure is 32 bar (any value on temperature is not available in the public domain). After the steam has been expanded to 3 bar, some of it is taken off for water production, air conditioning reheat, galleys, and laundry services. The plant layout is similar to that shown in Fig. 1, excluding the dearator and including steam extraction. Including the service steam supply, the fuel consumption is up to 7% lower than that of an equivalent medium-speed diesel installation [20]. In figures, the overall energy utilisation varies from 45% to 50%. The corre-

3471

sponding COGES efficiency, i.e. the ratio of power produced to the power provided via the fuel, is a few percent lower. As outlined in Part I, in order to maximise plant efficiency of combined cycles during part-load, it might be advantageous to shut one gas turbine off and run the remaining at full power where their efficiencies are high, rather than running all of them in part-load. When the Millennium ship is running at high-speed transition, which is typical for a long deployment route or when connecting between two long-distance ports, both gas turbines are running at high power in conjunction with the steam turbine. Whereas, if the ship is to operate on a short transit run, only one gas turbine is operated at high power providing steam to the steam turbine generator set. In port, only one gas turbine is operated at part power in conjunction with the steam turbine. Generally, when changing from a diesel engine-based plant running on HFO to a COGES arrangement, significant reductions of pollutant emissions are expected. According to Tinsley [21] the emissions of nitrogen oxides (NOx) are reduced by 80%, and sulphur dioxide (SO2) by 98%. Sannemann [20] reports that total NOx emissions of 5 g/kW h were measured for the COGES plant during GTS Millennium sea trials. Furthermore, visible smoke was not observed at any power setting [20]. The power plant supplies power to electrically driven podded propulsion units featuring fixed pitch propellers. The pod can be rotated through 360° to provide thrust in any direction, thus eliminating the need for stern tunnel thrusters (normally used for port operation), rudder shaft lines, steering gear bossing and brackets. The electricity to the pods is provided by speed-controlled alternating current electric motors. After a few years of operation it was decided to retrofit some of the Millennium ships. By the last quarter of 2007, with the intention of saving fuel, an auxiliary diesel generating set of 11.6 MW was planned to be added [22]. This engine will provide base load power, mainly for the hotel, both at sea and in port. It will be running on HFO; thus another fuel system needs to be added. Before it was decided to install a diesel engine, other possibilities to cut the operating cost were considered, including using less costly alternative grades of fuel. In addition, gas turbines from another manufacturer were considered and various fuel alternatives, including

Fig. 2. The Millennium cruise ship (picture by AKER YARDS).

3472

F. Haglind / Energy Conversion and Management 49 (2008) 3468–3475

biodiesel, were tested. Biodiesel will now be used where appropriate to reduce MGO consumption. 2.2.2. Russian merchant ships Combined cycle power plants have also been used for Russian merchant ships. About 30 years ago two Russian 25-knot Ro-Ro vessels powered by combined cycle power plants with direct power transmission (COGAS) were built. Not very much data on these ships are available in the public domain. However, a few fundamental design and performance data from Shipbuilding and Marine Engineering International [23] are gathered here. The plant is a two-screw configuration with one aero-derived gas turbine and one steam turbine connected to each shaft. At normal sea service output, each shaft delivers 17 280 kW, of which 14,120 kW comes from the gas turbine and 3160 kW from the steam turbine. The maximum capacity of the steam turbine is 4270 kW, but at sea service, some of the steam generated in the heat recovery steam generators is extracted before the steam turbine and used to power turbo-alternators and for the ship’s service. The steam conditions at steam turbine inlet are 15 bar and 310 °C. At maximum power output the specific fuel consumption is 238 g/kW h. Assuming a lower fuel heating value for the fuel of 42.8 MJ/kg, this corresponds to an efficiency of 35.3%. 3. Brief overview of implications Introducing combined cycles as prime movers will affect a range of items. In this section a number of issues are discussed, and differences compared with diesel engines are highlighted. 3.1. Plant efficiency An essential feature of the prime mover is the plant efficiency, which influences the operating cost and the pollutant emissions, primarily carbon dioxide. Large stationary modern combined cycle power plants of a few hundred MW used for power production, achieve thermal efficiencies of about 60%. When used for large ship propulsion, the plants are smaller (about 50–100 MW), resulting in smaller dimensions of turbo-machinery. In order to avoid very small dimensions, which would cause excessive losses, the pressure levels need to be kept at moderate levels. Lower component efficiencies and moderate pressure levels, particularly in the steam cycle, result in lower efficiencies for combined cycles used for ship propulsion. Estimating what efficiency levels can be expected of combined cycles designed for marine applications is, as pointed out in the introduction, beyond the scope of the present paper. Quite low efficiency values for combined cycles were quoted in Section 2; however, these plants either were based on old technology and/or were designed such that the potential of achieving maximum efficiency was not fully exploited. 3.2. Weight and volume considerations The power density of gas turbines is very high, especially if related to diesel engines, making the space requirement of a gas turbine-based plant low. The additional space can be used to increase the cargo/passenger capacity. Furthermore, for a given power level, the weight of a plant consisting of gas turbines is lower than that consisting of diesel engines. A reduced main engine weight is beneficial because it lowers the ship displacement, which in turn reduces the power requirement. Alternatively, unless the ship is volume constrained, the cargo/passenger capacity can be increased while retaining the displacement. Using a gas and steam turbine combined cycle as prime mover for a post-panmax container ship, the lower weight and volume

requirement of the combined cycle compared with the slow-speed diesel engine results in payload benefits. A recent study suggests that with respect to the volume constraint, the vessel can carry another 74 TEU containers and with respect to the weight constraint, the vessel can carry another 30 TEU containers [24]. Conventionally, the payload of the vessel is 6000 TEU containers, which implies that the payload can be increased by 1.2% or 0.5%, respectively, depending on whether the volume or the weight is considered the limiting factor. By designing the whole vessel from start for a combined cycle prime mover is likely to bring even larger benefits. Figures of the machine room with diesel engine and combined cycle are shown in Figs. 3 and 4. Preferably for freeing as much space as possible, the heat recovery steam generators should be located within the funnel (and not within the machine room as in Fig. 4), which they are for conventional vessels recovering heat of the diesel engine exhausts. The benefits in terms of weight and space requirement of the COGES plant over the diesel-electric system, which is the traditional choice for most cruise ships, is illustrated for the Millennium ship (Section 2.2.1) [20]. As a consequence of the lower space requirement of the COGES plant, space for up to 50 additional passenger cabins was freed. According to Sannemann [20], by other arrangements, such as locating the engines aft or in the funnel, it would be possible to gain even more space. In terms of power requirement, the COGES installation requires less auxiliary machinery, namely, about 500 kW. Moreover, due to the reduced weight of the COGES plant, the ship displacement was reduced by 1000 tons, giving a 1.6% reduction in propulsion power. In total, the vessel power requirement was reduced by 1150 kW. Another study [14] comparing various combined cycle concepts with diesel systems to be applied for a 2500 passenger cruise ship, suggests similar figures. Following this study, the reduced area requirement of the COGES system compared with the diesel electric system, results in another 23 (2.5%) passenger cabins, and the lower weight reduces the power requirement by 1–1.5%. Domachowski and Dzida [16] compared a COGES plant with a medium-speed diesel plant as prime mover for a fast Ro-Ro ship of 10,000 DWT loading capacity. Their results suggest that the specific mass, i.e. the ratio of total power plant mass to the rated output of the engine, is more than three times higher for the diesel plant. In terms of power density, i.e. the ratio of the rated output of the engine to the engine room volume, their results suggest an even larger advantage for the COGES plant, namely, a power density about 16 times higher for the COGES plant. 3.3. Costs Based on previous work it is difficult to present any reliable cost figures of marine combined cycle power plants, and making a detailed cost analysis is beyond the scope of the current paper. There are several reasons for the difficulties of presenting reliable cost figures. Assessing costs of unconventional technologies is always difficult, because costs of components are not always known. Furthermore, the operating cost is governed by the fuel cost, which vary widely, and particularly considering future restrictions on fuel sulphur content, future fuel prices are unknown. Results of previous studies comparing costs of combined cycles with diesel engines are not consistent, and they are not based on current technologies; nevertheless, some figures are compiled below. As for the first cost, when comparing engines for cruise ships, Koehler [25] claims that this is lower for diesels, whereas Ahlqvist [14] concludes that these are similar. Diesel engines consume large quantities of lubricating oil, whereas gas turbines consume only very small quantities, typically 1% of that of diesel engines [25]. This is favourable for the gas turbine operating cost, but the

F. Haglind / Energy Conversion and Management 49 (2008) 3468–3475

3473

Fig. 3. Machine room with diesel engine of a post-panmax container ship [24].

advantage is fairly small because the lubricating oil cost makes up only a minor portion of the total operating cost. When comparing medium-speed diesel engines with a combined cycle with electric transmission (i.e. COGES configuration), Domachowski and Dzida [16] estimate that the specific power cost is 12% lower for the combined cycle. These values were derived assuming that both engines were running on marine diesel oil. However, generally diesel engines (particularly those of slow-speed type) run on heavy fuel oil which is remarkably cheaper than distillate fuels (see Section 1). Koehler [25] does not quantify the difference, but claims that the operating cost is higher for combined cycles. Experience shows that the required maintenance by crew personnel for the combined cycle is significantly less than for a typical diesel-powered vessel [20]. Furthermore, the modular structure of systems with gas turbines makes the replacement of failed parts fast. 3.4. Operational issues and start-up The fact that gas turbines generally are not reversible, while diesel engines are may, in some circumstances, affect the manoeuvrability of the vessel. However, for a combined cycle featuring an electric transmission, there is no disadvantage (because the electric engine powering the propeller is reversible). Neither is there a disadvantage for a combined cycle with a directly-driven propeller, because for such arrangements reversible transmission gears or controllable pitch propellers can be used [16]. Both engine types

have shown to have reliable operation, and there is no obvious advantage for either of the concepts in this respect. Due to the higher frequencies of gas turbines (which are more easily damped than lower frequencies), the vibrations of gas turbines are lower than those of diesel engines. In addition, the noise level is less from gas turbines. Diesel engines respond faster to a load change than combined cycles [25]. Large two-stroke diesel engines can accelerate from zero to 30–40% load in a few seconds; the remaining acceleration to full load, however, takes significantly longer. Typically, when used for ship propulsion such engines are accelerated from zero to full load in about 0.5–1 h [26]. In theory, this can be done much faster, but it is avoided because it would cause high thermal stresses. A gas turbine with a free power turbine typically needs five seconds or more for accelerating from idle to full load [27]. That is, about two-thirds of the power of a combined cycle can be obtained within this timeframe. The remaining power contribution coming from the steam turbine, takes longer. However, owing to the large vessel inertia, engine acceleration time is generally not critical for ships [27]. The start-up time is very much dependent on whether engine parts are cold or warm before start-up, especially for the combined cycles. In practice, diesel engines used for ship propulsion are not allowed to become cold (i.e. equal to the ambient temperature). When the ship is not in operation, the engines are kept at a reasonable temperature using the cooling water. The temperature of the cooling water is that normally used during operation. The cooling

3474

F. Haglind / Energy Conversion and Management 49 (2008) 3468–3475

Fig. 4. Machine room with combined cycle of a post-panmax container ship [24].

water, in turn, is heated by the exhausts of the auxiliary engines, which consist of four-stroke diesel engines, so called, generating sets. From this condition, full load can be reached within 1 h, without exceeding thermal stresses for materials. The gas turbine start-up time is independent of standstill time, while the start-up time for the steam cycle is dependent on the time required to heat up parts of the machine without exceeding thermal stresses imposed by the material. Kehlhofer et al. [28] state that for gas turbines used for large stationary combined cycle power plants, it takes less than 30 min from when the start-up sequence has been activated until it has reached full load. For marine gas turbines, a start time of less than 90 s throughout the operational envelope is mandatory [27]. A quick start of the gas turbines of the COGES plant powering the Millennium cruise ship (see Section 2.2.1) takes typically 3 min or less to reach synchronous idle, after which load can be taken within seconds [20]. In light of these figures, start times for gas turbines adapted for marine combined cycles of less than 10 min seem feasible. The starting-up process of a combined cycle can be divided into three main phases [28]:  purging of the HRSG;  gas turbine speed-up, synchronisation and loading;  steam turbine speed-up, synchronisation and loading. The process is initiated by purging of the HRSG in order to prevent the explosion of any unburned hydrocarbons left in the system from earlier operations. This is done by running the gas turbine at ignition speed (approximately 30% of nominal speed) with the generator as motor, or similar equipment, and blowing air through the HRSG. The purge time depends on the volume that should be purged; it is advisable that the equipment is exchanged

with a factor of five with ‘‘clean air” before ignition of the gas turbine can take place. Subsequently, the gas turbine is ignited and run up to the nominal speed, synchronised and loaded to the desired load. During the gas turbine start-up sequence, steam is generated in the HRSG. Depending on actual start-up conditions, i.e. temperature of equipment, the appropriate steam properties for a steam turbine start-up is reached at approximately 50–60% gas turbine load, corresponding to 40–60% nominal pressure and a sufficient degree of superheat (around 50 K). Before this point the steam flows across the steam turbine by-pass. Subsequently, after having evacuated the condenser, the steam turbine is started and run up to the nominal speed, synchronised and loaded to the desired load. In total, starting up a combined cycle in the range of 50– 400 MW after 8 h standstill (i.e. components are still hot) takes about 40–50 min [28]. A warm start (60 h standstill) is expected to take 75–110 min, while a cold start (120 h standstill) is expected to take 75–150 min. Since a combined cycle to be applied as the marine main engine is in the lower power range, the start-up times are also expected to be in the lower range. Thus, provided parts within the steam cycle are warm, the start-up times of diesel engines and combined cycles are similar. Since the start-up time is very much dependent on the initial material temperature, keeping parts within the steam cycle at a reasonable temperature in port is advisable. This can be done by employing supplementary fired HRSGs. That is, the HRSGs are provided with an additional burner that can be used either in combination with the gas turbines or alone. Such burner can be used in port for generating sufficient steam to satisfy the auxiliary power requirement when the gas turbines are taken out of operation. This avoids that parts within the steam cycle become cold, and at the same time the need for auxiliary machinery is eliminated.

F. Haglind / Energy Conversion and Management 49 (2008) 3468–3475

4. Summary and conclusions In this second part of the paper (consisting of three parts) previous work and experience are reviewed. Moreover, an overview of the implications of introducing combined cycles as prime movers is included. A few decades ago combined cycles of COGAS configuration were considered as an option for naval and commercial applications. These efforts were driven mainly by concerns of depletion of world petroleum reserves and the will to render the vessels more effective. More recently, a number of different studies on the use of combined cycle power plants of COGES configuration as prime movers have been published. One of the reasons for studying the COGES combined cycle was environmental concerns. Another application where gas turbine-based systems have been considered as prime movers is for LNG ships, which are traditionally powered by steam turbines with gas-burning boilers. Because of advantages in terms of economics (considering both first costs and operating costs) and redundancy, the COGES plant is believed to become a viable competitor to the steam plant for LNG ships. In terms of practical experience, two ship types have been powered by a combined cycle. GTS Millennium is a cruise ship which began commercial operation in year 2000. This ship and others of her class use a combined cycle power plant with a turbo-electric transmission, i.e. a COGES plant, consisting of two aero-derivative gas turbines, delivering 25 MW apiece, and one steam turbine, delivering 9 MW. Compared with a cruise ship with a conventional diesel-electric main engine, the Millennium ship produces significantly lower levels of pollutant emissions. In addition, with the COGES plant, the main engine mass and volume requirement are reduced, resulting in increased passenger capacity and reduced propulsion power requirement. The other example is two Russian 25-knot Ro-Ro vessels, which were powered by combined cycle power plants with direct power transmission (COGAS). These ships were built about 30 years ago. For a given power output, the weight and space requirement of gas turbines are much smaller than those of two-stroke diesel engines. Also, studies and practical experience for cruise ships and Ro-Ro ferries have shown that combined cycles offer improvements in these aspects. Smaller dimensions and weight of the main engine might increase the cargo/passenger capacity and/or reduce the propulsion power requirement. Evaluating the effects on voyage performance of changed volume and weight of prime movers for large ships is the subject of further work. Diesel engines respond faster to a load change than do combined cycles. The gas turbine(s), which is providing about twothirds of the power output of the combined cycle, reacts as quickly as the diesel engine, but the response time of the steam cycle is longer. However, owing to the large vessel inertia, particularly for large ships, engine acceleration time is generally not critical. The start-up time is very much dependent on whether engine parts are cold or warm before start-up, especially for the combined cycles. The gas turbine start-up time is independent of standstill time, while the start-up time for the steam cycle is dependent on the time required to heat up parts of the machinery without exceeding thermal stresses imposed by the material. Provided parts within the steam cycle are warm, the start-up times of diesel engine and combined cycles are similar – full power can be reached within about 1 h. Keeping parts within the steam cycle at a reasonable temperature in port can be done by employing supplementary fired HRSGs. That is, the HRSGs are provided with an additional burner that can be used either in combination with the gas turbines or alone. Such burner can be used in port for generating sufficient steam to satisfy the auxiliary power requirement when the gas turbines are taken

3475

out of operation. This avoids that parts within the steam cycle become cold, and at the same time the need for auxiliary machinery is eliminated. It is worth mentioning that such a system would need to be designed so that corrosion in the economiser is avoided, i.e. the stack temperature needs to be kept above the acid dew point. Acknowledgements Henri Doyer at AKER YARDS, France, is acknowledged for providing information on the Millennium cruise ship, and Niels Kjemtrup at MAN Diesel, Denmark, is thanked for the information on diesel engines. Kjeld Roar Jensen at FORCE Technology, Denmark, is thanked for having reviewed the manuscript and contributed with helpful comments. The funding from the Danish Center for Maritime Technology (DCMT) is acknowledged. References [1] Eyringer V, Köhler HW, Lauer A, Lemper B. Emissions from international shipping: 2. Impact of future technologies on scenarios until 2050. J Geophys Res 2005;110:D17306. [2] IMO. MARPOL, Consolidated Edition 2006. MARPOL Annex VI: regulations for the prevention of air pollution from ships. London: International Maritime Organization; 2006. [3] CEC. Directive 2005/33/EC of the European Parliament and of the Council of 6 July 2005 amending Directive 1999/32/EC. O J 2005; L191: 59–69. [4] CEC. Council Directive 1999/32/EC of 26 April 1999 relating to a reduction in the sulphur content of certain liquid fuels and amending Directive 93/12/EC. O J 1999; L121: 13–8. [5] Beicip-Franlab. Advice on the costs to fuel producers and price premia likely to result from a reduction in the level of sulphur in marine fuels marketed in the EU, European Commission – Study/C.1/01/2002, Contract ENV.C1/SER/2001/ 0063; 2002. [6] Horlock JH. Combined power plants, including combined cycle gas turbine (CCGT) plants. Malabar (FL): Krieger Publishing Company; 2002. [7] Mattson WS. Designing reliability and maintainability into the racer system. Nav Eng J 1983;95:202–13. [8] Brady EF, Dubios JP. Innovative concepts for naval ship systems. Nav Eng J 1982;94:214–31. [9] The Motor Ship. Alternatives to the diesel engine for marine propulsion. The Motor Ship 1982; 62(January): 49–50. [10] Brady EF. Energy conservation for propulsion of naval vessels. Nav Eng J 1981;93:131–44. [11] Merz CA, Pakula TJ. The design and operational characteristics of a combined cycle marine power plant. ASME paper 72-GT-90; 1972. [12] Mills RG. Greater ship capability and energy saving with combined-cycle machinery. Nav Eng J 1977;89:17–25. [13] Abott JW, McIntire JG, Rubis CJ. A dynamic analysis of a COGAS propulsion plant. Nav Eng J 1977;89:19–34. [14] Ahlqvist I. Increasing availability through introduction of redundancy. In: Papers and programme: electric propulsion, the effective solution. The Institute of Marine Engineers; 5–6 October 1995. [15] Shipping World and Shipbuilder, Gas turbine system integration. Ship World Shipbuild 2002; 203: 21–3. [16] Domachowski Z, Dzida M. An analysis of characteristics of ship gas turbine propulsion system (in the light of the requirements for ship operation in the Baltic Sea). Pol Marit Res 2004: 73–78 [special issue]. [17] Nurmi J. Are gas turbines an option for LNG tanker propulsion? The Naval Arch 1998; (February): 9–10. [18] Kalyanaraman K, Jeffs E. New applications for trent. Turbomach Int 2005;46:9–12. [19] MER – Marine Engineers Review. Next step in hybrid LNG propulsion? MER – Mar Eng Rev 2005; (November): 43–5. [20] Sannemann BN. Pioneering gas turbine–electric system in cruise ships: a performance update. Mar Technol 2004;41:161–6. [21] Tinsley D. Seminal gas turbine application arrives in Millennium cruise class COGES plant. Mar Propul Int 2000:14–5. [22] Kalosh A. Precision surgery. Seatrade Cruise Rev 2006;10:7–9. [23] Shipbuilding & Marine Engineering International. Russian COGAS in the Thames. Shipbuild Mar Eng Int 1980; 103: 233–7. [24] Dahlkvist J. Examination Project, ship design, combined cycle vs. diesel engine. Bachelor Thesis, Technical University of Denmark, Department of Mechanical Engineering; 2007. [25] Koehler H. Diesel engines and gas turbines for cruise vessels. Ship World Shipbuild 2001;202:14–7. [26] Kjemtrup N. MAN Diesel, Denmark. Private communication; 2007. [27] Walsh PP, Fletcher P. Gas turbine performance. London: Blackwell Science Ltd.; 1998. [28] Kehlhofer RH, Warner J, Nielsen H, Bachmann R. Combined-cycle gas & steam turbine power plants. 2nd ed. Tulsa (OK): PennWell; 1999.