Energy 142 (2018) 264e276

Contents lists available at ScienceDirect

Energy journal homepage: www.elsevier.com/locate/energy

Utilizing the scavenge air cooling in improving the performance of marine diesel engine waste heat recovery systems Mohamed T. Mito a, Mohamed A. Teamah a, b, Wael M. El-Maghlany b, *, Ali I. Shehata a a

Mechanical Engineering Department, College of Engineering and Technology, Arab Academy for Science, Technology and Maritime Transport, Abu-Qir, Alexandria, Egypt b Mechanical Engineering Department, Faculty of Engineering, Alexandria University, Egypt

a r t i c l e i n f o

a b s t r a c t

Article history: Received 19 April 2017 Received in revised form 19 September 2017 Accepted 9 October 2017 Available online 21 October 2017

This paper aims at improving power generation efﬁciency of marine diesel engine waste heat recovery systems. It presents a novel technique of integrating the heat rejected in the scavenge air cooling process and the exhaust gas in operating a single and dual pressure steam power generation cycles. Moreover, a thermodynamic analysis of proposed systems was performed to identify the optimum operating parameters for achieving an overall efﬁciency improvement. The analysis considered the exergy destruction in each component and the energy/exergy efﬁciencies. A performance analysis was conducted to assess applicability and power output at off design conditions. An evaluation of achieved improvements by suggested designs was presented from both an economical and environmental standpoint. In conclusion, results show that, the recommended cycle increased overall efﬁciency improvement from 2.8% for the conventional system to 5.1%, with an additional power output of 1210 kW, representing 9.7% of the engine's power. Also, exergy efﬁciency increased signiﬁcantly by 6.6% when using the presented system. Furthermore, the waste heat recovery system attained a reduction in fuel consumption of 1538 Ton/year, reducing carbon dioxide emission by 4790 Ton/year. © 2017 Elsevier Ltd. All rights reserved.

Keywords: Marine diesel engine WHR Ship power plant Rankine cycle Emission reduction Energy management

1. Introduction Maritime transport, an essential contributor to world trade, has risen over the last decades where total transported payload increased from 2.6 billion tons in 1970 to 9.5 billion tons in 2013 [1]. Lately, due to current climate change and CO2 emissions reduction efforts for environmental protection, a need emerged to enhance diesel engines power generation efﬁciency using waste heat recovery for energy utilization [1e3]. The IMO estimate the total maritime shipping CO2 emissions in 2012 at 938 million tons; expected to rise to 250% by 2050 [1]. On the other hand, the new ships' design efﬁciency has dropped 10% since 1990, due to less hydrodynamic hull design as a need to maximize cargo capacity [4]. Diesel engines currently dominate the ﬁeld of marine propulsion due to its high thermal efﬁciency compared to other prime movers as it can exceed 50% [5]. Different waste heat recovery techniques have been used, such as; turbocharging, turbo-compounding,

* Corresponding author. E-mail addresses: [email protected], [email protected] (W.M. ElMaghlany). https://doi.org/10.1016/j.energy.2017.10.039 0360-5442/© 2017 Elsevier Ltd. All rights reserved.

Brayton cycle, Rankine engine cycle and thermoelectric generators (TEG) [6,7]; such techniques have increased engine thermal efﬁciency from 2% to 20%, depending on system design, quality of energy recovery, component efﬁciency, and implementation [8]. Accordingly, the ideal technique available for emissions reduction is to utilize the waste heat from fuel combustion that nearly reaches 50% for additional power generation. Hence, an efﬁcient approach to increase power generation efﬁciency is through employing a power cycle driven by diesel engine waste heat. Main engine exhaust gas energy is the most attractive amongst waste heat sources due to its high mass ﬂow rate and high temperature. This will result in fuel consumption and CO2 emissions reduction [5,9]. The beneﬁts of applying waste heat recovery (WHR) systems are not only constrained to environmental beneﬁts, but also provide a commercial aspect for shipping companies constituting in a lower annual fuel bill as well as lower emission of CO2 and NOX which results in the ‘green’ ship image improving their position in freight competition [9,10]. The waste heat energy from a diesel engine mainly resides in the exhaust gas where, about 25.4% of fuel energy is still available for use [9]. However, due to adiabatic compression of the scavenge air

M.T. Mito et al. / Energy 142 (2018) 264e276

Nomenclature A C Cp df E_ e F g HV H h K l m Nuo P Pt Pr Q_ Re Rth S T Uo _ W

Surface area, m2 Cost, $ Speciﬁc heat capacity, kJ/kg C Fin diameter, m Exergy rate, kW Speciﬁc exergy, kJ/kg Correction factor Gravitational acceleration, m/s2 Heating value of fuel, kJ/kg C Heat transfer coefﬁcient, W/m2 C Speciﬁc enthalpy, kJ/kg Thermal conductivity, W/m C Fin length, m Mass ﬂow rate, kg/s Nusselt number Pressure, bar Tube pitch, m Prandtl number Heat transfer rate, kW Reynolds number Thermal resistance, C/W Fin spacing, m Temperature, C Overall heat transfer coefﬁcient, W/m2 C Power input or output, kW

Greek symbol d Fin thickness h Efﬁciency j Exergy efﬁciency l Exergy destruction ratio lc Taylor instability

in the turbocharger; an increase in temperature occurs. Hence, cooling the scavenge air is required to enhance the engine volumetric efﬁciency. The reduction of scavenge air temperature is done by the charge air cooler, where a signiﬁcant amount of fuel energy (nearly 14.1%) is wasted [9]. A promising contributor for waste heat recovery applications other than exhaust gas is heat available in the scavenge air which can be used to operate a Rankine cycle. The scavenge air offers nearly the same mass ﬂow rate as the exhaust gas and a relatively high temperature. Additionally, the exhaust gas has a minimum temperature that needs to be avoided; to evade its dew point temperature caused by high HFO sulfur content. The scavenge air is not associated with this technical problem, thus its temperature can be decreased without interrupting engine's performance, which offers full heat utilization and enhanced management of waste heat sources. Numerous scientiﬁc studies targeted improvement in power generation efﬁciency and emission reduction of diesel engines by mainly using waste heat from exhaust gas for additional power generation. In reference to these, different approaches and systems were considered for achieving such objective. Macian et al. [11] introduced a methodology for design optimization of WHR bottoming Rankine cycle from heavy-duty diesel engines. The methodology was based on considering different heating and working ﬂuids. Bonilla et al. [12] discussed the potential of different WHR technologies to use waste heat from industry located in Basque Country in Spain. Zheshu ma et al. [13] performed an analysis for

265

Subscripts 0 Dead state cogen Cogeneration cond Condenser dest Destruction e Exit eco Economizer engine Engine related parameter eva Evaporator f Fuel g Exhaust gas heat Heat added i Inlet OV Overall PP Pump Pinch Pinch point rej Heat rejected s Water-steam SG Steam generator SU Superheater Turb Turbine WH Waste heat Acronyms CEPCI Chemical engineering plant cost index HFO Heavy fuel oil HP High pressure IMO International Maritime Organization LP Low pressure ORC Organic Rankine cycle RPM Revolution per minute WHR Waste heat recovery WHRS Waste heat recovery system

WHR from the exhaust of a container ship's diesel engine at variable engine load and exhaust gas boiler operating pressure. Bari and Hossain [14] studied numerically the effect of using a shell and tube heat exchanger in WHR from a diesel engine using a CFD program. The effect of different parameters such as shell diameter, number of shells, and tubes diameter have been studied to obtain the optimum WHR unit performance. Butcher and Reddy [15] presented a performance analysis for a WHRS operating on a Rankine cycle. The main focus was on the effect of pinch point on the power output, second law efﬁciency and outlet temperature of exhaust gas. Various cogeneration schemes were studied in previous literature where Abusoglu and Kanoglu [16] performed a ﬁrst and second law performance analysis of a diesel engine powered cogeneration system. The study was based on characteristics of an existing diesel powered cogeneration plant in which thermodynamic performance and efﬁciency were studied at full engine-load range. Furthermore, Yilmaz [17] analyzed the performance of a cogeneration system operating on a reversible Carnot cycle while investigating variable operating conditions. It was concluded that all performance criteria increases when the ratio between temperature of heating ﬂuid and cooling ﬂuid increases. Different trigeneration systems were presented including a novel trigeneration scheme that was introduced by Mohan et al. [18]. Based on the exhaust supplied by the gas turbine operating AlHamra power plant. The tri-generation involved power generation by a steam Rankine cycle, water treatment by an air gap

266

M.T. Mito et al. / Energy 142 (2018) 264e276

membrane desalination and cooling by a single pressure absorption chiller. Choi and Kim [19] introduced a novel trilateral cycle for WHR from the diesel engine of a container ship. The system is composed of an exhaust gas boiler for producing saturated steam for heating applications, a steam Rankine cycle, and an organic Rankine cycle for power generation. The ideas handled in the previous review were conﬁned to using exhaust gas in operating diesel engine WHRS, while analyzing the effects of different operating parameters on initial design or on system performance. The scavenge air was not used as a main heating ﬂuid despite its high waste heat content and better waste heat utilization when compared to exhaust gas. The current study innovation will rely on three main points; ﬁrstly, this paper will present a comparative study for different cycle conﬁgurations of WHR systems performing on a Rankine cycle, aiming at achieving higher energy availability, while introducing the use of scavenge air as a main heating ﬂuid. As a whole, the scavenge air role will be established in improving power generation efﬁciency and WHRS performance. Secondly, the integration between scavenge air and exhaust gas will be introduced in operating a single and dual pressure WHRS. The dual pressure system was not introduced previously for marine WHRS despite that it offers the operation at higher pressures while allowing full use of available waste heat. Thirdly, the economic and environmental impacts shall be considered for achieving reduction of both fuel consumption and carbon footprint. 2. System description 2.1. Speciﬁed diesel engine The diesel engine used is a HYUNDAI-WARSTILA 6RT-ﬂex58T-E with maximum output power of 13.94 MW. The two-stroke marine diesel engine includes a constant pressure turbocharging. The design of the systems provided in this study will be based on the engine's continuous service rating, presented in Table 1. The shop test provides the scavenge air and exhaust gas temperature and mass ﬂow rate at different operating loads, as shown in Table 2. 2.2. Proposed models Model I considers a simple Rankine cycle conventional WHRS, depending primarily on diesel engine exhaust gases for generating superheated steam. This will enable a comparison with the conventional approach tackled by previous literature, while using operating preferences of the speciﬁed Diesel engine. Also, model I will help in scoping improvements achieved by cycle conﬁgurations demonstrated in this study. The schematic diagram for model I is presented in Fig. 1. Model II integrates the exhaust gas and scavenge air to operate a single pressure Rankine cycle. The design consists of two boilers, one for generating dry saturated steam by means of heat available from scavenge air called ‘air boiler’ and the other for generating superheated steam by means of exhaust gas, called ‘gas boiler’. The Table 1 Operational characteristics of the diesel engine at the continuous service rating. Output power Scavenge air Exhaust gas Fuel (HFO)

Mass ﬂow rate Temperature Mass ﬂow rate Temperature Mass ﬂow rate Heating value

12.456 28.2 213 28.8 251 0.61 38.76

MW kg/s C kg/s C kg/s MJ/kg

gas boiler consists of an evaporator for producing dry saturated steam and a superheater that superheats dry steam from air boiler and gas boiler. The cycle also contains a feed water heater that acts as an economizer for both boilers where it preheats the feed water before entering the boilers by using scavenge air exiting the air boiler. Also minimizing the thermal effect in the exhaust gas boiler, which is the same function of regeneration but without power loss that accompanies steam extraction. The main advantage of using the scavenge air in preheating is that its temperature can be lowered without the constraint of the dew point temperature and risk of sulfuric acid formation as for the exhaust gas. The presented cycle is set to operate at a single operating pressure, where dry saturated steam produced from the air boiler is mixed with dry saturated steam produced from the gas boiler, to be eventually superheated in the gas boiler before its admission into the turbine. The schematic diagram for model II is described in Fig. 2. Model III purposefully aims to introduce a dual pressure cycle to the proposed system in order to beneﬁt from each energy source individually for maximizing waste heat utilization. The main difference to model II is the addition of a superheater within the air boiler allowing generation of superheated steam and operation of both boilers at two different pressures owing to the difference in heat available from scavenge air and exhaust gas. The feed heater will speciﬁcally serve the gas boiler in increasing its ﬂow rate, whereas the exhaust heat that would have been needed to preheat the feed water will be used for evaporation. Fig. 3 presents the schematic diagram for model III. Before initiating the thermodynamic analysis, some design considerations were assumed for all models, presented in Table 3. 3. Thermodynamic analysis In this study, the steady-state physical and thermodynamics equations were used based on the ﬁrst and second thermodynamic laws. The steady state equations are applicable during the ship's voyage. However, they were excluded at heavy sea conditions, bad weather, maneuvering, and docking. The thermodynamic analysis was founded on the pinch point temperature difference as a dominant parameter governing the performance of any WHR boiler. It deﬁnes the WHRS optimum operating pressure based on heating ﬂuid temperature. The WHRS designs proposed in this study will be based on minimum pinch point temperature difference of 15 [5]. The following assumptions are considered throughout the study: The system is assumed to be at steady state. No pressure drops on steam side or heating ﬂuid side in the heat exchanger. Approach point is negligible. The dead state condition is assumed at T0 ¼ 20 C and P0 ¼ 100 kPa [22]. The feed pump compression process is assumed isentropic, thus, the pump will not be considered in the exergy analysis [23].

3.1. Energy analysis The governing equations used for designing WHRS are energy balance equations based on the ﬁrst thermodynamics law for energy conversion while neglecting kinetic and potential energy changes. The conservation of energy equation is represented in equation (1) as follows:

M.T. Mito et al. / Energy 142 (2018) 264e276

267

Table 2 Temperature and mass ﬂow rate distribution of exhaust gas and scavenge air at different operating loads. Load (%)

25

50

75

90

100

Engine speed (rpm)

65.9

83

95

101

104.6

7.1 290 6.9 52

15.1 262 14.8 134

24.1 249 23.5 190

28.8 251 28.2 213

32.7 261 31.9 230

Mass ﬂow rate (kg/s) Temperature ( C) Mass ﬂow rate (kg/s) Temperature ( C)

Exhaust gas Scavenge air

Fig. 1. Schematic diagram for model I. Fig. 3. Schematic diagram for model III.

maximize work output. Consequently, the exergy will represent the useful work potential of the system at the speciﬁed conditions [24,25].

3.2.1. Formulations for the exergy analysis The exergy balance for a control volume undergoing steady state process with negligible changes in kinetic and potential energy is represented as:

_ ¼ E_ heat ±W

Fig. 2. Schematic diagram model II.

_ ¼ Q_ ±W

X

_ e he m

X

_ i hi m

(1)

_ are the net heat and power input, m _ is the mass where Q_ and W ﬂow rate, h is the speciﬁc enthalpy and the subscripts i and e stands for the inlet and exit states, respectively [24]. The energy balance equations for model I are presented in Table 4, whereas for models II and III same equations are applied to each component in a similar manner. 3.2. Exergy analysis An exergy analysis based on the second law of thermodynamics will be performed to assess WHRS performance with varying engine load. The exergy analysis implies that the system will be compared from an initial speciﬁed state to the dead state, where at the end of the process, the system should reach dead state to

X

_ e ee m

X

_ i ei þ E_ dest m

(2)

where E_ heat is the exergy transferred to the system at a speciﬁed temperature, E_ dest is the rate of exergy destruction due to irreversibility and W is the exergy change as a result of work input or output to system [26]. Additionally, e and E_ are the speciﬁc ﬂow exergy and the rate of total exergy which are presented by equations (3) and (4) respectively, where h is the speciﬁc enthalpy, s is the speciﬁc entropy and the subscript 0 indicate the dead state [22].

e ¼ ðh h0 Þ T0 ðs s0 Þ

(3)

_ E_ ¼ me

(4)

The exergy transferred E_ heat associated with heat transfer at a speciﬁed temperature T is presented in equation (5). Where Q_ is the heat transfer rate and T0 is the temperature of the dead state [26].

E_ heat ¼

X

1

T0 _ Q T

(5)

The exergy destruction due to irreversibilities within the system E_ dest is given by equation (6), where the total exergy destruction is calculated for the whole system while considering exergy destruction in each component of the cycle [22].

268

M.T. Mito et al. / Energy 142 (2018) 264e276

Table 3 Design considerations assumed for all models. Parameter

Value

Cooling water temperature at ISO condition Temperature difference between the saturation temperature of steam and inlet of cooling water Steam saturation Temperature at condenser pressure Condenser pressure, PCond Minimum exhaust gas temperature based on the dew point of the exhaust gas produced by burning HFO Superheater approach temperature Minimum degree of superheat

25 C 15 C 40 C 0.07 bar 150 C 20 C 20 C

ef ¼ 1:065 HV

Table 4 Energy balance equations for model I.

Evaporator

_ s ðhs;4 hs;1 Þ ¼ m _ g cp;g ðTg;1 Tg;4 Þ Q_ ov ¼ m _ s ðhs;2 hs;1 Þ ¼ m _ g cp;g ðTg;3 Tg;4 Þ Q_ eco ¼ m _ s ðh h Þ ¼ m _ g cp;g ðT T Þ Q_ ¼m

Superheater

_ s ðhs;4 hs;3 Þ ¼ m _ g cp;g ðTg;1 Tg;2 Þ Q_ su ¼ m

Overall energy balance Economizer

eva

s;3

s;2

g;2

g;3

[5] [20]

[5] [21] [21]

(11)

The energy and exergy balance equations for the system's components are presented in Table 5, where subscripts i and e indicate the inlet and exit states for a ﬂuid stream at each component, respectively. 3.3. Cost analysis

E_ dest ¼

X

_ i ei m

X

_ e ee þ m

X T _ 1 0 Q_ ±W T

(6)

The exergy destruction ratio l, presented in equation (7), will be considered to assess the ratio of energy lost and irreversibilities caused by each component with regards to total exergy destruction. Another important parameter in performing an exergy analysis is the exergy efﬁciency j, which represents the ratio between exergy of the WHRS E_ WHRS and exergy input to the system E_ heat associated with heat transfer at a speciﬁed heating temperature T. The exergy efﬁciency is represented in equation (8).

l¼

E_ dest;cmponent E_

(7)

dest;total

E_

j ¼ _WHRS Eheat

(8)

The ﬁrst law efﬁciency of WHRS is a ratio between total power output and heat added through fuel burned. The total power output is presented by engine power and additional WHRS power. However, the second law efﬁciency for a cogeneration system is a comparison between whole system exergy rate and fuel exergy. The formulas of both ﬁrst and second law efﬁciencies are presented by equations (9) and (10), respectively [26].

hcogen;I

_ _ W engine þ WWHRS ¼ _ f HV m

hcogen;II ¼

(9)

_ _ W engine þ EWHRS _E f

_ _ s ððhe hi Þ T0 ðse si ÞÞ W engine þ m ¼ _ f ef m

(10)

. Brzustowski and Brena [27] discussed the ratio ef HV for different hydrocarbons and their study concluded a constant value of 1.065 for the ratio between fuel exergy and fuel heating value. The authors also added that; previous ratio is applicable to heavy hydrocarbons with undetermined composition, which applies to HFO used onboard ships and in this study. Therefore the formula in equation (11) will be used to calculate fuel exergy ef .

An economic analysis will be performed to compare between presented models regarding their initial cost. The steam generator will be the only component considered in the economic analysis as it is the main point of difference between each model. The economic analysis includes estimating the initial cost of the proposed systems and reduction in fuel consumption achieved by the additional power of the WHRS. 3.3.1. Heat transfer modeling The steam generator initial cost depends mainly on its surface area. Thus, the steam generator surface area is calculated for each model. The heat transfer modeling is performed regarding shell and tube heat exchanger based on cross ﬂow between the heating and working ﬂuids. The thermal load for each heat exchanger is used to calculate the required heat transfer surface area from equation (12), where Uo is the overall heat transfer coefﬁcient, DTLM is the logarithmic mean temperature difference between the heating and working ﬂuids across the heat exchangers and Ao is the required surface area. In equation (12), the logarithmic mean temperature difference is multiplied by a correction factor F to compensate for the cross ﬂow arrangement [24]. The value of the correction factor is unity for the boiling ﬂow in evaporators. However, the correction factor is calculated for the economizer and superheater for single pass cross ﬂow exchanger where both ﬂuids are unmixed [28].

Q Ex ¼ Uo Ao FDTLM

(12)

The modules suggested involve radial ﬁns on the heating ﬂuid side to increase the surface area per cubic meter of exchanger volume [29]. The heat transfer coefﬁcient across the ﬁn and the ﬁn material thermal conductivity are assumed constant. The ﬁn temperature is assumed to vary in one dimension and any variation among other dimensions is neglected. This assumption is valid for the case of thin ﬁns where studies have shown that the error involved with the assumption of one dimensional heat transfer is less than 1% when [28]:

Hd < 0:2 k

(13)

3.3.1.1. Heat transfer correlations on the heating ﬂuid side. The heating ﬂuid heat transfer coefﬁcient will be dependent on the ﬂow characteristics around the radially ﬁnned tubes. The average Nusselt

M.T. Mito et al. / Energy 142 (2018) 264e276

269

Table 5 The energy and exergy balance equations for the WHRS. Component

Energy analysis

Exergy analysis

Heat exchanger

_ e hi Þ Q_ SG ¼ mðh

_ i he Þ mT _ 0 ðsi se Þ þ Q_ WH 1 TT0 E_ dest; SG ¼ mðh

Turbine

_ _ i he Þ W Turb ¼ mðh _ i he Þ Q_ cond ¼ mðh

_ 0 ðse si Þ E_ dest; Turb ¼ mT

Condenser Pump

_ PP ¼ mðh _ e hi Þ W

_ 0 ðse si Þ E_ dest; PP ¼ mT

_ i he Þ mT _ 0 ðsi se Þ þ Q_ rej 1 TT0 E_ dest; Cond ¼ mðh

number for low radial ﬁnned tube cross ﬂow heat exchangers can be calculated from the following empirical correlation [29]:

Nuo ¼ 0:183Re

0:7

s0:36 P t l df

!0:06

l df

!0:11 pr0:36

(14)

The above mentioned correlation is applicable for 103 Re 8103 , 0:19 sl 0:66, 1:1 dpt 4:92, f l 0:058 d 0:201. The outside heat transfer is thus calculated f from equation (15) as a function of Nusselt number [30].

Ho ¼

Nuo ko do

(15)

1 Uo Ao ¼ P Rth

NUi ¼ 0:023Re0:8 pr0:4

(16)

The inside heat transfer coefﬁcient in the evaporator is based on the evaporation of the saturated feed water circulating the riser. This particular case involves speciﬁcally ﬁlm boiling heat transfer correlations [31]. Stable ﬁlm boiling can be solved analytically because the ﬂow pattern is relatively simple where an exact solution can be obtained. The following equation considers the interfacial shear and curvative effects in calculating the heat transfer coefﬁcient (micro and macro convection) [32].

0:59 þ 0:069

3.3.2. Capital cost estimation Cost evaluation equations are widely used for estimating the initial cost of thermal equipment for the preliminary design stage. The module cost for shell and tube heat exchangers is given by equation (20) [33].

#0:75 " 3 kv g hfg ðrl rv Þrv di mi ðTw Tsat Þlc

lc

logCp ¼ K1 þ K2 logðASG Þ þ K3 ðlogðASG ÞÞ2

logFp ¼ C1 þ C2 logð10P 1Þ þ C3 ðlogð10P 1ÞÞ2

3.4. Solution procedure

gc s gðrl rv Þ

(18)

The overall heat transfer coefﬁcient for each section of the steam Table 6 Economic cost parameters of the WHRS [37]. K1

K2

K3

B1

B2

4.3247

0.3030

0.1634

1.63

1.66

C1

C2

C3

Fm

0.0388

0.11272

0.08183

1.4

(22)

where C1 , C2 and C3 are the coefﬁcients of pressure factor which are stated in Table 6 [33]. The estimated purchased cost is calculated from equation (23) while considering the chemical engineering plant cost index CEPCI. CEPCI reﬂects the effect of time on the purchased equipment cost [34,35].

The minimum wavelength for Taylor instability, lc , is given by equation (18), where s is the surface tension of the working ﬂuid (N/m). Also, rl and rv are the working ﬂuid densities at the liquid and vapor state. Additionally, gc is the proportionality constant in Newton's second law of motion which equals unity for SI units [32].

0:5

(21)

where K1 , K2 and K3 are the coefﬁcients of equipment costs which are shown in Table 6. Furthermore, the pressure factor Fp for the heat exchangers can be calculated from equation (22) [33].

Cestimated ¼ ðCSG Þ2001

(20)

where Fm is the material factor and B1 and B2 are constants given in Table 6. Furthermore, the purchased cost of equipment when using carbon steel construction Cp , can be calculated from equation (21) [33].

(17)

lc ¼ 2p

(19)

CSG ¼ Cp B1 þ B2 Fm Fp

3.3.1.2. Heat transfer correlations on the working ﬂuid side. There are two ﬂow regimes for the working ﬂuid through the steam generator which are single phase ﬂow through the economizer and superheater and two phase boiling ﬂow through the evaporator. Equation (16) represents a correlation for calculating the Nusselt number for a single phase ﬂuid ﬂowing inside tubes [30].

Hi ¼

generator is deﬁned based on the electrical analogy of the heat transfer resistances, as shown in equation (19) [30].

CEPCI2014 CEPCI2001

(23)

The thermodynamic models presented in this study are formulated by using a series of iterative procedures developed on an in-house MATLAB program. The program solves governing equations, while maintaining design constraints held throughout this study. The properties of water/steam were obtained from the National Institute of Standards and Technology (NIST) database REFPROP 8.0 [36]. The solution procedure used in MATLAB is presented in Fig. 4. 4. Results & discussion 4.1. Design investigation based on the operating pressure Firstly, the effect of different operating pressures on the output

270

M.T. Mito et al. / Energy 142 (2018) 264e276

power will be studied while maintaining design criteria to determine the optimum pressure for utilizing waste heat and satisfying proper system operation. The solution is achieved by iterating the boiler pressure while maintaining a minimum pinch point of 15 C [5] between heating ﬂuids and feed water at the evaporator inlet. A common relation trend between operating pressure and output power is presented in Fig. 5 (a, b), which is the increase in power while increasing operating pressure until reaching a maximum value. Whereas the pressure increases past this value a uniform decrease in output power occurs. The reason for this trend is the pinch point temperature, where at low pressures the saturation temperature is low and difference in temperature between heating ﬂuid and steam is high, allowing for full use of waste heat. At higher pressure, the pinch point decreases past the set value due to increasing saturation temperature. As a result, the mass ﬂow rate of steam produced decreases; to decrease heat transferred and increase the temperature of heating ﬂuid at the boiler exit for maintaining a high pinch point. This reduces output power disregarding the increase in enthalpy drop achieved in the turbine. As a result of increasing the heating ﬂuid temperature at the economizer exit, less heat was transferred in the evaporator and superheater, thus, decreasing the superheated steam yield of the steam generator as presented in Fig. 6 (a, b). Subsequently, this led to a decrease in output power. Therefore, the relation between mass ﬂow rate of steam and operating pressure is affected mainly by waste heat source temperature and pinch point. In Fig. 6 (a), the steam mass ﬂow rate for model I is nearly constant until an operating pressure of 5 bar because the pinch point is at design value and consequently heat added by exhaust is constant. However, for

Fig. 4. Flow diagram describing the solution procedure.

model II, its mass ﬂow rate decreases constantly with increasing pressure; this is caused by scavenge air low temperature in which the pinch point is less than design value and heat added by scavenge air decreases with increasing pressure. The solution procedure also emphasized the importance of the exhaust steam dryness fraction where it is an important factor in determining both optimum operating pressure and presented cycle applicability. According to the author's knowledge, the dryness fraction was not discussed in previous literature. As shown in Fig. 7 (a, b), at constant superheated steam temperature, the dryness fraction decreased at higher pressures owing to water properties and change in the degree of superheat where water is known to be a wet ﬂuid as it has a tendency to prematurely condense during expansion owing to its bell-shaped temperature-entropy diagram [37]. The low dryness fraction is associated with low temperature of exhaust gas and scavenge air supplied by the engine, where the superheated steam temperature is set based on the heating ﬂuids temperature. The optimum operating pressure was selected based on the results of the MATLAB program where deﬁning the optimum pressure regarded output power, a dryness fraction over 0.88 [20] and a temperature distribution which utilized the full capacity of waste heat offered by the diesel engine while maintaining boiler lifetime. Even though the selected operating pressures do not yield maximum power, they still satisfy design condition. The optimum operating conditions for the three models at the continuous service rating are presented in Table 7. 4.2. Exergy destruction for each model The exergy analysis implies calculating the exergy destruction for each model with varying operating pressure to determine the rate of potential energy lost and the utilization of available energy. Fig. 8 (a, b & c) presents the variation in exergy destruction with changing operating pressure for all presented models. The rate of exergy destruction for the steam generator is clearly the highest among all components due to the boiling process. Additionally, it decreases uniformly for all models with increasing pressure, where the latent heat of vaporization decreases when operating at higher pressures, this allows for lower irreversibilities and higher energy utilization. Furthermore, the condenser exergy destruction is prominently dependent on the steam mass ﬂow rate, due to small difference between temperatures of cooling water and surrounding. This gives a low signiﬁcance to the Carnot efﬁciency term; used to calculate exergy transferred accompanying heat rejected from the condenser. The exergy destruction caused by the turbine is signiﬁcantly lower compared to other components where it reﬂects the turbine isentropic efﬁciency. On the other hand, in Fig. 8, the total exergy destruction varied between all three models where the lowest exergy destruction was achieved by model I as it requires the least heat input and achieves the lowest steam mass ﬂow rate. As for models II and III, there is a noticeable increase in total exergy destruction from model I, where models II and III dictate more exergy transfer due to increased heat input as a result of integrating heat available from exhaust gas and scavenge air. However, when dealing with WHRS the total exergy destruction indicates lower utilization of previously wasted heat. But as shown by Fig. 5, operating at higher pressures leads to a decrease in power output as a result of the decreasing pinch point and steam mass ﬂow rate. The results of the exergy analysis at the continuous service rating are presented in Table 8. The steam generator exergy destruction ratio is highest for model I reaching 64.4% as a result of high exhaust exit temperature. This indicates high exergy destruction related to available energy and low energy utilization

M.T. Mito et al. / Energy 142 (2018) 264e276

271

(a)

(b)

Fig. 5. Effect of operating pressure on the output power.

(a)

(b)

Fig. 6. Effect of operating pressure on the superheated steam yield.

(a)

(b)

Fig. 7. Effect of operating pressure on the dryness fraction.

Table 7 Operation characteristics for each model at the optimum operating pressure. Model

Operating pressure (bar)

Steam yield (kg/s)

Heating ﬂuid exit temperature ( C)

Dryness fraction

I II

3 3

1.16 2.08

150 150

0.894 0.89

III

Air boiler

Gas boiler

2.25

Scavenge air

Exhaust gas

Air boiler

Gas boiler

2

3

109.2

150

0.898

0.894

272

M.T. Mito et al. / Energy 142 (2018) 264e276

Fig. 8. Effect of operating pressure on the exergy destruction for a) model I, b) model II and c) model III.

when compared to other models. As for the exergy efﬁciency, model I achieved the lowest exergy efﬁciency of 57.8% as a result of high exhaust gas exit temperature, caused by the high dew point temperature of the exhaust produced by burning HFO. However, the exergy efﬁciency increased signiﬁcantly to 64.4% and 62.4% for models II and III respectively. The reason for this improvement is the decrease in heating ﬂuids exit temperature which indicates a better waste heat utilization in case of models II and III. In conclusion, although models II and III show higher total exergy destruction when compared to model I, however, this is due to the increase in energy input to the system as a result of using multiple heating ﬂuids. Nonetheless, when considering the exergy

efﬁciency, models II and III dictate better energy management when compared to model I. 4.3. First and second law efﬁciencies The overall improvement of WHRS operating efﬁciency was taken into consideration regarding the ﬁrst and second law efﬁciencies, as shown in Fig. 9 (a, b). The ﬁrst and second law efﬁciencies are directly proportional to the superheated steam yield achieved by each model which is dependent on the operating pressure and pinch point at this pressure. The values of the second law efﬁciency are lower than the energy efﬁciency, where exergy

M.T. Mito et al. / Energy 142 (2018) 264e276

utilization.

Table 8 Results of the exergy analysis at the continuous service rating. Model

Component

E_ dest (kW)

l (%)

E_ dest;total (kW)

j (%)

I

Steam generator Condenser Turbine Steam generator Condenser Turbine Steam generator Condenser LP Turbine HP Turbine

352.7 162.42 33.1 469 292.6 59.6 459.5 315.7 22.9 37.7

64.4 29.6 6 57.1 35.6 7.3 55 37.8 2.7 4.5

548.2

57.8

II

III

821.2

64.4

835.2

62.4

efﬁciency also considers system irreversibilities. 4.4. Comparison between the suggested models A comparison between suggested models was made in terms of output power, superheated steam yield and improvement in system efﬁciency at engine continuous service rating. 4.4.1. Evaluating the power output The proposed systems power outputs are compared in Table 9 at engine continuous service rating. Regarding model I, it generated the least power at 672 kW. As for models II and III, they generated roughly the same power output at 1210 kW and 1230 kW respectively. This proves that the dual pressure cycle did not yield a signiﬁcant improvement in output power, as the pressure difference should be more distinctive to yield the increase in enthalpy drop achieved in the turbine while operating at higher pressures. Due to low temperature of waste heat available; pressure levels of the dual pressure system cannot be increased signiﬁcantly to improve power output. 4.4.2. Evaluating the superheated steam yield The proposed systems steam yield is compared at engine continuous service rating. The results, presented in Table 9, conform to the power output of each system, as the steam yield also depends on heat available from the waste heat source used. The system relying on exhaust gas produced 1.16 kg/s, while there was a signiﬁcant increase in steam yield for model II which produced 2.08 kg/s of steam owing to addition of heat available by scavenge air. Model III produced the largest superheated steam yield at 2.25 kg/s mainly due to the dual pressure conﬁguration where the air boiler operated at low pressure allowing for more waste heat

(a)

273

4.5. Performance analysis The suggested WHRS performance was investigated to study the effect of changing engine load on functionality and output power for each system. The performance analysis will be very beneﬁcial in reviewing the behavior of each system at real operating conditions which include full load operation at open seas and lower engine loads at docking or maneuvering. The WHRS operation depends mainly on temperature and mass ﬂow rate of exhaust gas and scavenge air, which are dependent on engine operating condition. The performance analysis for each model is presented in Fig. 10. 4.5.1. Model I The proposed system in model I operates at full range of engine operation due to high exhaust gas temperature. Additionally, the power output changes regarding the change in mass ﬂow rate of the engine's exhaust with respect to its operating load. At 25% engine load, the cycle output power is 228 kW, which is 6.5% of the engine's power at mentioned load. Furthermore, at 100% load, the exhaust gas temperature is 261 C, while the mass ﬂow rate of exhaust is 32.66 kg/s. Consequently, the WHRS output power is 837 kW at full engine load which is 6% of the engine's output power. This settles that, the power generated by the conventional WHRS proposed in model I is useful at different operating conditions including docking, maneuvering, and open seas. 4.5.2. Model II During the operation at loads lower than 75% which include docking and maneuvering of the ship, the scavenge air would not be sufﬁcient for producing saturated steam at the speciﬁed pressure. At this condition, the scavenge air will be only used for heating the feed water used in the exhaust gas boiler. The function of feed water heater arises at this point because it will be used to supply heat to feed water entering the gas boiler at low operating loads, as the air boiler will be bypassed due to low scavenge air temperature. 4.5.3. Model III At engine loads less than 90%, the scavenge air would not be sufﬁcient for generating superheated steam at 193 C and the lowpressure boiler would not be operational. Thus, scavenge air waste heat will be used for feed water heating in the feed heater to increase the steam yield generated by the high-pressure exhaust gas boiler.

(b)

Fig. 9. Effect of operating pressure on the a) ﬁrst and b) second law efﬁciencies of the whole system.

274

M.T. Mito et al. / Energy 142 (2018) 264e276

Table 9 Comparison between models at the continuous service rating. Model

Superheated Steam yield (kg/s)

WHRS Power (kW)

WHRS Power (As a percentage from the engine power)

I II III

1.16 2.08 2.25

672 1210 1230

5.4% 9.7% 9.9%

Table 10 Comparison between the fuel saving for each model. Model

Fuel saving (Ton/year)

I II III

854 1538 1564

Model II Model II Model I $-

$100 $200 $300 $400 Cost of fuel saved (x103 $/year)

Fig. 10. Performance curves for each model at variable engine load. Fig. 11. Cost of fuel saved per year.

By reviewing the performance analysis, it is clear that; models II and III offer higher power output than model I as a result of using the scavenge air alongside the exhaust gas. This setup offers a consistent power output due to use of the exhaust gas and an overall higher power than model I due to use of the scavenge air in preheating the feed water at low loads. When comparing models II and III, model II offers higher functionality and power output than model III at loads less than 90%, as a result of using the scavenge air in producing saturated steam, in model II, which is more suitable based on scavenge air temperature. 4.6. Economical aspect of the WHRS The economic analysis for the proposed WHRS is presented in the form of reduction in fuel consumption and fuel cost per year as a result of the additional WHRS power. The reduction in engine running cost is considered extremely beneﬁcial for shipping companies as it translates to a reduction in shipping fees and thus improving their position in freight competition. The speciﬁc fuel consumption is calculated using equation (24) by considering the engine power and the additional WHRS power. The reduction in speciﬁc fuel consumption represents the amount of fuel saved when using the WHRS. The WHRS is assumed to operate for 300 days per year at the continuous service rating [3].

S:F:C ¼

mfuel _ þW

_ W engine

(24)

WHRS

The results presented in Table 10 indicate that; the WHRS operating on the diesel engine exhaust gas (model I) achieved a noticeable reduction in fuel consumption of 854 Ton/year by generating additional power from the WHRS. This amount translates to 5.4% of HFO used daily by the engine. The integration of exhaust gas and scavenge air achieved a signiﬁcant reduction in

fuel consumption of 1538 Ton/year and 1563 Ton/year for the single and dual pressure systems respectively, this is equivalent to 9.7% and 9.9% of the engine's consumption. The reduction in fuel consumption achieved by applying the WHRS can be translated to a reduction in engine running cost. According to the French National Institute of Statistics and Economic Studies (INSEE), the price of HFO in May 2016 was 243 $/Ton. Consequently, the HFO cost for running the engine 300 days/year at the continuous service rating can be estimated to nearly 3.84 million dollars [38]. Fig. 11 displays the cost of fuel saved for each model where model I achieved a reduction of approximately 208 103 $/year, while model II reduced fuel cost signiﬁcantly by nearly 374 103 $/year. The results displayed in Fig. 11 indicates a signiﬁcant economic value of using the scavenge air alongside the exhaust gas where model II achieved an 80% higher reduction in fuel cost compared to the conventional system (model I).

Model III Model II Model I 0.00

0.50 1.00 1.50 Cost of steam generator (x106 $)

Fig. 12. Initial cost of the steam generator.

2.00

M.T. Mito et al. / Energy 142 (2018) 264e276

Model III Model II Model I 0

1000 2000 3000 4000 5000 6000 CO2 emission reduction (Ton/year) Fig. 13. CO2 emission reduction.

The feasibility of the WHRS system can be determined by comparing the initial cost and the fuel consumption reduction for each model. The initial cost of the steam generator for each model is presented in Fig. 12, where model I offered the lowest initial cost of 0.83 106 $ as a result of the low steam output, thus, demanding smaller equipment. Nonetheless, model I offered the lowest reduction in fuel consumption. Additionally, model III had the highest initial cost of 1.54 106 $ due to the complexity of the system and implementation an additional superheater for the air boiler. 4.7. Reduction in CO2 emission The environmental impact of the proposed WHRS is presented in the CO2 emission reduction that resulted from the additional power cycle performing alongside the Diesel engine. The WHRS output power was generated without additional fuel consumption and thus, a reduction in CO2 emission was achieved. The speciﬁc CO2 emission for burning HFO is 3.114 kg-CO2/kg-fuel [3]. Consequently, the reduction in CO2 emission is directly proportional to fuel saving achieved by the proposed WHR systems. As presented in Fig. 13, the conventional system operating on exhaust gas as heating ﬂuid (model I) accomplished a reduction in CO2 emissions of 2660 Ton/year, while the proposed systems utilizing heat available in both exhaust and scavenge air achieved 4790 Ton/year and 4870 Ton/year for the single and dual pressure systems respectively. Results indicate that models II and III are superior to model I due to the fact that more heat was successfully utilized when using the suggested systems. 5. Conclusion This study was set to investigate the improvements applicable to WHR systems involving marine diesel engines. The WHR method chosen is the Rankine cycle in which performance improvements were achieved by designing a system that integrates different sources of waste heat. A thermodynamic analysis was performed for all three suggested models based on energy and exergy balance equations. All models, showed decrease in output power and steam yield with increasing pressure caused by maintaining a constant pinch point at increasing saturation temperature. The integration of scavenge air and exhaust gas yielded the highest power outputs of 1210 kW and 1230 kW, equivalent to 9.7% and 9.9% of engine's power for single and dual pressure cycles respectively. Model I achieves the lowest improvement in ﬁrst law efﬁciency of 2.8%, while models II and III improved the efﬁciency by 5.1% and 5.2% respectively.

275

The exergy analysis showed that models II and III imply higher total exergy destruction when compared to model I, as a result of increasing energy input to the system when using multiple heating ﬂuids. However, when considering exergy efﬁciency, it increased signiﬁcantly from 57.8% for the conventional system to 64.4% and 62.4% for models II and III respectively. Thus, models II and III present better energy management when compared to model I. The performance analysis shows that model I is functional at full range of engine load with variable steam yield, owing to high exhaust gas temperature. However, the scavenge air temperature for models II and III varied with the engine load where at certain loads it would not be sufﬁcient for generating saturated steam. Consequently, the scavenge air is used for preheating the feed water before entering the exhaust boiler for achieving higher steam yield at loads less than 75% and 90% for models II and III respectively. As a result, models II and III generate more power than model I at overall engine load range. Model I achieved the lowest saving in fuel consumption of 5.4% (854 Ton/year), on the other hand a signiﬁcant reduction in fuel consumption was achieved by models II and III of 9.7% (1538 Ton/year) and 9.9% (1563 Ton/year) respectively. Similarly, the CO2 emission reduction is the lowest for model I, where models II and III reduced CO2 emission by 4790 Ton/year and 4870 Ton/ year respectively. In conclusion, the recommended system for implementation in ships design is model II, which relies on the scavenge air and exhaust gas in operating a single pressure Rankine cycle while incorporating a common economizer and superheater. Model II will offer lower initial and maintenance costs and less complexity than the dual pressure cycle in model III with a difference of 0.1% improvement in efﬁciency. Meanwhile, model II achieves an efﬁciency improvement of 5.1% and fuel consumption reduction of 9.7%. Acknowledgements Financial support granted from the Arab Academy for Science, Technology and Maritime Transport for this research is greatly appreciated. References [1] IMO. Third IMO greenhouse gas study 2014. International Maritime Organization; 2015. [2] Gençsü I, Hino M. Raising ambition to reduce international aviation and maritime emissions. In: Contributing paper for Seizing the global opportunity: partnerships for better growth and a better climate. London and Washington, DC: New Climate Economy; 2015. Available at: http://newclimateeconomy. report/misc/working-papers. [3] Min-Hsiung Yang. Thermal and economic analyses of a compact waste heat recovering system for the marine diesel engine using transcritical Rankine cycle, vol. 106; 2015. p. 1082e96. [4] Faber J, Hoen MT. Historical trends in ship design efﬁciency. In: Seas at risk and transport & environment; 2015. [5] Grljusi c M, Medica V, Ra ci c N. Thermodynamic analysis of a ship power plant operating with waste heat recovery through combined heat and power production. Energies 2014;7:7368e94. [6] Edwards K Dean, Wagner Robert, Briggs Thomas. Investigating potential lightduty efﬁciency improvement through simulation of turbo-compounding and waste-heat recovery systems. SAE Int.; 2010. 2010-01-2209. [7] Weerasinghe WMSR, Stobart RK, Hounsham SM. Thermal efﬁciency improvement in high output diesel engines a comparison of a Rankine cycle with turbo-compounding. Appl Therm Eng 2010;30:2253e6. [8] Armstead John R, Miers Scott A. Review of waste heat recovery mechanisms for internal combustion engines. ASME paper, ICEF2010e35142. [9] Schmid H. Less emissions through waste heat recovery. In: Green ship Tech€rtsila € Corporation; 2004. nology conference. London: Wa [10] MAN Diesel & Turbo. Thermo efﬁciency system for reduction of fuel consumption and CO2 emission. 2014. nchez J. Methodology to design a bottoming [11] Maci an V, Serrano JR, Dolz V, Sa

276

[12]

[13]

[14] [15] [16] [17] [18]

[19]

[20] [21] [22]

[23]

M.T. Mito et al. / Energy 142 (2018) 264e276 Rankine cycle, as a waste energy recovering system in vehicles. Study in a HDD engine. Appl Energy 2013;104:758e71. pez L, Sala JM. Technological recovery potential of Bonilla JJ, Blanco JM, Lo waste heat in the industry of the Basque Country. Appl Therm Eng 1997;17(3):283e8. Ma Z, Yang D, Guo Q. Conceptual design and performance analysis of an exhaust gas waste heat recovery system for a 10000TEU container ship. Pol Marit Res 2012;19:31e8. Bari S, Hossain SN. Waste heat recovery from a diesel engine using shell and tube heat exchanger. Appl Therm Eng 2013;61:355e63. Butcher CJ, Reddy BV. Second law analysis of a waste heat recovery based power generation system. Int J Heat Mass Transf 2007;50:2355e63. Abusoglu A, Kanoglu M. First and second law analysis of diesel engine powered cogeneration systems. Energy Convers Manag 2008;49:2026e31. Yilmaz T. Optimization of cogeneration systems under alternative performance criteria. Energy Convers Manag 2004;45:939e45. Mohan G, Dahal S, Kumar U, Martin A, Kayal H. Development of natural gas ﬁred combined cycle plant for tri-generation of power, cooling and clean water using waste heat recovery: techno-economic analysis. Energies 2014;7: 6358e81. Choi BC, Kim YM. Thermodynamic analysis of a dual loop heat recovery system with trilateral cycle applied to exhaust gases of internal combustion engine for propulsion of the 6800 TEU container ship. Energy 2013;58: 404e16. Nag PK. Power plant engineering. third ed. McGraw Hill; 2008. Larsen U, Sigthorsson O, Haglind F. A comparison of advanced heat recovery power cycles in a combined cycle for large ships. Energy 2014;74:260e8. Etemoglu AB. Thermodynamic investigation of low-temperature industrial waste-heat recovery in combined heat and power generation systems. Int Commun Heat Mass Transf 2013;42:82e8. Baral S, Kim D, Yun E, Kim K. Energy, exergy and performance analysis of

[24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35]

[36] [37] [38]

small-scale organic rankine cycle systems for electrical power generation applicable in rural areas of developing countries. Energies 2015;8:684e713. Eastop TD, McConkey A. Applied thermodynamics. ﬁfth ed. Longman; 1993. Moran MJ, Shapiro HN, Boettner DD, Bailey MB. Fundamental of engineering thermodynamics. seventh ed. 1988. Abusoglu A, Kanoglu M. First and second law analysis of diesel engine powered cogeneration systems. Energy Convers Manag 2008;49:2026e31. Brzustowski TA, Brena A. Second law analysis of energy processes. Trans Can Soc Mech Eng 1986;10:121e8. Çengel YA. Heat transfer: a practical approach. 2002. Wais P. Heat exchangers - basic design application. 2012. p. 343e66. Holman JP. Heat transfer. McGraw Hill; 2010. Tien W, Yeh R, Hong J. Theoretical analysis of cogeneration system for ships. Energy Convers Manag 2007;48:1965e74. Kakac S, Bergles AE, Fernandes EO. Two-phase ﬂow heat exchangers: thermalhydraulic fundamentals and design. Kluwer Academic Publishers; 1987. Turton R, Bailie RC, Whiting WB. Analysis, synthesis and design of chemical processes. fourth ed. New Jersey: Prentice Hall PTR; 2013. Chemical engineering plant cost index 2016. Chemical Engineering; 2016. p. 64. www.chemengonline.com. Yang MH, Yeh RH. Analyzing the optimization of an organic Rankine cycle system for recovering waste heat from a large marine engine containing a cooling water system. Energy Convers M

Contents lists available at ScienceDirect

Energy journal homepage: www.elsevier.com/locate/energy

Utilizing the scavenge air cooling in improving the performance of marine diesel engine waste heat recovery systems Mohamed T. Mito a, Mohamed A. Teamah a, b, Wael M. El-Maghlany b, *, Ali I. Shehata a a

Mechanical Engineering Department, College of Engineering and Technology, Arab Academy for Science, Technology and Maritime Transport, Abu-Qir, Alexandria, Egypt b Mechanical Engineering Department, Faculty of Engineering, Alexandria University, Egypt

a r t i c l e i n f o

a b s t r a c t

Article history: Received 19 April 2017 Received in revised form 19 September 2017 Accepted 9 October 2017 Available online 21 October 2017

This paper aims at improving power generation efﬁciency of marine diesel engine waste heat recovery systems. It presents a novel technique of integrating the heat rejected in the scavenge air cooling process and the exhaust gas in operating a single and dual pressure steam power generation cycles. Moreover, a thermodynamic analysis of proposed systems was performed to identify the optimum operating parameters for achieving an overall efﬁciency improvement. The analysis considered the exergy destruction in each component and the energy/exergy efﬁciencies. A performance analysis was conducted to assess applicability and power output at off design conditions. An evaluation of achieved improvements by suggested designs was presented from both an economical and environmental standpoint. In conclusion, results show that, the recommended cycle increased overall efﬁciency improvement from 2.8% for the conventional system to 5.1%, with an additional power output of 1210 kW, representing 9.7% of the engine's power. Also, exergy efﬁciency increased signiﬁcantly by 6.6% when using the presented system. Furthermore, the waste heat recovery system attained a reduction in fuel consumption of 1538 Ton/year, reducing carbon dioxide emission by 4790 Ton/year. © 2017 Elsevier Ltd. All rights reserved.

Keywords: Marine diesel engine WHR Ship power plant Rankine cycle Emission reduction Energy management

1. Introduction Maritime transport, an essential contributor to world trade, has risen over the last decades where total transported payload increased from 2.6 billion tons in 1970 to 9.5 billion tons in 2013 [1]. Lately, due to current climate change and CO2 emissions reduction efforts for environmental protection, a need emerged to enhance diesel engines power generation efﬁciency using waste heat recovery for energy utilization [1e3]. The IMO estimate the total maritime shipping CO2 emissions in 2012 at 938 million tons; expected to rise to 250% by 2050 [1]. On the other hand, the new ships' design efﬁciency has dropped 10% since 1990, due to less hydrodynamic hull design as a need to maximize cargo capacity [4]. Diesel engines currently dominate the ﬁeld of marine propulsion due to its high thermal efﬁciency compared to other prime movers as it can exceed 50% [5]. Different waste heat recovery techniques have been used, such as; turbocharging, turbo-compounding,

* Corresponding author. E-mail addresses: [email protected], [email protected] (W.M. ElMaghlany). https://doi.org/10.1016/j.energy.2017.10.039 0360-5442/© 2017 Elsevier Ltd. All rights reserved.

Brayton cycle, Rankine engine cycle and thermoelectric generators (TEG) [6,7]; such techniques have increased engine thermal efﬁciency from 2% to 20%, depending on system design, quality of energy recovery, component efﬁciency, and implementation [8]. Accordingly, the ideal technique available for emissions reduction is to utilize the waste heat from fuel combustion that nearly reaches 50% for additional power generation. Hence, an efﬁcient approach to increase power generation efﬁciency is through employing a power cycle driven by diesel engine waste heat. Main engine exhaust gas energy is the most attractive amongst waste heat sources due to its high mass ﬂow rate and high temperature. This will result in fuel consumption and CO2 emissions reduction [5,9]. The beneﬁts of applying waste heat recovery (WHR) systems are not only constrained to environmental beneﬁts, but also provide a commercial aspect for shipping companies constituting in a lower annual fuel bill as well as lower emission of CO2 and NOX which results in the ‘green’ ship image improving their position in freight competition [9,10]. The waste heat energy from a diesel engine mainly resides in the exhaust gas where, about 25.4% of fuel energy is still available for use [9]. However, due to adiabatic compression of the scavenge air

M.T. Mito et al. / Energy 142 (2018) 264e276

Nomenclature A C Cp df E_ e F g HV H h K l m Nuo P Pt Pr Q_ Re Rth S T Uo _ W

Surface area, m2 Cost, $ Speciﬁc heat capacity, kJ/kg C Fin diameter, m Exergy rate, kW Speciﬁc exergy, kJ/kg Correction factor Gravitational acceleration, m/s2 Heating value of fuel, kJ/kg C Heat transfer coefﬁcient, W/m2 C Speciﬁc enthalpy, kJ/kg Thermal conductivity, W/m C Fin length, m Mass ﬂow rate, kg/s Nusselt number Pressure, bar Tube pitch, m Prandtl number Heat transfer rate, kW Reynolds number Thermal resistance, C/W Fin spacing, m Temperature, C Overall heat transfer coefﬁcient, W/m2 C Power input or output, kW

Greek symbol d Fin thickness h Efﬁciency j Exergy efﬁciency l Exergy destruction ratio lc Taylor instability

in the turbocharger; an increase in temperature occurs. Hence, cooling the scavenge air is required to enhance the engine volumetric efﬁciency. The reduction of scavenge air temperature is done by the charge air cooler, where a signiﬁcant amount of fuel energy (nearly 14.1%) is wasted [9]. A promising contributor for waste heat recovery applications other than exhaust gas is heat available in the scavenge air which can be used to operate a Rankine cycle. The scavenge air offers nearly the same mass ﬂow rate as the exhaust gas and a relatively high temperature. Additionally, the exhaust gas has a minimum temperature that needs to be avoided; to evade its dew point temperature caused by high HFO sulfur content. The scavenge air is not associated with this technical problem, thus its temperature can be decreased without interrupting engine's performance, which offers full heat utilization and enhanced management of waste heat sources. Numerous scientiﬁc studies targeted improvement in power generation efﬁciency and emission reduction of diesel engines by mainly using waste heat from exhaust gas for additional power generation. In reference to these, different approaches and systems were considered for achieving such objective. Macian et al. [11] introduced a methodology for design optimization of WHR bottoming Rankine cycle from heavy-duty diesel engines. The methodology was based on considering different heating and working ﬂuids. Bonilla et al. [12] discussed the potential of different WHR technologies to use waste heat from industry located in Basque Country in Spain. Zheshu ma et al. [13] performed an analysis for

265

Subscripts 0 Dead state cogen Cogeneration cond Condenser dest Destruction e Exit eco Economizer engine Engine related parameter eva Evaporator f Fuel g Exhaust gas heat Heat added i Inlet OV Overall PP Pump Pinch Pinch point rej Heat rejected s Water-steam SG Steam generator SU Superheater Turb Turbine WH Waste heat Acronyms CEPCI Chemical engineering plant cost index HFO Heavy fuel oil HP High pressure IMO International Maritime Organization LP Low pressure ORC Organic Rankine cycle RPM Revolution per minute WHR Waste heat recovery WHRS Waste heat recovery system

WHR from the exhaust of a container ship's diesel engine at variable engine load and exhaust gas boiler operating pressure. Bari and Hossain [14] studied numerically the effect of using a shell and tube heat exchanger in WHR from a diesel engine using a CFD program. The effect of different parameters such as shell diameter, number of shells, and tubes diameter have been studied to obtain the optimum WHR unit performance. Butcher and Reddy [15] presented a performance analysis for a WHRS operating on a Rankine cycle. The main focus was on the effect of pinch point on the power output, second law efﬁciency and outlet temperature of exhaust gas. Various cogeneration schemes were studied in previous literature where Abusoglu and Kanoglu [16] performed a ﬁrst and second law performance analysis of a diesel engine powered cogeneration system. The study was based on characteristics of an existing diesel powered cogeneration plant in which thermodynamic performance and efﬁciency were studied at full engine-load range. Furthermore, Yilmaz [17] analyzed the performance of a cogeneration system operating on a reversible Carnot cycle while investigating variable operating conditions. It was concluded that all performance criteria increases when the ratio between temperature of heating ﬂuid and cooling ﬂuid increases. Different trigeneration systems were presented including a novel trigeneration scheme that was introduced by Mohan et al. [18]. Based on the exhaust supplied by the gas turbine operating AlHamra power plant. The tri-generation involved power generation by a steam Rankine cycle, water treatment by an air gap

266

M.T. Mito et al. / Energy 142 (2018) 264e276

membrane desalination and cooling by a single pressure absorption chiller. Choi and Kim [19] introduced a novel trilateral cycle for WHR from the diesel engine of a container ship. The system is composed of an exhaust gas boiler for producing saturated steam for heating applications, a steam Rankine cycle, and an organic Rankine cycle for power generation. The ideas handled in the previous review were conﬁned to using exhaust gas in operating diesel engine WHRS, while analyzing the effects of different operating parameters on initial design or on system performance. The scavenge air was not used as a main heating ﬂuid despite its high waste heat content and better waste heat utilization when compared to exhaust gas. The current study innovation will rely on three main points; ﬁrstly, this paper will present a comparative study for different cycle conﬁgurations of WHR systems performing on a Rankine cycle, aiming at achieving higher energy availability, while introducing the use of scavenge air as a main heating ﬂuid. As a whole, the scavenge air role will be established in improving power generation efﬁciency and WHRS performance. Secondly, the integration between scavenge air and exhaust gas will be introduced in operating a single and dual pressure WHRS. The dual pressure system was not introduced previously for marine WHRS despite that it offers the operation at higher pressures while allowing full use of available waste heat. Thirdly, the economic and environmental impacts shall be considered for achieving reduction of both fuel consumption and carbon footprint. 2. System description 2.1. Speciﬁed diesel engine The diesel engine used is a HYUNDAI-WARSTILA 6RT-ﬂex58T-E with maximum output power of 13.94 MW. The two-stroke marine diesel engine includes a constant pressure turbocharging. The design of the systems provided in this study will be based on the engine's continuous service rating, presented in Table 1. The shop test provides the scavenge air and exhaust gas temperature and mass ﬂow rate at different operating loads, as shown in Table 2. 2.2. Proposed models Model I considers a simple Rankine cycle conventional WHRS, depending primarily on diesel engine exhaust gases for generating superheated steam. This will enable a comparison with the conventional approach tackled by previous literature, while using operating preferences of the speciﬁed Diesel engine. Also, model I will help in scoping improvements achieved by cycle conﬁgurations demonstrated in this study. The schematic diagram for model I is presented in Fig. 1. Model II integrates the exhaust gas and scavenge air to operate a single pressure Rankine cycle. The design consists of two boilers, one for generating dry saturated steam by means of heat available from scavenge air called ‘air boiler’ and the other for generating superheated steam by means of exhaust gas, called ‘gas boiler’. The Table 1 Operational characteristics of the diesel engine at the continuous service rating. Output power Scavenge air Exhaust gas Fuel (HFO)

Mass ﬂow rate Temperature Mass ﬂow rate Temperature Mass ﬂow rate Heating value

12.456 28.2 213 28.8 251 0.61 38.76

MW kg/s C kg/s C kg/s MJ/kg

gas boiler consists of an evaporator for producing dry saturated steam and a superheater that superheats dry steam from air boiler and gas boiler. The cycle also contains a feed water heater that acts as an economizer for both boilers where it preheats the feed water before entering the boilers by using scavenge air exiting the air boiler. Also minimizing the thermal effect in the exhaust gas boiler, which is the same function of regeneration but without power loss that accompanies steam extraction. The main advantage of using the scavenge air in preheating is that its temperature can be lowered without the constraint of the dew point temperature and risk of sulfuric acid formation as for the exhaust gas. The presented cycle is set to operate at a single operating pressure, where dry saturated steam produced from the air boiler is mixed with dry saturated steam produced from the gas boiler, to be eventually superheated in the gas boiler before its admission into the turbine. The schematic diagram for model II is described in Fig. 2. Model III purposefully aims to introduce a dual pressure cycle to the proposed system in order to beneﬁt from each energy source individually for maximizing waste heat utilization. The main difference to model II is the addition of a superheater within the air boiler allowing generation of superheated steam and operation of both boilers at two different pressures owing to the difference in heat available from scavenge air and exhaust gas. The feed heater will speciﬁcally serve the gas boiler in increasing its ﬂow rate, whereas the exhaust heat that would have been needed to preheat the feed water will be used for evaporation. Fig. 3 presents the schematic diagram for model III. Before initiating the thermodynamic analysis, some design considerations were assumed for all models, presented in Table 3. 3. Thermodynamic analysis In this study, the steady-state physical and thermodynamics equations were used based on the ﬁrst and second thermodynamic laws. The steady state equations are applicable during the ship's voyage. However, they were excluded at heavy sea conditions, bad weather, maneuvering, and docking. The thermodynamic analysis was founded on the pinch point temperature difference as a dominant parameter governing the performance of any WHR boiler. It deﬁnes the WHRS optimum operating pressure based on heating ﬂuid temperature. The WHRS designs proposed in this study will be based on minimum pinch point temperature difference of 15 [5]. The following assumptions are considered throughout the study: The system is assumed to be at steady state. No pressure drops on steam side or heating ﬂuid side in the heat exchanger. Approach point is negligible. The dead state condition is assumed at T0 ¼ 20 C and P0 ¼ 100 kPa [22]. The feed pump compression process is assumed isentropic, thus, the pump will not be considered in the exergy analysis [23].

3.1. Energy analysis The governing equations used for designing WHRS are energy balance equations based on the ﬁrst thermodynamics law for energy conversion while neglecting kinetic and potential energy changes. The conservation of energy equation is represented in equation (1) as follows:

M.T. Mito et al. / Energy 142 (2018) 264e276

267

Table 2 Temperature and mass ﬂow rate distribution of exhaust gas and scavenge air at different operating loads. Load (%)

25

50

75

90

100

Engine speed (rpm)

65.9

83

95

101

104.6

7.1 290 6.9 52

15.1 262 14.8 134

24.1 249 23.5 190

28.8 251 28.2 213

32.7 261 31.9 230

Mass ﬂow rate (kg/s) Temperature ( C) Mass ﬂow rate (kg/s) Temperature ( C)

Exhaust gas Scavenge air

Fig. 1. Schematic diagram for model I. Fig. 3. Schematic diagram for model III.

maximize work output. Consequently, the exergy will represent the useful work potential of the system at the speciﬁed conditions [24,25].

3.2.1. Formulations for the exergy analysis The exergy balance for a control volume undergoing steady state process with negligible changes in kinetic and potential energy is represented as:

_ ¼ E_ heat ±W

Fig. 2. Schematic diagram model II.

_ ¼ Q_ ±W

X

_ e he m

X

_ i hi m

(1)

_ are the net heat and power input, m _ is the mass where Q_ and W ﬂow rate, h is the speciﬁc enthalpy and the subscripts i and e stands for the inlet and exit states, respectively [24]. The energy balance equations for model I are presented in Table 4, whereas for models II and III same equations are applied to each component in a similar manner. 3.2. Exergy analysis An exergy analysis based on the second law of thermodynamics will be performed to assess WHRS performance with varying engine load. The exergy analysis implies that the system will be compared from an initial speciﬁed state to the dead state, where at the end of the process, the system should reach dead state to

X

_ e ee m

X

_ i ei þ E_ dest m

(2)

where E_ heat is the exergy transferred to the system at a speciﬁed temperature, E_ dest is the rate of exergy destruction due to irreversibility and W is the exergy change as a result of work input or output to system [26]. Additionally, e and E_ are the speciﬁc ﬂow exergy and the rate of total exergy which are presented by equations (3) and (4) respectively, where h is the speciﬁc enthalpy, s is the speciﬁc entropy and the subscript 0 indicate the dead state [22].

e ¼ ðh h0 Þ T0 ðs s0 Þ

(3)

_ E_ ¼ me

(4)

The exergy transferred E_ heat associated with heat transfer at a speciﬁed temperature T is presented in equation (5). Where Q_ is the heat transfer rate and T0 is the temperature of the dead state [26].

E_ heat ¼

X

1

T0 _ Q T

(5)

The exergy destruction due to irreversibilities within the system E_ dest is given by equation (6), where the total exergy destruction is calculated for the whole system while considering exergy destruction in each component of the cycle [22].

268

M.T. Mito et al. / Energy 142 (2018) 264e276

Table 3 Design considerations assumed for all models. Parameter

Value

Cooling water temperature at ISO condition Temperature difference between the saturation temperature of steam and inlet of cooling water Steam saturation Temperature at condenser pressure Condenser pressure, PCond Minimum exhaust gas temperature based on the dew point of the exhaust gas produced by burning HFO Superheater approach temperature Minimum degree of superheat

25 C 15 C 40 C 0.07 bar 150 C 20 C 20 C

ef ¼ 1:065 HV

Table 4 Energy balance equations for model I.

Evaporator

_ s ðhs;4 hs;1 Þ ¼ m _ g cp;g ðTg;1 Tg;4 Þ Q_ ov ¼ m _ s ðhs;2 hs;1 Þ ¼ m _ g cp;g ðTg;3 Tg;4 Þ Q_ eco ¼ m _ s ðh h Þ ¼ m _ g cp;g ðT T Þ Q_ ¼m

Superheater

_ s ðhs;4 hs;3 Þ ¼ m _ g cp;g ðTg;1 Tg;2 Þ Q_ su ¼ m

Overall energy balance Economizer

eva

s;3

s;2

g;2

g;3

[5] [20]

[5] [21] [21]

(11)

The energy and exergy balance equations for the system's components are presented in Table 5, where subscripts i and e indicate the inlet and exit states for a ﬂuid stream at each component, respectively. 3.3. Cost analysis

E_ dest ¼

X

_ i ei m

X

_ e ee þ m

X T _ 1 0 Q_ ±W T

(6)

The exergy destruction ratio l, presented in equation (7), will be considered to assess the ratio of energy lost and irreversibilities caused by each component with regards to total exergy destruction. Another important parameter in performing an exergy analysis is the exergy efﬁciency j, which represents the ratio between exergy of the WHRS E_ WHRS and exergy input to the system E_ heat associated with heat transfer at a speciﬁed heating temperature T. The exergy efﬁciency is represented in equation (8).

l¼

E_ dest;cmponent E_

(7)

dest;total

E_

j ¼ _WHRS Eheat

(8)

The ﬁrst law efﬁciency of WHRS is a ratio between total power output and heat added through fuel burned. The total power output is presented by engine power and additional WHRS power. However, the second law efﬁciency for a cogeneration system is a comparison between whole system exergy rate and fuel exergy. The formulas of both ﬁrst and second law efﬁciencies are presented by equations (9) and (10), respectively [26].

hcogen;I

_ _ W engine þ WWHRS ¼ _ f HV m

hcogen;II ¼

(9)

_ _ W engine þ EWHRS _E f

_ _ s ððhe hi Þ T0 ðse si ÞÞ W engine þ m ¼ _ f ef m

(10)

. Brzustowski and Brena [27] discussed the ratio ef HV for different hydrocarbons and their study concluded a constant value of 1.065 for the ratio between fuel exergy and fuel heating value. The authors also added that; previous ratio is applicable to heavy hydrocarbons with undetermined composition, which applies to HFO used onboard ships and in this study. Therefore the formula in equation (11) will be used to calculate fuel exergy ef .

An economic analysis will be performed to compare between presented models regarding their initial cost. The steam generator will be the only component considered in the economic analysis as it is the main point of difference between each model. The economic analysis includes estimating the initial cost of the proposed systems and reduction in fuel consumption achieved by the additional power of the WHRS. 3.3.1. Heat transfer modeling The steam generator initial cost depends mainly on its surface area. Thus, the steam generator surface area is calculated for each model. The heat transfer modeling is performed regarding shell and tube heat exchanger based on cross ﬂow between the heating and working ﬂuids. The thermal load for each heat exchanger is used to calculate the required heat transfer surface area from equation (12), where Uo is the overall heat transfer coefﬁcient, DTLM is the logarithmic mean temperature difference between the heating and working ﬂuids across the heat exchangers and Ao is the required surface area. In equation (12), the logarithmic mean temperature difference is multiplied by a correction factor F to compensate for the cross ﬂow arrangement [24]. The value of the correction factor is unity for the boiling ﬂow in evaporators. However, the correction factor is calculated for the economizer and superheater for single pass cross ﬂow exchanger where both ﬂuids are unmixed [28].

Q Ex ¼ Uo Ao FDTLM

(12)

The modules suggested involve radial ﬁns on the heating ﬂuid side to increase the surface area per cubic meter of exchanger volume [29]. The heat transfer coefﬁcient across the ﬁn and the ﬁn material thermal conductivity are assumed constant. The ﬁn temperature is assumed to vary in one dimension and any variation among other dimensions is neglected. This assumption is valid for the case of thin ﬁns where studies have shown that the error involved with the assumption of one dimensional heat transfer is less than 1% when [28]:

Hd < 0:2 k

(13)

3.3.1.1. Heat transfer correlations on the heating ﬂuid side. The heating ﬂuid heat transfer coefﬁcient will be dependent on the ﬂow characteristics around the radially ﬁnned tubes. The average Nusselt

M.T. Mito et al. / Energy 142 (2018) 264e276

269

Table 5 The energy and exergy balance equations for the WHRS. Component

Energy analysis

Exergy analysis

Heat exchanger

_ e hi Þ Q_ SG ¼ mðh

_ i he Þ mT _ 0 ðsi se Þ þ Q_ WH 1 TT0 E_ dest; SG ¼ mðh

Turbine

_ _ i he Þ W Turb ¼ mðh _ i he Þ Q_ cond ¼ mðh

_ 0 ðse si Þ E_ dest; Turb ¼ mT

Condenser Pump

_ PP ¼ mðh _ e hi Þ W

_ 0 ðse si Þ E_ dest; PP ¼ mT

_ i he Þ mT _ 0 ðsi se Þ þ Q_ rej 1 TT0 E_ dest; Cond ¼ mðh

number for low radial ﬁnned tube cross ﬂow heat exchangers can be calculated from the following empirical correlation [29]:

Nuo ¼ 0:183Re

0:7

s0:36 P t l df

!0:06

l df

!0:11 pr0:36

(14)

The above mentioned correlation is applicable for 103 Re 8103 , 0:19 sl 0:66, 1:1 dpt 4:92, f l 0:058 d 0:201. The outside heat transfer is thus calculated f from equation (15) as a function of Nusselt number [30].

Ho ¼

Nuo ko do

(15)

1 Uo Ao ¼ P Rth

NUi ¼ 0:023Re0:8 pr0:4

(16)

The inside heat transfer coefﬁcient in the evaporator is based on the evaporation of the saturated feed water circulating the riser. This particular case involves speciﬁcally ﬁlm boiling heat transfer correlations [31]. Stable ﬁlm boiling can be solved analytically because the ﬂow pattern is relatively simple where an exact solution can be obtained. The following equation considers the interfacial shear and curvative effects in calculating the heat transfer coefﬁcient (micro and macro convection) [32].

0:59 þ 0:069

3.3.2. Capital cost estimation Cost evaluation equations are widely used for estimating the initial cost of thermal equipment for the preliminary design stage. The module cost for shell and tube heat exchangers is given by equation (20) [33].

#0:75 " 3 kv g hfg ðrl rv Þrv di mi ðTw Tsat Þlc

lc

logCp ¼ K1 þ K2 logðASG Þ þ K3 ðlogðASG ÞÞ2

logFp ¼ C1 þ C2 logð10P 1Þ þ C3 ðlogð10P 1ÞÞ2

3.4. Solution procedure

gc s gðrl rv Þ

(18)

The overall heat transfer coefﬁcient for each section of the steam Table 6 Economic cost parameters of the WHRS [37]. K1

K2

K3

B1

B2

4.3247

0.3030

0.1634

1.63

1.66

C1

C2

C3

Fm

0.0388

0.11272

0.08183

1.4

(22)

where C1 , C2 and C3 are the coefﬁcients of pressure factor which are stated in Table 6 [33]. The estimated purchased cost is calculated from equation (23) while considering the chemical engineering plant cost index CEPCI. CEPCI reﬂects the effect of time on the purchased equipment cost [34,35].

The minimum wavelength for Taylor instability, lc , is given by equation (18), where s is the surface tension of the working ﬂuid (N/m). Also, rl and rv are the working ﬂuid densities at the liquid and vapor state. Additionally, gc is the proportionality constant in Newton's second law of motion which equals unity for SI units [32].

0:5

(21)

where K1 , K2 and K3 are the coefﬁcients of equipment costs which are shown in Table 6. Furthermore, the pressure factor Fp for the heat exchangers can be calculated from equation (22) [33].

Cestimated ¼ ðCSG Þ2001

(20)

where Fm is the material factor and B1 and B2 are constants given in Table 6. Furthermore, the purchased cost of equipment when using carbon steel construction Cp , can be calculated from equation (21) [33].

(17)

lc ¼ 2p

(19)

CSG ¼ Cp B1 þ B2 Fm Fp

3.3.1.2. Heat transfer correlations on the working ﬂuid side. There are two ﬂow regimes for the working ﬂuid through the steam generator which are single phase ﬂow through the economizer and superheater and two phase boiling ﬂow through the evaporator. Equation (16) represents a correlation for calculating the Nusselt number for a single phase ﬂuid ﬂowing inside tubes [30].

Hi ¼

generator is deﬁned based on the electrical analogy of the heat transfer resistances, as shown in equation (19) [30].

CEPCI2014 CEPCI2001

(23)

The thermodynamic models presented in this study are formulated by using a series of iterative procedures developed on an in-house MATLAB program. The program solves governing equations, while maintaining design constraints held throughout this study. The properties of water/steam were obtained from the National Institute of Standards and Technology (NIST) database REFPROP 8.0 [36]. The solution procedure used in MATLAB is presented in Fig. 4. 4. Results & discussion 4.1. Design investigation based on the operating pressure Firstly, the effect of different operating pressures on the output

270

M.T. Mito et al. / Energy 142 (2018) 264e276

power will be studied while maintaining design criteria to determine the optimum pressure for utilizing waste heat and satisfying proper system operation. The solution is achieved by iterating the boiler pressure while maintaining a minimum pinch point of 15 C [5] between heating ﬂuids and feed water at the evaporator inlet. A common relation trend between operating pressure and output power is presented in Fig. 5 (a, b), which is the increase in power while increasing operating pressure until reaching a maximum value. Whereas the pressure increases past this value a uniform decrease in output power occurs. The reason for this trend is the pinch point temperature, where at low pressures the saturation temperature is low and difference in temperature between heating ﬂuid and steam is high, allowing for full use of waste heat. At higher pressure, the pinch point decreases past the set value due to increasing saturation temperature. As a result, the mass ﬂow rate of steam produced decreases; to decrease heat transferred and increase the temperature of heating ﬂuid at the boiler exit for maintaining a high pinch point. This reduces output power disregarding the increase in enthalpy drop achieved in the turbine. As a result of increasing the heating ﬂuid temperature at the economizer exit, less heat was transferred in the evaporator and superheater, thus, decreasing the superheated steam yield of the steam generator as presented in Fig. 6 (a, b). Subsequently, this led to a decrease in output power. Therefore, the relation between mass ﬂow rate of steam and operating pressure is affected mainly by waste heat source temperature and pinch point. In Fig. 6 (a), the steam mass ﬂow rate for model I is nearly constant until an operating pressure of 5 bar because the pinch point is at design value and consequently heat added by exhaust is constant. However, for

Fig. 4. Flow diagram describing the solution procedure.

model II, its mass ﬂow rate decreases constantly with increasing pressure; this is caused by scavenge air low temperature in which the pinch point is less than design value and heat added by scavenge air decreases with increasing pressure. The solution procedure also emphasized the importance of the exhaust steam dryness fraction where it is an important factor in determining both optimum operating pressure and presented cycle applicability. According to the author's knowledge, the dryness fraction was not discussed in previous literature. As shown in Fig. 7 (a, b), at constant superheated steam temperature, the dryness fraction decreased at higher pressures owing to water properties and change in the degree of superheat where water is known to be a wet ﬂuid as it has a tendency to prematurely condense during expansion owing to its bell-shaped temperature-entropy diagram [37]. The low dryness fraction is associated with low temperature of exhaust gas and scavenge air supplied by the engine, where the superheated steam temperature is set based on the heating ﬂuids temperature. The optimum operating pressure was selected based on the results of the MATLAB program where deﬁning the optimum pressure regarded output power, a dryness fraction over 0.88 [20] and a temperature distribution which utilized the full capacity of waste heat offered by the diesel engine while maintaining boiler lifetime. Even though the selected operating pressures do not yield maximum power, they still satisfy design condition. The optimum operating conditions for the three models at the continuous service rating are presented in Table 7. 4.2. Exergy destruction for each model The exergy analysis implies calculating the exergy destruction for each model with varying operating pressure to determine the rate of potential energy lost and the utilization of available energy. Fig. 8 (a, b & c) presents the variation in exergy destruction with changing operating pressure for all presented models. The rate of exergy destruction for the steam generator is clearly the highest among all components due to the boiling process. Additionally, it decreases uniformly for all models with increasing pressure, where the latent heat of vaporization decreases when operating at higher pressures, this allows for lower irreversibilities and higher energy utilization. Furthermore, the condenser exergy destruction is prominently dependent on the steam mass ﬂow rate, due to small difference between temperatures of cooling water and surrounding. This gives a low signiﬁcance to the Carnot efﬁciency term; used to calculate exergy transferred accompanying heat rejected from the condenser. The exergy destruction caused by the turbine is signiﬁcantly lower compared to other components where it reﬂects the turbine isentropic efﬁciency. On the other hand, in Fig. 8, the total exergy destruction varied between all three models where the lowest exergy destruction was achieved by model I as it requires the least heat input and achieves the lowest steam mass ﬂow rate. As for models II and III, there is a noticeable increase in total exergy destruction from model I, where models II and III dictate more exergy transfer due to increased heat input as a result of integrating heat available from exhaust gas and scavenge air. However, when dealing with WHRS the total exergy destruction indicates lower utilization of previously wasted heat. But as shown by Fig. 5, operating at higher pressures leads to a decrease in power output as a result of the decreasing pinch point and steam mass ﬂow rate. The results of the exergy analysis at the continuous service rating are presented in Table 8. The steam generator exergy destruction ratio is highest for model I reaching 64.4% as a result of high exhaust exit temperature. This indicates high exergy destruction related to available energy and low energy utilization

M.T. Mito et al. / Energy 142 (2018) 264e276

271

(a)

(b)

Fig. 5. Effect of operating pressure on the output power.

(a)

(b)

Fig. 6. Effect of operating pressure on the superheated steam yield.

(a)

(b)

Fig. 7. Effect of operating pressure on the dryness fraction.

Table 7 Operation characteristics for each model at the optimum operating pressure. Model

Operating pressure (bar)

Steam yield (kg/s)

Heating ﬂuid exit temperature ( C)

Dryness fraction

I II

3 3

1.16 2.08

150 150

0.894 0.89

III

Air boiler

Gas boiler

2.25

Scavenge air

Exhaust gas

Air boiler

Gas boiler

2

3

109.2

150

0.898

0.894

272

M.T. Mito et al. / Energy 142 (2018) 264e276

Fig. 8. Effect of operating pressure on the exergy destruction for a) model I, b) model II and c) model III.

when compared to other models. As for the exergy efﬁciency, model I achieved the lowest exergy efﬁciency of 57.8% as a result of high exhaust gas exit temperature, caused by the high dew point temperature of the exhaust produced by burning HFO. However, the exergy efﬁciency increased signiﬁcantly to 64.4% and 62.4% for models II and III respectively. The reason for this improvement is the decrease in heating ﬂuids exit temperature which indicates a better waste heat utilization in case of models II and III. In conclusion, although models II and III show higher total exergy destruction when compared to model I, however, this is due to the increase in energy input to the system as a result of using multiple heating ﬂuids. Nonetheless, when considering the exergy

efﬁciency, models II and III dictate better energy management when compared to model I. 4.3. First and second law efﬁciencies The overall improvement of WHRS operating efﬁciency was taken into consideration regarding the ﬁrst and second law efﬁciencies, as shown in Fig. 9 (a, b). The ﬁrst and second law efﬁciencies are directly proportional to the superheated steam yield achieved by each model which is dependent on the operating pressure and pinch point at this pressure. The values of the second law efﬁciency are lower than the energy efﬁciency, where exergy

M.T. Mito et al. / Energy 142 (2018) 264e276

utilization.

Table 8 Results of the exergy analysis at the continuous service rating. Model

Component

E_ dest (kW)

l (%)

E_ dest;total (kW)

j (%)

I

Steam generator Condenser Turbine Steam generator Condenser Turbine Steam generator Condenser LP Turbine HP Turbine

352.7 162.42 33.1 469 292.6 59.6 459.5 315.7 22.9 37.7

64.4 29.6 6 57.1 35.6 7.3 55 37.8 2.7 4.5

548.2

57.8

II

III

821.2

64.4

835.2

62.4

efﬁciency also considers system irreversibilities. 4.4. Comparison between the suggested models A comparison between suggested models was made in terms of output power, superheated steam yield and improvement in system efﬁciency at engine continuous service rating. 4.4.1. Evaluating the power output The proposed systems power outputs are compared in Table 9 at engine continuous service rating. Regarding model I, it generated the least power at 672 kW. As for models II and III, they generated roughly the same power output at 1210 kW and 1230 kW respectively. This proves that the dual pressure cycle did not yield a signiﬁcant improvement in output power, as the pressure difference should be more distinctive to yield the increase in enthalpy drop achieved in the turbine while operating at higher pressures. Due to low temperature of waste heat available; pressure levels of the dual pressure system cannot be increased signiﬁcantly to improve power output. 4.4.2. Evaluating the superheated steam yield The proposed systems steam yield is compared at engine continuous service rating. The results, presented in Table 9, conform to the power output of each system, as the steam yield also depends on heat available from the waste heat source used. The system relying on exhaust gas produced 1.16 kg/s, while there was a signiﬁcant increase in steam yield for model II which produced 2.08 kg/s of steam owing to addition of heat available by scavenge air. Model III produced the largest superheated steam yield at 2.25 kg/s mainly due to the dual pressure conﬁguration where the air boiler operated at low pressure allowing for more waste heat

(a)

273

4.5. Performance analysis The suggested WHRS performance was investigated to study the effect of changing engine load on functionality and output power for each system. The performance analysis will be very beneﬁcial in reviewing the behavior of each system at real operating conditions which include full load operation at open seas and lower engine loads at docking or maneuvering. The WHRS operation depends mainly on temperature and mass ﬂow rate of exhaust gas and scavenge air, which are dependent on engine operating condition. The performance analysis for each model is presented in Fig. 10. 4.5.1. Model I The proposed system in model I operates at full range of engine operation due to high exhaust gas temperature. Additionally, the power output changes regarding the change in mass ﬂow rate of the engine's exhaust with respect to its operating load. At 25% engine load, the cycle output power is 228 kW, which is 6.5% of the engine's power at mentioned load. Furthermore, at 100% load, the exhaust gas temperature is 261 C, while the mass ﬂow rate of exhaust is 32.66 kg/s. Consequently, the WHRS output power is 837 kW at full engine load which is 6% of the engine's output power. This settles that, the power generated by the conventional WHRS proposed in model I is useful at different operating conditions including docking, maneuvering, and open seas. 4.5.2. Model II During the operation at loads lower than 75% which include docking and maneuvering of the ship, the scavenge air would not be sufﬁcient for producing saturated steam at the speciﬁed pressure. At this condition, the scavenge air will be only used for heating the feed water used in the exhaust gas boiler. The function of feed water heater arises at this point because it will be used to supply heat to feed water entering the gas boiler at low operating loads, as the air boiler will be bypassed due to low scavenge air temperature. 4.5.3. Model III At engine loads less than 90%, the scavenge air would not be sufﬁcient for generating superheated steam at 193 C and the lowpressure boiler would not be operational. Thus, scavenge air waste heat will be used for feed water heating in the feed heater to increase the steam yield generated by the high-pressure exhaust gas boiler.

(b)

Fig. 9. Effect of operating pressure on the a) ﬁrst and b) second law efﬁciencies of the whole system.

274

M.T. Mito et al. / Energy 142 (2018) 264e276

Table 9 Comparison between models at the continuous service rating. Model

Superheated Steam yield (kg/s)

WHRS Power (kW)

WHRS Power (As a percentage from the engine power)

I II III

1.16 2.08 2.25

672 1210 1230

5.4% 9.7% 9.9%

Table 10 Comparison between the fuel saving for each model. Model

Fuel saving (Ton/year)

I II III

854 1538 1564

Model II Model II Model I $-

$100 $200 $300 $400 Cost of fuel saved (x103 $/year)

Fig. 10. Performance curves for each model at variable engine load. Fig. 11. Cost of fuel saved per year.

By reviewing the performance analysis, it is clear that; models II and III offer higher power output than model I as a result of using the scavenge air alongside the exhaust gas. This setup offers a consistent power output due to use of the exhaust gas and an overall higher power than model I due to use of the scavenge air in preheating the feed water at low loads. When comparing models II and III, model II offers higher functionality and power output than model III at loads less than 90%, as a result of using the scavenge air in producing saturated steam, in model II, which is more suitable based on scavenge air temperature. 4.6. Economical aspect of the WHRS The economic analysis for the proposed WHRS is presented in the form of reduction in fuel consumption and fuel cost per year as a result of the additional WHRS power. The reduction in engine running cost is considered extremely beneﬁcial for shipping companies as it translates to a reduction in shipping fees and thus improving their position in freight competition. The speciﬁc fuel consumption is calculated using equation (24) by considering the engine power and the additional WHRS power. The reduction in speciﬁc fuel consumption represents the amount of fuel saved when using the WHRS. The WHRS is assumed to operate for 300 days per year at the continuous service rating [3].

S:F:C ¼

mfuel _ þW

_ W engine

(24)

WHRS

The results presented in Table 10 indicate that; the WHRS operating on the diesel engine exhaust gas (model I) achieved a noticeable reduction in fuel consumption of 854 Ton/year by generating additional power from the WHRS. This amount translates to 5.4% of HFO used daily by the engine. The integration of exhaust gas and scavenge air achieved a signiﬁcant reduction in

fuel consumption of 1538 Ton/year and 1563 Ton/year for the single and dual pressure systems respectively, this is equivalent to 9.7% and 9.9% of the engine's consumption. The reduction in fuel consumption achieved by applying the WHRS can be translated to a reduction in engine running cost. According to the French National Institute of Statistics and Economic Studies (INSEE), the price of HFO in May 2016 was 243 $/Ton. Consequently, the HFO cost for running the engine 300 days/year at the continuous service rating can be estimated to nearly 3.84 million dollars [38]. Fig. 11 displays the cost of fuel saved for each model where model I achieved a reduction of approximately 208 103 $/year, while model II reduced fuel cost signiﬁcantly by nearly 374 103 $/year. The results displayed in Fig. 11 indicates a signiﬁcant economic value of using the scavenge air alongside the exhaust gas where model II achieved an 80% higher reduction in fuel cost compared to the conventional system (model I).

Model III Model II Model I 0.00

0.50 1.00 1.50 Cost of steam generator (x106 $)

Fig. 12. Initial cost of the steam generator.

2.00

M.T. Mito et al. / Energy 142 (2018) 264e276

Model III Model II Model I 0

1000 2000 3000 4000 5000 6000 CO2 emission reduction (Ton/year) Fig. 13. CO2 emission reduction.

The feasibility of the WHRS system can be determined by comparing the initial cost and the fuel consumption reduction for each model. The initial cost of the steam generator for each model is presented in Fig. 12, where model I offered the lowest initial cost of 0.83 106 $ as a result of the low steam output, thus, demanding smaller equipment. Nonetheless, model I offered the lowest reduction in fuel consumption. Additionally, model III had the highest initial cost of 1.54 106 $ due to the complexity of the system and implementation an additional superheater for the air boiler. 4.7. Reduction in CO2 emission The environmental impact of the proposed WHRS is presented in the CO2 emission reduction that resulted from the additional power cycle performing alongside the Diesel engine. The WHRS output power was generated without additional fuel consumption and thus, a reduction in CO2 emission was achieved. The speciﬁc CO2 emission for burning HFO is 3.114 kg-CO2/kg-fuel [3]. Consequently, the reduction in CO2 emission is directly proportional to fuel saving achieved by the proposed WHR systems. As presented in Fig. 13, the conventional system operating on exhaust gas as heating ﬂuid (model I) accomplished a reduction in CO2 emissions of 2660 Ton/year, while the proposed systems utilizing heat available in both exhaust and scavenge air achieved 4790 Ton/year and 4870 Ton/year for the single and dual pressure systems respectively. Results indicate that models II and III are superior to model I due to the fact that more heat was successfully utilized when using the suggested systems. 5. Conclusion This study was set to investigate the improvements applicable to WHR systems involving marine diesel engines. The WHR method chosen is the Rankine cycle in which performance improvements were achieved by designing a system that integrates different sources of waste heat. A thermodynamic analysis was performed for all three suggested models based on energy and exergy balance equations. All models, showed decrease in output power and steam yield with increasing pressure caused by maintaining a constant pinch point at increasing saturation temperature. The integration of scavenge air and exhaust gas yielded the highest power outputs of 1210 kW and 1230 kW, equivalent to 9.7% and 9.9% of engine's power for single and dual pressure cycles respectively. Model I achieves the lowest improvement in ﬁrst law efﬁciency of 2.8%, while models II and III improved the efﬁciency by 5.1% and 5.2% respectively.

275

The exergy analysis showed that models II and III imply higher total exergy destruction when compared to model I, as a result of increasing energy input to the system when using multiple heating ﬂuids. However, when considering exergy efﬁciency, it increased signiﬁcantly from 57.8% for the conventional system to 64.4% and 62.4% for models II and III respectively. Thus, models II and III present better energy management when compared to model I. The performance analysis shows that model I is functional at full range of engine load with variable steam yield, owing to high exhaust gas temperature. However, the scavenge air temperature for models II and III varied with the engine load where at certain loads it would not be sufﬁcient for generating saturated steam. Consequently, the scavenge air is used for preheating the feed water before entering the exhaust boiler for achieving higher steam yield at loads less than 75% and 90% for models II and III respectively. As a result, models II and III generate more power than model I at overall engine load range. Model I achieved the lowest saving in fuel consumption of 5.4% (854 Ton/year), on the other hand a signiﬁcant reduction in fuel consumption was achieved by models II and III of 9.7% (1538 Ton/year) and 9.9% (1563 Ton/year) respectively. Similarly, the CO2 emission reduction is the lowest for model I, where models II and III reduced CO2 emission by 4790 Ton/year and 4870 Ton/ year respectively. In conclusion, the recommended system for implementation in ships design is model II, which relies on the scavenge air and exhaust gas in operating a single pressure Rankine cycle while incorporating a common economizer and superheater. Model II will offer lower initial and maintenance costs and less complexity than the dual pressure cycle in model III with a difference of 0.1% improvement in efﬁciency. Meanwhile, model II achieves an efﬁciency improvement of 5.1% and fuel consumption reduction of 9.7%. Acknowledgements Financial support granted from the Arab Academy for Science, Technology and Maritime Transport for this research is greatly appreciated. References [1] IMO. Third IMO greenhouse gas study 2014. International Maritime Organization; 2015. [2] Gençsü I, Hino M. Raising ambition to reduce international aviation and maritime emissions. In: Contributing paper for Seizing the global opportunity: partnerships for better growth and a better climate. London and Washington, DC: New Climate Economy; 2015. Available at: http://newclimateeconomy. report/misc/working-papers. [3] Min-Hsiung Yang. Thermal and economic analyses of a compact waste heat recovering system for the marine diesel engine using transcritical Rankine cycle, vol. 106; 2015. p. 1082e96. [4] Faber J, Hoen MT. Historical trends in ship design efﬁciency. In: Seas at risk and transport & environment; 2015. [5] Grljusi c M, Medica V, Ra ci c N. Thermodynamic analysis of a ship power plant operating with waste heat recovery through combined heat and power production. Energies 2014;7:7368e94. [6] Edwards K Dean, Wagner Robert, Briggs Thomas. Investigating potential lightduty efﬁciency improvement through simulation of turbo-compounding and waste-heat recovery systems. SAE Int.; 2010. 2010-01-2209. [7] Weerasinghe WMSR, Stobart RK, Hounsham SM. Thermal efﬁciency improvement in high output diesel engines a comparison of a Rankine cycle with turbo-compounding. Appl Therm Eng 2010;30:2253e6. [8] Armstead John R, Miers Scott A. Review of waste heat recovery mechanisms for internal combustion engines. ASME paper, ICEF2010e35142. [9] Schmid H. Less emissions through waste heat recovery. In: Green ship Tech€rtsila € Corporation; 2004. nology conference. London: Wa [10] MAN Diesel & Turbo. Thermo efﬁciency system for reduction of fuel consumption and CO2 emission. 2014. nchez J. Methodology to design a bottoming [11] Maci an V, Serrano JR, Dolz V, Sa

276

[12]

[13]

[14] [15] [16] [17] [18]

[19]

[20] [21] [22]

[23]

M.T. Mito et al. / Energy 142 (2018) 264e276 Rankine cycle, as a waste energy recovering system in vehicles. Study in a HDD engine. Appl Energy 2013;104:758e71. pez L, Sala JM. Technological recovery potential of Bonilla JJ, Blanco JM, Lo waste heat in the industry of the Basque Country. Appl Therm Eng 1997;17(3):283e8. Ma Z, Yang D, Guo Q. Conceptual design and performance analysis of an exhaust gas waste heat recovery system for a 10000TEU container ship. Pol Marit Res 2012;19:31e8. Bari S, Hossain SN. Waste heat recovery from a diesel engine using shell and tube heat exchanger. Appl Therm Eng 2013;61:355e63. Butcher CJ, Reddy BV. Second law analysis of a waste heat recovery based power generation system. Int J Heat Mass Transf 2007;50:2355e63. Abusoglu A, Kanoglu M. First and second law analysis of diesel engine powered cogeneration systems. Energy Convers Manag 2008;49:2026e31. Yilmaz T. Optimization of cogeneration systems under alternative performance criteria. Energy Convers Manag 2004;45:939e45. Mohan G, Dahal S, Kumar U, Martin A, Kayal H. Development of natural gas ﬁred combined cycle plant for tri-generation of power, cooling and clean water using waste heat recovery: techno-economic analysis. Energies 2014;7: 6358e81. Choi BC, Kim YM. Thermodynamic analysis of a dual loop heat recovery system with trilateral cycle applied to exhaust gases of internal combustion engine for propulsion of the 6800 TEU container ship. Energy 2013;58: 404e16. Nag PK. Power plant engineering. third ed. McGraw Hill; 2008. Larsen U, Sigthorsson O, Haglind F. A comparison of advanced heat recovery power cycles in a combined cycle for large ships. Energy 2014;74:260e8. Etemoglu AB. Thermodynamic investigation of low-temperature industrial waste-heat recovery in combined heat and power generation systems. Int Commun Heat Mass Transf 2013;42:82e8. Baral S, Kim D, Yun E, Kim K. Energy, exergy and performance analysis of

[24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35]

[36] [37] [38]

small-scale organic rankine cycle systems for electrical power generation applicable in rural areas of developing countries. Energies 2015;8:684e713. Eastop TD, McConkey A. Applied thermodynamics. ﬁfth ed. Longman; 1993. Moran MJ, Shapiro HN, Boettner DD, Bailey MB. Fundamental of engineering thermodynamics. seventh ed. 1988. Abusoglu A, Kanoglu M. First and second law analysis of diesel engine powered cogeneration systems. Energy Convers Manag 2008;49:2026e31. Brzustowski TA, Brena A. Second law analysis of energy processes. Trans Can Soc Mech Eng 1986;10:121e8. Çengel YA. Heat transfer: a practical approach. 2002. Wais P. Heat exchangers - basic design application. 2012. p. 343e66. Holman JP. Heat transfer. McGraw Hill; 2010. Tien W, Yeh R, Hong J. Theoretical analysis of cogeneration system for ships. Energy Convers Manag 2007;48:1965e74. Kakac S, Bergles AE, Fernandes EO. Two-phase ﬂow heat exchangers: thermalhydraulic fundamentals and design. Kluwer Academic Publishers; 1987. Turton R, Bailie RC, Whiting WB. Analysis, synthesis and design of chemical processes. fourth ed. New Jersey: Prentice Hall PTR; 2013. Chemical engineering plant cost index 2016. Chemical Engineering; 2016. p. 64. www.chemengonline.com. Yang MH, Yeh RH. Analyzing the optimization of an organic Rankine cycle system for recovering waste heat from a large marine engine containing a cooling water system. Energy Convers M