Professional Documents
Culture Documents
a r t i c l e i n f o a b s t r a c t
Article history: Different studies on both 2- and 4-stroke engines have shown how the choice of different supercharging
Received 19 March 2015 architectures can influence engine performance. Among them, architectures coupling one turbocharger
Accepted 3 June 2015 with a mechanical compressor or two turbochargers are found to be the most performing in terms of
engine output power and efficiency. However, defining the best supercharging architecture for aircraft
2-stroke engines is a quite complex task because the supercharging system as well as the ambient con-
Keywords: ditions influence the engine performance/efficiency. This is due to the close interaction between super-
Engine supercharging
charging, trapping, scavenging and combustion processes. The aim of the present work is the comparison
Supercharging architecture
2-stroke Diesel engine
between different architectures (single turbocharger, double turbocharger, single turbocharger combined
Trapping efficiency with a mechanical compressor, single turbocharger with an electrically-assisted turbocharger, with inter-
Scavenging efficiency cooler or aftercooler) designed to supercharge an aircraft 2-stroke Diesel engine for general aviation and
General aviation unmanned aerial vehicles characterized by a very high altitude operation and long fuel distance. A 1D
UAV model of the engine purposely designed has been used to compare the performance of the different
supercharging systems in terms of power, fuel consumption, and their effect on trapping and scavenging
efficiency at different altitudes.
The analysis shows that the engine target power is reached by a 2 turbochargers architecture; in this
way, in fact, the cylinder filling, and consequently the engine performance, are maximized. Moreover, it is
shown that the performance of a 2 turbochargers architecture performance can be further improved con-
necting electrically and not mechanically the low pressure compressor and turbine (electrically-assisted
turbocharger). From an energetic point of view, this system has also proved to be particularly convenient
at high engine speed and load, because it is possible to extract power from the electric turbocharger with-
out a penalty on specific fuel consumption.
Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction In 2-stroke engines, the cylinder filling and emptying phases are
driven by the difference between inlet and outlet pressure.
One of the crucial points of the use of a reciprocating internal Therefore, the supercharging system of a 2-stroke engine is respon-
combustion engine for aircraft propulsion is the performance sible for the air density increase and cylinder scavenging. Due to
decline with the increase in altitude. This is due to the negative the strict interaction among supercharging, scavenging and com-
effect of air density reduction on cylinder filling process. The solu- bustion, 2-stroke engines are very sensitive to the variation of
tion to this problem is the engine supercharging [1]. To super- engine working parameters and ambient conditions with altitude
charge a 2- and 4-stroke engines a turbocharger (TC), and/or a [2,3].
mechanical compressor, can be used: the first involves an air com- Several studies on both 2- and 4-stroke engines have shown
pressor fed by a turbine driven by exhaust gases; the second one how the supercharging architecture can influence the engine per-
consists of a compressor mechanically coupled to the engine formance. Pohorelsky et al. [4] simulated numerically different
crankshaft [1]. supercharging systems, characterized by positive displacement or
dynamic compressors mechanically coupled to the engine before
or after a TC with Waste-Gate (WG) or variable nozzle turbine to
⇑ Corresponding author at: Systems for Energy and Environment, University of ensure power and scavenging targets of 2-stroke Diesel engines.
Salento, Dept. Engineering for Innovation (CREA), Via per Monteroni, 73100 Lecce, They proved that the best configuration is to place the positive dis-
Italy. Tel.: +39 0832 297751, mobile: +39 340 6925653; fax: +39 0832 297777. placement compressor upstream the TC with WG. Systems with
E-mail address: paolo.carlucci@unisalento.it (A.P. Carlucci).
http://dx.doi.org/10.1016/j.enconman.2015.06.009
0196-8904/Ó 2015 Elsevier Ltd. All rights reserved.
A.P. Carlucci et al. / Energy Conversion and Management 101 (2015) 470–480 471
Nomenclature
electric compressor and Electrically Assisted Turbocharger (EAT) Divekar et al. [11] simulated the performance of a supercharg-
required excessive electric power. ing scheme decoupling the intake-boost system from the exhaust
Knoll [5] modeled numerically and compared different super- recovery system using electric supercharging and
charging architectures for a 2-stroke engine prototype: a mechan- turbo-generation with two separate electric machines and a dedi-
ical compressor, a TC coupled with a crankcase scavenging pump, cated energy storage buffer. In this way, the system transient
and a TC combined with an external mechanical compressor. The response is faster if compared to a conventional turbocharged sys-
last one proved to be the best option, thanks to the compression tem, and a reduction in fuel consumption over driving cycles char-
power reduction, and beneficial effect of the exhaust backpressure acterized by frequent transient operation is obtained. These
on trapping efficiency. conclusions are confirmed by Katrasnik et al. [12].
Mattarelli [6] analyzed two different architectures for a Terdich et al. [13] showed that the hybridization of the TC is the
4-stroke Diesel engine. The first one comprises two TCs; the second most effective way to recover the transient response of a TC; VGT,
one comprises a TC and a mechanical positive displacement com- moreover, decreases the fuel consumption. Mamalis et al. [14]
pressor. The analysis revealed that, using a mechanical compressor showed that the system with two TCs leads to the best extension
and a Variable Geometry Turbocharger (VGT), the engine perfor- of the operational range of a 4-stroke Homogeneous Combustion
mance at full-load are improved together with the fuel consump- Compression Ignition engine, i.e. an engine in which well-mixed
tion and pollutants control [7]. fuel and oxidizer are compressed to the point of auto-ignition.
Chen et al. [8] studied the influence of two-stage turbocharging The aim of the present work is to compare different architec-
system parameters on the engine performance. It was found that tures – single turbocharger, double turbocharger, single tur-
engine efficiency and performance are strongly influenced by high bocharger combined with a mechanical compressor, single
pressure (HP) and Low Pressure (LP) compressors efficiency, by turbocharger with an EAT, with intercooler or aftercooler – to
intercooler efficiency and cooling water temperature. In supercharge an aircraft 2-stroke Diesel engine for general aviation
off-design operation, the turbocharging system shall be designed and Unmanned Aerial Vehicles (UAV). This engine is characterized
to ensure the two compressors to operate with a similar value of by a very high altitude operation and long fuel endurance. To carry
the compressor efficiency. A similar analysis with similar results out tests, a 1D model of the engine has been realized with AVL
has been presented by Liu et al. [9], in which a matching method BOOST software. The performance of the different supercharging
for two-stage turbocharging system has been proposed. Shan systems were compared to determine the most performing system
et al. [10] proposed a simulation method for the design and analy- in terms of power and specific fuel consumption. Moreover, the
sis of an aerial turbocharging system, composed by two TCs, two results have been validated analyzing the cylinder scavenging
intercoolers, and two turbine by-pass valves. It was concluded that, and filling processes. Lastly, the effect of the EAT compression ratio
operating the two by-pass valves correctly, it is possible to meet on the power produced or required by the EAT at part load has
the design objectives at different altitudes, and to operate the com- been also analyzed to define the best control strategies of fuel
pressors to minimize the compression work. consumption.
472 A.P. Carlucci et al. / Energy Conversion and Management 101 (2015) 470–480
Table 1 4.E+03
Intake conditions and required performance per engine bank with altitude variation. geometric flow area
Altitude p_amb T_amb Power (% referred to effecve flow area
take-off sea level) per 3.E+03
bank at 2000 rpm
[m] [kft] [bar] [°C] [%]
[mm2]
0 (sea level) 0 1.01 15 100
3050 10 0.69 6 100 2.E+03
5180 17 0.53 18 89
10,670 35 0.23 54 54
1.E+03
mtas
TE ¼ ð2Þ
mas all closed
Table 2 Fig. 3. Scavenging process (a) and following compression stroke (b) with ports and
Engine main specifications. valves closed.
(4 cylinders)
(3 plenum)
(2 aercooler)
(5 turbine) (1 compressor)
Fig. 5. Scheme of one bank of the prototype used during the experiments.
(4) through the inlet ports. After the combustion process, the
Fig. 4. Scavenging model for Uniflow system compared with perfect mixing and exhaust gases are discharged in the exhaust ducts and then at
perfect displacement models.
the turbine inlet.
In order to run the model, the values for the fuel mass flow rate
The scavenging models define SE as a function of the Delivery (14 g/s, measured by means of a AVL 733S fuel balance) and pres-
Ratio (DR), which is the ratio between mas and the air mass, at sure ratio of the compressor (1.85, obtained averaging over time
ambient conditions, ideally required to fill the engine displacement the instantaneous pressure measured downstream and upstream
(equal to the product between the ambient density and the cylin- the compressor by means of two AVL piezoresistive sensors type
der displacement). In this work, a scavenging model, correspond- LP11DA) have been set equal to those measured during the exper-
ing to Eq. (4) and shown in Fig. 4, has been implemented. This iments on a real engine prototype performed at Sea Level. Engine
model is halfway between the ideal models of perfect mixing and output power was measured by means of a Borghi & Saveri
perfect displacement, as suggested in literature [16] on the basis eddy-current dynamometer, while the air mass flow was estimated
of experimental tests on 2-stroke Uniflow Diesel engines: by means of a calibrated diaphragm.
According to real data, the characteristic settings of the model
SE ¼ 1 expð4:35 þ 1:36DR 0:22DR2 Þ ð4Þ are reported in Table 3.
Moreover, the engine friction mean effective pressure – exper-
In order to describe the combustion development, the Vibe
imentally set during motored conditions on a range of ‘‘n’’ – varies
function was used [1,15] thanks to its easiness. The Vibe function
almost linearly with ‘‘n’’ and is equal to 2.6 bar when n = 2000 rpm.
models the heat release rate into the cylinder as a function of
The engine model, and the Vibe model of the combustion phase,
CAD based only on the start of combustion angle, combustion
have been calibrated with experimental data acquired on the real
angular duration and a shape parameter.
engine prototype, operating at n = 2000 rpm and full load. In detail,
Regarding the supercharging system, each component can be
the angular start of combustion was set equal to 13 CAD ATDC,
simulated either in simplified or full mode. The simplified calcula-
the combustion angular duration was set equal to 80 CAD while
tion is based on the power balance and requires as input the
the shape parameter was fixed equal to 1.5. In Table 4, the relative
desired value for the compressor pressure ratio p2/p1; referring to
turbochargers (compressor mechanically coupled to turbine), the
power balance leads to the following equation:
" k1 # Table 3
p k 1 1 Characteristic settings of the model.
_ T gsT cpT T 3 1 4
m _C
m c T
p3 gm gsC pC 1 Aftercooler
" k1 # V 19.2 L
p2 k e 0.7
1 tcPLUS ¼ 0 ð5Þ
p1 Dp Varying with the square of the volumetric flow rate
and equal to 0.2 bar in operating conditions used for
tcPLUS, defined as the difference between the power produced by the model calibration
the turbine and the power required by the compressor, takes into Intake ports
account the possible production or demand of mechanical/electric Number per cylinder 14
Maximum cylinder 3700 mm2
power.
overall area
The full mode, instead, is based on the dynamic balance of tur- CA variation See Fig. 1
bomachineries, and therefore on the characteristic maps of turbine Exhaust valves
and compressor. Number per cylinder 2
Maximum lift 12 mm
CA variation See Fig. 2
3. Settings and calibration of the engine model for the
simulation of the different architectures Turbocharger
Model Simplified
gsC 0.7 (according to compressor and turbine
In Fig. 5, the model of the engine in its initial configuration is characteristic maps)
shown. In particular, a compressor (1) driven by the turbine (5) gsT 0.65 (according to compressor and turbine
delivers the compressed air in the aftercooler (2) and then in the characteristic maps)
gm 0.98 (as suggested in [15])
plenum (3); from the plenum, the air enters the three cylinders
474 A.P. Carlucci et al. / Energy Conversion and Management 101 (2015) 470–480
put power and specific fuel consumption of the engine when Architecture Computational Goals
supercharged with different supercharging systems at different settings
altitudes. Once assumed the symmetry and independence of the TC (Fig. 5) One bank, 1 TC and AFRcmb equal to Calculation of the
two banks, the analysis was done on a single bank to reduce the one aftercooler per 20; TC in engine
computing duty. bank simplified mode; performance once
n = 2000 rpm; fixed the
Table 5 summarizes the supercharging architectures and the simulations at 0, supercharging
targets of the simulations. During the simulations, the trapped 3050, 5180, conditions
Air/Fuel Ratio (AFRcmb) was always imposed equal to 20 if not para- 10,670 m altitude
metrized (like in TTCHybrid system, please refer to Table 5). This TTC-1, One bank, 2 TCs AFRcmb equal to Evaluation of the
choice was made because the amount of fuel burning in the engine TTC-2 and one aftercooler 20; TC in supercharging
depends on the air trapped in the cylinder, and consequently on (Fig. 7) (TTC-1) or simplified mode; conditions (mass
intercooler (TTC-2) n = 2000 rpm; flow and
the supercharging architecture.
per bank simulations at 0, compression ratio)
In all the previous architectures, the presence of the WG valve 3050, 5180, required to
was not considered. This decision was taken because the goal of 10,670 m altitude guarantee the
this analysis was to determine the supercharging system in work- target
ing conditions at maximum efficiency. performance;
comparison
between
architectures with
4. Results and discussion aftercooler (TTC-1)
and intercooler
4.1. Performance with one TC per bank (TC model) (TTC-2)
TCC, CTC One bank, 1 TC, 1 AFRcmb equal to Evaluation of the
The scheme with one TC and one aftercooler per bank (Fig. 5) (Fig. 10) mechanical 20; TC in supercharging
has been simulated setting the model as specified in Table 5. In compressor on LP simplified mode; conditions (mass
(CTC) or HP (TCC) n = 2000 rpm; flow and
Fig. 6a and following, Pb is reported in % of Pb target at sea level
side and one simulations at 0, compression ratio)
for different compressor pressure ratios and different operating aftercooler per 3050, 5180, required to
altitudes; in the same figure, the Pb targets for the different alti- bank 10,670 m altitude guarantee the
tudes are reported. In Fig. 6b and following, the engine relative target
BSFC are also reported in % of a target value calculated as average performance;
comparison
of the target values at different altitudes. It can be concluded that between
TC supercharging system does not guarantee the achievement of Pb architectures with
target operating with acceptable compressor ratios. In fact, while the mechanical
the target at sea level and 3050 m could be reached with a com- compressor on LP
(CTC) and HP (TCC)
pressor pressure ratio respectively equal to 3 and 4.2, the target
side with respect
at 5180 and 10,670 m would require a compressor pressure ratio to the TC
higher than 4.5; this value cannot be currently reached by a single
TTCHybrid One bank, 1 HP TC HP TC in complete Performance
compressor. It is therefore necessary to study alternative super- (Fig. 13) and one aftercooler mode with a evaluation for
charging architectures. per bank, 1 LP commercial map; different ‘‘n’’ (1500
compressor and 1 simulation at 0, and 2000 rpm) and
LP turbine 3050, 5180, altitudes; maps
4.2. Performance with two turbochargers (TTC model) connected to the 10,670 m altitude; setting for the EAT
engine in order to pressure ratio of and fuel mass flow
represent an EAT electric rate management
Two architectures with two turbochargers have been compared
system compressor in order to
with different position of the cooler: as aftercooler (TTC-1 model, (eCOM) and guarantee the Pb
see Fig. 7a); as intercooler (TTC-2 model, see Fig. 7b). AFRcmb target for all the
Considering the limitations in weight and bulk, the possibility parametrized operating
of having a second cooler per bank was not contemplated. conditions
Scheme TTC-1 (a), theoretically identical to TC scheme, has been
analyzed to highlight that the two compressors would work with
compression ratios acceptable for the current state of the art in
order to reach the engine output targets.
the HP compressor pressure ratio is slightly lower than the LP com-
The model was set as specified in Table 5, while Fig. 7 also
pressor pressure ratio. As reported by Liu et al. [9], this is also the
reports the compressor pressure ratios used for simulations.
working condition corresponding to the lowest compression work.
These values not only ensure the achievement of the Pb target at
In Fig. 8, the behavior of the two architectures is compared.
different altitudes; they also correspond to the working condition
The scheme with the intercooler (TTC-2) leads to a longer life of
with the lowest BSFC. In particular, this condition is obtained when
the second compressor, which works at a lower temperature. The
scheme with aftercooler (TTC-1), on the contrary, leads to better
Table 4 engine efficiency. In fact Fig. 8a shows that BSFC is lower with
Relative error of the computed with respect to the measured values. TTC-1 architecture than TTC-2. Since TTC-1 is also characterized
Value Relative error [%] by a higher TE (see Fig. 8b), the final result is that the compressed
air mass flow rate in TTC-1 is lower than in TTC-2. Additionally,
Engine brake power (Pb) 1.8
Brake specific fuel consumption (BSFC) +1.9 using TTC-1 requires lower overall compression ratio (Fig. 8c),
Air mass flow 10.0 and therefore a lower supercharging power in order to reach the
Turbine pressure ratio +4.2 same Pb target.
A.P. Carlucci et al. / Energy Conversion and Management 101 (2015) 470–480 475
120 125
(a) (b)
100 100
80 75
ΔBSFC [%]
%Pb@sea level
60 50
40 25
20 0
1.5 2.0 2.5 3.0 3.5 4.0 4.5
0 - 25
1.5 2.0 2.5 3.0 3.5 4.0 4.5 Compressor Pressure Ratio [-]
Compressor Pressure Ratio [-]
TC-0 TC-3050 TC-5180 TC-10670 TC-0 TC-3050 TC-5180 TC-10670
target 0-3050 m target 5180 m target 10670 m
Fig. 6. Engine brake power Pb (a) and brake specific fuel consumption BSFC (b) at different altitudes and compression ratios with TC architecture.
(4 cylinders) (4 cylinders)
(a) (b)
(3 plenum) (3 plenum)
(2 aftercooler)
(2 intercooler)
(5 –LP turbine) (5 –LP turbine)
(1 –LP compressor) (1 –LP compressor)
altitude Low pressure – high pressure altitude Low pressure – high pressure
Fig. 7. TTC-1 (a) e TTC-2, (b) schemes. In the table, the values of low pressure and high pressure compressors ratio used during simulations.
The reason of this behavior lies in the architecture of the com- 4.3. Performance with a mechanical compressor and a TC per bank
pression system; in particular, once the compression ratios and
efficiency and the cooler efficiency are fixed, the architecture with To obtain a lower bulk and weight configuration a supercharg-
intercooler causes a higher outlet (plenum) temperature; this, in ing system with a TC, a mechanical compressor on the LP side (CTC,
turn, leads to less air mass trapped into the cylinders, to a lower Fig. 10a) or on the HP side (TCC, Fig. 10b) and an aftercooler per
amount of fuel burned and lower output power. Therefore, to bank has been analyzed.
increase the air mass trapped, the outlet pressure must be The model was set as specified in Table 5. In Fig. 10, the values
increased, this leading to the higher compression ratios values of the compression ratio set for the TC and the mechanical com-
for TTC-2 architecture scheme reported in Fig. 7. This is expected pressor are reported. Similarly to TTC simulations, these values
to cause the difference between the pressure at cylinder inlet not only guarantee the attainment of the Pb target at different alti-
and exhaust to increase. In Fig. 9, the difference between the pres- tudes, they also correspond to the working condition with the low-
sure at cylinder inlet and exhaust is compared during the scaveng- est BSFC. In Fig. 11, the behavior of the two systems is compared.
ing phase with TTC-1 and TTC-2 schemes for the lowest and Still in Fig. 11, the results obtained with TTC-1 scheme are reported
highest operating altitude. As expected, this difference is higher for comparison purposes.
with TTC-2 with beneficial effects on scavenging efficiency or AP As well known, the power required by the mechanical compres-
(Fig. 8d) but detrimental effects on TE. The overall effect is there- sor is provided by the engine crankshaft. Having the power to drive
fore a better behavior of TTC-1 scheme, becoming more evident the mechanical compressor, the indicated power must be higher
increasing the altitude. This scheme has been chosen for further than the target value. Therefore, a higher quantity of air and fuel
considerations. must be trapped and injected into the cylinder. Consequently, a
476 A.P. Carlucci et al. / Energy Conversion and Management 101 (2015) 470–480
10
BSFC [%] 0.7
TE [ ]
5
0.6
0
TTC 1
TTC 1
TTC 2 TTC 2
5 0.5
0 3000 6000 9000 12000 0 3000 6000 9000 12000
altitude [m] altitude [m]
(c) 11
(d) 0.94
Overall Compressor
Pressure Ratio [ ]
9
0.93
7
AP [ ]
0.92
5
0.91
3 TTC 1
TTC 1
TTC 2 TTC 2
1 0.90
0 3000 6000 9000 12000 0 3000 6000 9000 12000
(4 cylinders) (4 cylinders)
(3 plenum) (3 plenum)
(2 aftercooler) (2 aftercooler)
(1 –LP compressor)
(a) (b)
altitude Turbocharger – mechanical compr. altitude Turbocharger – mechanical compr.
Fig. 10. TCC (a) e CTC (b) schemes. In the table, the values of turbocharger and mechanical compressor ratio used during simulations.
(a) 40
TTC 1
(b) 5
TTC 1
30 TCC 4 TCC
CTC CTC
BSFC [%]
20
pplenum [bar]
10 2
0 1
10 0
0 3000 6000 9000 12000 0 3000 6000 9000 12000
altitude [m] altitude [m]
TCC 0.93
CTC
0.60
0.91 TTC 1
TCC
CTC
0.50 0.89
0 3000 6000 9000 12000 0 3000 6000 9000 12000
altitude [m] altitude [m]
n 3
simulations have been run varying the load at different ‘‘n’’ (1500
PðnÞ ¼ P n¼2000 ð7Þ
and 2000 rpm) and altitudes. The power is that required by the 2000
propeller directly coupled to the engine crankshaft, whose varia- Concerning the HP TC, the full mode has been used, as previ-
tion with ‘‘n’’ is described by the Renard law: ously specified in Table 5. This choice was made since the aim of
these simulations was to describe the system behavior at part
PðnÞ ¼ C P qa n3 R5 ð6Þ loads, where the turbomachineries cannot be assumed working
Knowing the values of the power required at the different alti- at design point. The full mode has been implemented based on
tudes at 2000 rpm (target values), it is possible to know the power commercial maps for the turbomachineries, and chosen consider-
required at 1500 rpm as: ing the mass flow rates and compression ratios calculated with
478 A.P. Carlucci et al. / Energy Conversion and Management 101 (2015) 470–480
0.9
0.75 0 TTC 1
0 CTC
0.6 0 TTC
pinlet pexhaust [bar]
0.45
0.3
0.15
0
0.15
0.3
IPO IPC
0.45
BDC
0.6
80 100 120 140 160 180 200 220 240
CAD ATDC
Fig. 12. Difference between the pressure at cylinder inlet and exhaust during the
scavenging phase; comparison between TTC-1, CTC and TCC architectures at sea
level.
(4 cylinders)
(3 plenum)
(2 aercooler)
(6 –mech.
conn. between
LP TC and
engine shas) tcPLUS<0 (1’ – HP compressor)
tcPLUS>0
(5’ – HP turbine)
(1 – LP compressor)
(5 – LP turbine)
Fig. 13. TTCHybrid scheme. tcPLUS > 0 if the power produced by the turbine
exceeds the power required by the compressor; tcPLUS < 0 if the power required by
the compressor exceeds the power produced by the turbine.
TTC-1 model; in particular, the maps of the selected TC are Fig. 14. Compressor and turbine characteristic maps of the chosen commercial
reported in Fig. 14. In this way, no assumptions were required con- turbocharger.
cerning the value of the pressure ratio of the HP compressor.
Moreover, a TC mechanical efficiency of 0.98 for the HP TC has
been set [15].
As for the LP side of the supercharging system, given the double
electric connection, a mechanical efficiency equal to 0.97 has been
% Pb@sealevel
used both on the compressor and on the turbine sides, resulting in
a 0.95 overall mechanical efficiency. During the simulations, the
pressure ratio of the eCOM was parameterized in the range 1–2. 160
A further limit on the maximum power required by the eCOM 120
equal to 10 kW has been imposed for design constraints of the
electric system. Finally, in order to vary the engine load, AFRcmb 80
has been varied in the range 20–35. 40
In this way, a 3D surface has been obtained for each value of
altitude and ‘‘n’’, fitting the power delivered by the engine for each
value of both AFRcmb and eCOM pressure ratio; this surface, inter-
sected with a planar surface at constant power equal to Pb target,
gives the values of PR,eCOM and AFRcmb corresponding to the equi-
librium engine and load (see for example Fig. 15). Among these val-
Fig. 15. 3D surface plotting engine brake power as a function of both AFRcmb and
ues, excluding those related to a power of the eCOM higher than eCOM pressure ratio for n = 2000 rpm at Sea Level (the dotted line is the
10 kW, it is possible to plot the graphs shown in the following intersection between the 3D surface and the Pb = Pb@sea level plane).
A.P. Carlucci et al. / Energy Conversion and Management 101 (2015) 470–480 479
BSFC [%]
PR,eCOM [-]
PR,eCOM [-]
tcPLUS [kW]
BSFC [%]
tcPLUS [kW]
+15.4 1.25 +25.0 1.5 0
-5
-23
0 [m] 1 10670 [m]
-10 +15.4 1 -10
20 23.3 26.6 30 20 24 28 32
AFRcmb [-] AFRcmb [-]
Fig. 16. BSFC, tcPLUS and PR,eCOM at different altitudes for a constant output power at 1500 rpm.
PR,eCOM [-]
BSFC [%]
BSFC [%]
tcPLUS [kW]
tcPLUS [kW]
tcPLUS [kW]
-4.5838 1.385 0.65 +8.78 1.9 0.5
Fig. 17. BSFC, tcPLUS and PR,eCOM at different altitudes for a constant output power at 2000 rpm.
compressor, EAT, with intercooler or aftercooler. This engine must [3] Ravi M, Marathe A. Effect of port sizes and timings on the scavenging
characteristics of a uniflow scavenged engine. SAE Paper 920782; 1992.
be characterized by a very high altitude operation and long fuel
[4] Pohorelsky L, Brynych P, Macek J. Air system conception for a downsized two-
distance. Therefore, a 1D model of the engine has been realized stroke diesel engine. SAE Paper 2012-01-0831; 2012.
with AVL BOOST software. The performance of the above men- [5] Knoll R. AVL two-stroke diesel engine. SAE Paper 981038; 1998.
tioned supercharging systems have been compared to determine [6] Mattarelli E. Comparison among different 2-stage supercharging systems for
HSDI diesel engines. SAE Paper 2009-24-0072; 2009.
the most performing one in terms of power and specific fuel con- [7] Mattarelli E, Rinaldini CA, Mazza A, Oliva M. Development of a 2-stage
sumption to supercharge an aircraft 2-stroke Diesel engine for gen- supercharging system for a HSDI diesel engine. SAE Paper 2009-01-2757;
eral aviation and UAV. Moreover, the effect of the EAT compression 2009.
[8] Chen J, Zhuge W, Zheng X, Zhang Y. Investigation of influence of two-stage
ratio on the power produced or required by the EAT at part load turbocharging system on engine performance using a pre-design model. In:
has been also analyzed to define best performing the control Proceedings of ASME Turbo Expo 2013: turbine technical conference and
strategies in terms of fuel consumption. exposition GT2013, San Antonio, Texas, USA; June 3–7, 2013.
[9] Liu Y, Zhuge W, Zhang Y, Zhang S, Zhang J, Huo X. A matching method for two-
The analysis with the two turbochargers system showed that it stage turbocharging system. In: Proceedings of ASME Turbo Expo 2014:
was possible to reach the target power for each altitude. Moreover, turbine technical conference and exposition GT2014, Dusseldorf, Germany;
positioning the air cooler after the two compression stages reduces June 16–20, 2014.
[10] Shan P, Zhou Y, Zhu D. Mathematical model of two-stage turbocharging
the specific fuel consumption. The two turbochargers architecture gasoline engine propeller propulsion system and analysis of its flying
was shown to be more effective even if compared with architec- characteristic. In: Proceedings of ASME Turbo Expo 2014: turbine technical
tures in which a single TC was used together with a mechanical conference and exposition GT2014, Dusseldorf, Germany; June 16–20, 2014.
[11] Divekar PS, Ayalew B, Prucka R. Coordinated Electric supercharging and turbo-
compressor.
generation for a diesel engine. SAE Paper 2010-10-1228; 2010.
Conclusively, a hybrid supercharging system has been analyzed. [12] Katrasnik T, Medica V, Trenc F. Analysis of the dynamic response improvement
It comprised a classic HP TC and a LP compressor and turbine elec- of a turbocharged diesel engine driven alternating current generating set.
trically connected. The analysis revealed that the hybrid super- Energy Convers Manage 2005;46:2838–55.
[13] Terdich N, Romagnoli A, Martinez-Botas RF, Pesiridis A. Mild hybridization via
charging architecture is characterized by a fuel consumption electrification of the air system: electrically assisted and variable geometry
lower than that of the two turbochargers architecture and compa- turbocharging impact on an off-road diesel engine. In: Proceedings of ASME
rable with single TC architecture; compared to this last one, the Turbo Expo 2013: turbine technical conference and exposition GT2013, San
Antonio, Texas, USA; June 3–7, 2013.
hybrid architecture is also able to reach the power targets for all [14] Mamalis S, Nair V, Andruskiewicz P, Assanis D, Babajimopoulos A. Comparison
operation altitudes. Though more complex and high in weight of different boosting strategies for homogeneous charge compression ignition
and bulk, this scheme allows to extract power from the EAT with- engines – a modeling study. SAE Paper 2010-01-0571; 2010.
[15] Avl BOOST, Users Guide for AVL BOOST – AVL List GmbH; 2008.
out a penalty on fuel consumption. [16] Blair GP. Design and simulation of two-stroke engines. Warrendale: SAE
International; 1996.
References [17] Heywood JB, Sher E. The two-stroke cycle engine: its development, operation
and design. USA: Taylor and Francis; 1999.
[18] Baines NC. Fundamentals of turbocharging. USA: Concepts NREC; 2005.
[1] Heywood J. Internal combustion engine fundamentals. McGraw-Hill; 1988.
[2] Ravi M, Marathe A. Effect of inlet and exhaust pressures on the scavenging
characteristics of a carbureted uniflow scavenged engine. SAE Paper 920840;
1992.