- Email: [email protected]
Performance analysis of a two-stage expansion air engine
Energy 115 (2016) 140e148
Contents lists available at ScienceDirect
Energy journal homepage: www.elsevier.com/locate/energy
Performance analysis of a two-stage expansion air engine Chi-Min Liu*, Chin-Lun Huang, Cheng-Kuo Sung, Chih-Yung Huang Department of Power Mechanical Engineering, National Tsing Hua University, Hsinchu, 300, Taiwan, ROC
a r t i c l e i n f o
a b s t r a c t
Article history: Received 26 November 2015 Received in revised form 5 August 2016 Accepted 4 September 2016
This study proposes an air engine consisting of one small and one large cylinder to conduct two-stage expansion in series, in which high-pressure air first expands in the small cylinder and then residualpressure air is transferred to the large cylinder for another expansion, by fully using the high-pressure air and increasing the power output and efficiency of the engine. First, mathematical models of a single-cylinder engine and a two-stage expansion engine were constructed. Second, the relations between the rotational speed and the output power, torque, efficiency, and cylinder pressure were established using MATLAB simulation software for analyzing the air engine in comparison with experimental approaches. The experimental results indicated that the two-stage expansion engine generated up to 1.7 kW of power and 12.42 Nm of torque at air pressure of 12 bar, which was superior to the performance of a single-cylinder engine. By varying the intake and exhaust timing sequences, the relations among the rotational speed, output power, torque, efficiency, and cylinder pressure were investigated. The results showed that early intake in the first cylinder improved power output by 5.3% as the speed increased, whereas early intake and exhaust in the second cylinder increased power output by 7%. © 2016 Published by Elsevier Ltd.
Keywords: Piston-type air-powered engine Two-stage expansion Effects of timing Performance Analysis Improvement
1. Introduction Reducing carbon dioxide (CO2) emissions is a crucial challenge worldwide. A major source of CO2 emissions is the exhaust gas of internal combustion (IC) engines. Studies have explored alternative energy sources that can be used in engines to reduce emissions [1]. Compressed air, as a type of medium, can be easily obtained from the power generation process of renewable energies such as solar energy, wind energy, and tidal energy [2]. In addition, compressed air is a potential alternative to battery electric systems because of its high power density, low cost, and minimal environmental impact [3]. Air engines that use compressed air as fuel have attracted substantial attention and have been investigated for feasibility in vehicles [4]. Compressed air expands in the engine cylinder, driving the piston to output work, and discharges in the form of breathable gas at low temperature. A compressed-air engine, when used as the main engine of a motor vehicle, can produce power with zero CO2 emissions [5]. Compressed air can also be used as an auxiliary energy source to enable an IC engine to operate at the optimal fuel consumption rate
* Corresponding author. E-mail address: [email protected] (C.-M. Liu). http://dx.doi.org/10.1016/j.energy.2016.09.023 0360-5442/© 2016 Published by Elsevier Ltd.
when powering the air compressor, thus improving the IC engine fuel efficiency and reducing CO2 emissions [6]. Compressed-air engines use a conventional IC engine to activate an onboard air compressor that provides compressed air to an onboard air motor, which serves as the main power system for motor vehicles. This approach allows the IC engine to operate at the optimal fuel consumption rate to improve its efficiency and reduce emissions while powering the air compressor [7]. This combination of an IC engine and an air motor can improve fuel efficiency by up to 22% compared with conventional IC engines [8]. A favorable feature of the compressed-air engine is its capability of recovering energy during braking to further improve the engine efficiency by using approaches such as heat recovery [9]. The success of applications using a compressed-air system, especially with regard to its advantage of easy integration with IC engines, has attracted great attention from the energy sector. The applications of a compressedair system can be further extended to a hybrid system with IC engines or electric motors, where it can serve as a secondary power system [10]. As described previously, a simple air-powered engine can provide sufficient power output to drive motor vehicle, and its energy efficiency is more favorable than that of an IC engine [11]. For full use of the air pressure, the intake stroke of an air-powered engine is shorter, which extends the expansion stroke and avoids incomplete
C.-M. Liu et al. / Energy 115 (2016) 140e148
expansion caused by taking in excessive high-pressure air. However, the high-pressure air in a single-cylinder engine cannot be reduced to atmosphere pressure after the expansion stroke. With a higher power output, the air-powered engine requires a higher intake of air pressure, and the residual pressure increases after the expansion stroke, resulting in wastage of energy if air with residual pressure is directly exhausted to the atmosphere. Liu et al. presented the architecture of an air-powered engine consisting of a large cylinder in series with a small cylinder. The exhaust of the small cylinder passed through a heat exchanger and then flowed into the large cylinder. A simulation showed that the two-cylinder design was more favorable than a single-cylinder engine when the intake pressure was 10 bar, rotation speed was 1000 rpm, phase difference between the two cylinders was 180 , and intake stroke exceeded90 [12]. The Scuderi Group designed a hybrid pneumatic engine (HPE) that divides the four strokes of a conventional internal combustion engine cycle over two paired cylinders, i.e., one compression cylinder and one power cylinder, connected by an air tank [13]. However, no experiments or in-depth discussion was reported for this special design. Therefore, this study proposes a method to improve the efficiency of air-powered engines by constructing an engine with two single cylinders in series, one small and one large, to form a two-stage expansion engine. The residualpressure air after the expansion stroke in the small cylinder is transferred to the large one, where it undergoes another expansion and performs work, resulting in two air-powered engine cycles that use high-pressure air more effectively. 2. Thermodynamic model Fig. 1 shows a two-stage expansion engine. The exhaust process in the first cylinder completely overlaps with the intake process in the second cylinder to avoid the compression caused by the exhaust stroke in the first cylinder. Because the second cylinder reuses the gas exhausted by the first cylinder, the first cylinder must produce more output to push the gas into the second cylinder if its volume is smaller than that of the second one. Assuming that the two cylinders have the same pressure while the first cylinder is exhausting gas to the second cylinder, if the first cylinder is larger than the second cylinder, then the negative torque produced by the exhaust in the first cylinder is greater than the positive torque produced by the intake in the second cylinder. Therefore, the first cylinder should be smaller than the second one. Accordingly, the volumes of the first and second cylinders were taken as 50 and 100 cm3, respectively. A thermodynamic model of the air engine is developed using compressible flow [14] and the thermodynamic models [15] of the piston-type air engine. The following assumptions are made [16].
141
First, the tank supplies high-pressure air at a fixed pressure and temperature. Second, the piston-type air engine has no heat exchange with the external environment during the expansion and compression stokes. Third, the piston operates with a simple harmonic motion. Finally, the model ignores the influence of flow resistance, leakage, and friction. 2.1. Intake stroke The Mach number is calculated by using the pressure ratio before and after the intake valve, which can be approximated as the pressure in the high-pressure air source and the pressure inside the cylinder, respectively. Then, the mass flow rate is calculated using the Mach number. The intake amount is determined from the mass flow rate multiplied by time. The cylinder pressure is then calculated by using the equation of state for an ideal gas.
(" Min ¼
Ptank Pc
m_ in ¼ Ptank
Pc ¼
g1 g
#
2 g1
)12
!
rffiffiffiffiffiffiffiffiffiffiffiffiffi
g
RTtank
(1)
AC 1þ
g1 2 2Min
mc R Tc Vc
1þg 2ð1gÞ
(2)
(3)
where Min is the Mach number of inlet air, P is the pressure, the subscripts c and in respectively indicate the cylinder and inlet air, g is the heat capacity, m_ in is the mass flow rate, R is the air constant, A is the valve area, T is the temperature, and C is the discharge coefficient (¼0.75).In the interest of simplicity, approximation through experimental tuning of the discharge coefficients was deemed sufficient [17]. 2.2. Expansion and compression strokes Because it is assumed that gas has no heat exchange with the exterior, the processes of expansion and compression are considered as isentropic processes. The pressure, temperature, and work of the whole process can be calculated by using the isentropic process formula.
Vst Vfi
Pfi ¼ Pc Tfi ¼ Tc
W¼
Pfi Pc
!g (4)
g1 g
1 Pfi Vfi Pst Vst 1g
(5)
(6)
where W is work, the subscripts fi and st respectively indicates the end and start of the isentropic process. 2.3. Exhaust stroke
Fig. 1. Schematic of two-stage expansion engine.
Because there is a greater pressure difference between the cylinder pressure and the atmospheric pressure during the exhaust process, fluid resistance is generated. Therefore, the mass flow rate is calculated by the fluid resistance method.
142
C.-M. Liu et al. / Energy 115 (2016) 140e148
m_ ¼ r* a* A* C
r* ¼ rc T * ¼ Tc
a* ¼
2 1þg
2 1þg
(7)
1
g1
(8)
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi g R T*
(9)
(10)
where * is the throat. During the calculation, the processes of four strokes (intake, expansion, exhaust, and compression) are computed in 40 million time steps. The results are then subjected to iterate in one revolution and then return to the initial condition. By the same computation, the results are substituted into a loop computation to achieve convergence in a stable state, and therefore, the theoretical computing results are closer to the actual operation of the engine. The calculated results of each stroke are entered into a matrix to plot the in-cylinder P- diagram for the engine at various speeds. The integration of the P-V area in the graph can provide the indicative power of the air engine output. In the process of exhaust from the first cylinder to the second cylinder, the Mach number is calculated by using the ratio of the pressure in the first cylinder to that in the second cylinder. The volume of the pipeline connecting the first and second cylinders is disregarded. In the determination of the total flow of air with the resultant Mach number, the first cylinder has a smaller area of exhaust valve with respect to the cross-sectional area, and the flow is limited in the first cylinder at the exhaust valve. Because the air flows from the first cylinder and goes straight into the second cylinder without passing through any connecting pipeline in the simulation, the air does not expand inside the connecting pipes and is not affected by the flow resistance of the pipes. 2.4. Efficiency The efficiency is calculated as follows [18]. The numerator is the output power during the engine operation, and the denominator is the flow work obtained from the specified intake air pressure, P, multiplied by the amount of flow, Q, during operation.
eff ¼
Power 2p t N ¼ PQ DP DPþ1 Q nor
(11)
where t is the torque, N is the revolution per second, DP is the pressure difference at the inlet/outlet of the air engine (the outlet pressure is assumed to be atmospheric pressure in the calculation), and Qnor is the air-flow rate when converted to atmospheric pressure. The two-stage expansion proposed in this study is designed to enhance the efficiency of the engine for using high-pressure gas. Thus, a single-cylinder engine (50 cm3) and a two-stage expansion engine (50 cm3 þ 100 cm3) were selected for comparison. 3. Numerical simulation of two-stage expansion engine As shown in Fig. 2, no noticeable phenomenon of uneven exhaust is found at a low speed of 1000 rpm. Under the same intake pressure at 12 bar, the total pressure volume (P-V) area of the twostage expansion engine is larger than that of the single-cylinder engine, because the gas undergoes a second expansion in the second cylinder. Therefore, the output power of the two-stage expansion engine is higher than that of a single-cylinder engine. During the exhaust stroke, because the single-cylinder engine directly exhausts gas to the atmosphere, the pressure can be reduced to 1 bar. However, because the first cylinder of the twostage expansion engine must exhaust gas to the second cylinder, the exhaust pressure decreases slowly during the exhaust process, resulting in a cylinder pressure of 3 bar at the end of the exhaust stroke. This means that the gas is not completely discharged. As a result, the intake amount of the two-stage expansion engine is less than that of the single-cylinder engine at the next intake. Because of the higher power output of the two-stage expansion engine, the efficiency of the two-stage expansion engine is better. As shown in Figs. 3e5, the power output, torque, and efficiency of the two-stage expansion engine are more favorable than those of a single-cylinder engine when the intake pressure is simulated at 8, 10, and 12 bar. However, at higher speeds, the performance of both engines is similar. Because the gas of the first cylinder has to flow into the second cylinder, it causes lower intake pressure when the back-pressure of the first cylinder becomes excessively high at high speeds. Whereas the exhaust of the single-cylinder engine is discharged directly into the atmosphere, the performance of the
Fig. 2. PeV diagram (at 12 bar and 1000 rpm) for two-stage expansion and a single-cylinder engines.
C.-M. Liu et al. / Energy 115 (2016) 140e148
Fig. 3. Power output diagram (at 8, 10, and 12 bar) for two-stage expansion and singlecylinder engines.
143
During experiments, a high-pressure air system was integrated with a 300 bar air compressor (MACO BAM06, Denver, USA). To achieve various experimental conditions, two pressure regulators (HRP-350-V and SR-290, 3Arrow) arranged in series were used to lower the pressure of the 300 bar compressed air stored in the cylinders in two stages. The compressed air was stored in a 400 L buffer tank to provide a steady airflow rate during experiments. The test bench included a 50 Nm torque transducer (TPS-A-50NM, KYOWA, Tokyo, Japan) combined with an electromagnetic brake (ZKB005AA, Chain Tail Co.) to measure the test engine's output torque and power. The electromagnetic brake applied a load between 0 and 50 Nm. Two pressure transducers (PVL and PVB, KYOWA) and two k-type thermocouples were installed at the inlet and exit of the engine, respectively, to monitor the pressure and temperature variations during experiments. To record the cylinder pressure, a high-pressure sensor (PHL-A-2MP-B, KYOWA) was used; this sensor replaced the spark plug, which was no longer used for the compressed-air operation. A flow meter (TF-4150, Tokyo Keiso Co., Tokyo, Japan) was installed at the engine inlet to record the flow rate. An optical counter (FS-N11N, Keyence, Osaka, Japan), which also functioned as a tachometer, recorded the piston location in the cylinder during the engine operation. All data acquired during experiments were transferred to a data acquisition unit (GL900, GRAPHTEC, Yokohama, Japan) for recording and further analysis. The power output was calculated as the product of the measured torque and rotational speeds. Fig. 6 shows a schematic overview of the test bench for the compressed-air engine. 4.1. Measurement uncertainty
Fig. 4. Torque diagram (at 8, 10, and 12 bar) for two-stage expansion and singlecylinder engines.
The uncertainty of the measurement system is analyzed by calculating the uncertainty from the instruments and sensors integrated in the system, and the errors can be evaluated by balancing the uncertainties of each of them [19]. When estimating the measurement uncertainty, there are factors that must be considered, including: (1) the operator proficiency, (2) instrument precision, and (3) environment [20]. First, the average value of the experimental data is calculated by summing them and dividing by the number of results ‘n’ [21].
X¼
X1 þ X2 þ … þ Xn n
(12)
where X is the arithmetic mean of the n results considered. Second, this average value is subtracted from each data to find the deviation from each measured result.
d1 ¼ X1 X; d2 ¼ X2 X; …dn ¼ Xn X
Fig. 5. Efficiency diagram (at 8, 10, and 12 bar) for two-stage expansion and singlecylinder engines.
single-cylinder engine is better at high speeds. 4. Experimental setup In this study, series-connected 50 and 100 cm3 four-stroke IC engines supplied by a Taiwanese motorcycle company, KYMCO, were used. By modifying the design of the engine, the four-stroke IC engine was changed to a two-stroke air engine. The modified air engine used high-pressure air as the driving source, and the output performance and fluid characteristics of the air engine were measured.
(13)
where dn is the deviation. Finally, the estimated standard deviation for n measurements is calculated as follows:
sffiffiffiffiffiffiffiffiffiffiffiffi P 2 dn s¼ n1
(14)
where s is the standard deviation. The measurement results in the experiment are affected by factors such as the operator proficiency, instrument precision, and other changes in the environment. As a result, the measured values deviate from the actual values. The operator proficiency is caused by random and unpredictable factors that cannot be completely eliminated, and errors can be reduced only by increasing the number of measurements. The environmental conditions are controlled in the same way as the ambient conditions in the
144
C.-M. Liu et al. / Energy 115 (2016) 140e148
Fig. 6. Experimental setup for the compressed-air engine test bench.
laboratory. The instrument precision is estimated by the instrument specifications from the manufacturer's instructions from a probability distribution correction, which are the accuracy of the torque transducer, flow meter, pressure transducer, and optical counter (as a tachometer for rotational speed). The measurement uncertainty can be obtained by calculating the aforementioned factors. As shown in Fig. 7, the experimental results obtained under 12 bar and continuous rotation at 1300 rpm from 100 cycles are recorded to determine the accuracy of the measurements as the average and deviation from the torque and efficiency per lap in the experiment.
The calculated results indicate that the average values and measurement uncertainties of the torque and efficiency are10.368 ± 0.314 Nm and 70.689 ± 2.702%, respectively. During the experiment, the instrument precision, error of engine mounting, and vibration during engine operation are factors influencing the measurement values. 5. Results and discussion The experimental investigations of a two-stage expansion air engine have been performed using a 50 and 100 cm3 four-stroke IC engine for inlet pressures varying from 8 to 12 bar. The experimental results of a single-cylinder air engine have also been measured and plotted side-by side for comparison. 5.1. Experimental comparison of two-stage expansion and singlecylinder engines
Fig. 7. Torque and efficiency of each cycle.
The P-V area comparisons of the two-stage expansion and single-cylinder engines are shown in Figs. 8 and 9. As shown in Fig. 8, the P-V area of the two-stage expansion engine is larger than that of the single-cylinder engine, according to the experimental results under 12 bar intake pressure and 1000-rpm speed. However, Fig. 9 indicates that the obvious rough intake results in reduced output power as the two-stage expansion engine operates at 12 bar intake pressure and 2500 rpm speed, which may be attributed to the excessively high back-pressure of the first cylinder. In addition, the differences of the pressure drop between the experimental data (in Figs. 8 and 9) and the simulation data (in
C.-M. Liu et al. / Energy 115 (2016) 140e148
Fig. 8. PeV diagram (at 12 bar and 1000 rpm) for two-stage expansion and singlecylinder engines.
145
Fig. 11. Torque diagram (at 8, 10, and 12 bar) for two-stage expansion and singlecylinder engines.
generated when the back-pressure of the first cylinder becomes excessively high at high speeds. However, the exhaust of the singlecylinder engine is discharged directly into the atmosphere; thus, the performance of the single-cylinder engine is more favorable at high speeds. As shown in Fig. 12, the efficiency of the two-stage expansion engine is higher than that of the single-cylinder engine at 2000 rpm or less, and the efficiency of the two-stage expansion engine is lower than that of the single-cylinder engine when the speed is higher than 2000 rpm because of the increased backpressure of the exhaust of the first cylinder and the decreased intake pressure in the second cylinder. 5.2. Effects of timing
Fig. 2) are attributed to the volume of the connecting manifold between the first and the second cylinders, which was disregarded in the simulation for ease of calculation. The volume of the connecting manifold is large compared to that of the cylinders in the two-stage expansion air engine; therefore, the pressure drop during the expansion process is small. As shown in Figs. 10 and 11, the performance of the two-stage expansion engine is less favorable than that of the single-cylinder engine after ~2500 rpm. Because the gas from the first cylinder must flow into the second cylinder, lower intake pressure is
As observed from the simulation and experiment of the twostage expansion engine, the pressure of either the first or second cylinder tends to increase relatively slowly, with inadequate intake during the intake process, resulting in rough exhaust and high back-pressure during the exhaust process. Tai et al. used four fullyflexible cam-less valves (two intakes and two exhausts) for each cylinder, where one intake valve is switchable and connects either the intake manifold or the air tank to the cylinder through a threeway valve. Their air-hybrid engine improved fuel efficiency by 64% for city driving and 12% for highway driving [22]. Trajkovic et al. examined the intake and exhaust systems using the conventional crankshaft-driven cam system [23]. To ensure smooth running and fast response, the airflow was controlled using a simple cam mechanism in the compressed air system. For continuous running
Fig. 10. Power output diagram (at 8, 10, and 12 bar) for two-stage expansion and single-cylinder engines.
Fig. 12. Efficiency diagram (at 8, 10, and 12 bar) for two-stage expansion and singlecylinder engines.
Fig. 9. PeV diagram (at 12 bar and 2500 rpm) for two-stage expansion and singlecylinder engines.
146
C.-M. Liu et al. / Energy 115 (2016) 140e148
operation and fast response, the air flow is simply controlled by a proper cam design mechanism in many corresponding compressed-air engine designs [24]. Therefore, to improve the overall efficiency, timing adjustments are necessary to ensure that the pressure can rapidly increase during the intake process and that the back-pressure in the cylinder can be quickly eliminated during the exhaust process.
5.2.1. Early intake in the first cylinder As determined from the simulation diagrams in Figs. 13 and 14, the greater the degree of intake, the higher is the efficiency at higher speeds. If the degree of intake is increased earlier, the gas in the cylinder would be compressed by the upward piston during intake. Therefore, the pressure would increase rapidly, and the overall efficiency would increase. An excessively large angle of advance would cause the overall power to decrease because the intake process takes place while the piston starts moving up, resulting in an increase in the negative power. Owing to insufficient intake time during the stroke, the piston begins to move down without maximum pressure inside the cylinder. However, if the intake begins earlier at the point where the piston approximates the TDC, then the pressure can reach its maximum earlier during the down stroke of the piston, thus increasing the power. This improvement becomes more obvious with an increase in the rotation speed. As shown in Figs. 15 and 16, early intake in the first cylinder can increase the total power output to 1.79 kW (without early intake: 1.7 kW; 5.3% increase).
Fig. 15. Power and torque output for an early intake of 0 and 20 with 12 bar air supply pressure.
Fig. 16. Output efficiency for an early intake of 0 and 20 with 12 bar air supply pressure.
Fig. 13. Power output diagram (at 12 bar) for different degrees in the first cylinder early intake.
5.2.2. Early intake and exhaust in the second cylinder As shown in Fig. 9, the pressure in the second cylinder increases slowly during the intake process, and the back-pressure increases during the exhaust process at high speeds. Therefore, advancing the intake timing in the second cylinder can improve the intake process; on the other hand, advancing the exhaust timing can reduce the backpressure. As shown in the simulation diagrams in Figs. 17 and 18, early intake and exhaust in the second cylinder can increase its power output and efficiency during the exhaust process from the first to
Fig. 14. Efficiency diagram (at 12 bar) for different degrees in the first cylinder early intake.
Fig. 17. Power output diagram (at 12 bar) for various degrees of early intake and exhaust in the second cylinder.
C.-M. Liu et al. / Energy 115 (2016) 140e148
147
in the second cylinder may result in a long compression stroke, and the overall performance will be poor. Early intake and exhaust in the second cylinder can increase the total power output to 1.82 kW (without early intake and exhaust: 1.7 kW; 7% increase). 6. Conclusions
Fig. 18. Efficiency diagram (at 12 bar) for various degrees of early intake and exhaust in the second cylinder.
Fig. 19. Power and torque output for an early intake and exhaust of 0 and 20 with 12 bar air supply pressure.
the second cylinders. However, if the intake is too early, the negative output caused by the second cylinder increases. In this case, the piston compresses during intake, reducing the rate of increase of the overall power output. As shown in Figs. 19 and 20, early intake and exhaust in the second cylinder is intended to increase the pressure during the intake process and reduce the back-pressure during the exhaust process, both caused by rough exhaust. Early intake and exhaust are more favorable when the speed is increased; because early exhaust
Fig. 20. Efficiency for an early intake and exhaust of 0 and 20 with 12 bar air supply pressure.
In a comparison of two-stage expansion and single-cylinder engines, the simulations and experiments showed that the twostage expansion engine is more favorable than the single-cylinder engine at low speeds. However, at high speeds, because the gas in the first cylinder cannot smoothly exhaust to the second cylinder, the power output of the two-stage expansion engine becomes similar to that of the single-cylinder engine. Through experiments, it became obvious that at high speeds, because the gas in the first cylinder cannot smoothly exhaust to the second cylinder, the power output and efficiency of the two-stage expansion engine are less favorable than those in the single-cylinder engine. Therefore, the two-stage expansion engine is more favorable for applications with low speed and high torque. The two-stage expansion engine can generate up to 1.7 kW of power and 12.42 Nm of torque at air pressure of 12 bar, which is more favorable than the performance of a single-cylinder engine. Through different intake and exhaust timing sequences, early intake in the first cylinder is found to improve the power output by 5.3% as the speed increases, whereas early intake and exhaust in the second cylinder can increase the power output by 7%. Timing adjustments are needed to address the challenge of rough intake and exhaust, which are obvious at high speeds. The results of both the simulation and the experiment indicate that the higher the speed, the greater is the improvement. Therefore, the camshaft valve can be changed to a solenoid valve, and the valve timing can be adjusted in concert with the speed to achieve the greatest efficiency. Furthermore, the two-stage expansion engine can be employed as an auxiliary system in conventional gasoline engines during idle-stop/start and regenerative braking. Acknowledgments The authors would like to thank Taiwanese motorcycle manufacturer KYMCO for providing a 50 cm3 and a 100 cm3 IC engines used in this study. References [1] Drury E, Denholm P, Sioshansi R. The value of compressed air energy storage in energy and reserve markets. Energy 2011:4959e73. [2] Foley A, Lobera ID. Impacts of compressed air energy storage plant on an electricity market with a large renewable energy portfolio. Energy 2013: 85e94. [3] Agarwalla DK, Sethi S. Estimation of run time parameters of compressed air engine prototype. Int J Enhanc Res Sci Technol Eng 2014:108e12. [4] Huang C-Y, Hu C-K, Yu C-J, Sung C-K. Experimental investigation on the performance of a compressed-air driven piston engine. Energies 2013: 1731e45. [5] Papson A, Creutzig F, Schipper L. Compressed air vehicles: drive-cycle analysis of vehicle performance, environmental impacts, and economic costs. J Transp Res Board 2010:67e74. [6] Higelin P, Charlet A, Chamaillard Y. Thermodynamic simulation of a hybrid pneumatic-combustion engine concept. Int J Appl Thermodyn 2002:1e11. [7] Schechter MM. New cycles for automobile engines. 1999-01-0623. SAE Technical Paper; 1999. [8] Fazeli A, Khajepour A, Devaud C, Azad NL. A new air hybrid engine using throttle control. 2009-01-1319. SAE Technical Paper; 2009. [9] Li D, Xu H, Wang L, Fan Z, Dou W, Yu X. Simulation and analysis of a hybrid pneumatic engine based on in-cylinder waste heat recovery. 2014-01-2355. SAE Technical Paper; 2014. [10] Yu Q, Cai M, Shi Y, Yuan C. Dimensionless study on efficiency and speed characteristics of a compressed air engine. J Energy Resour Technol 2015:1e9. [11] Kumar V, Takkar J, Chitransh M, Kumar N, Banka U, Gupta U. Development of an advanced compressed air engine kit for small engine. 2014-01-1666. SAE
148
C.-M. Liu et al. / Energy 115 (2016) 140e148
Technical Paper; 2014. [12] Liu H, Chen Y, Tao Gl. Modeling and simulation of two-stage expansion airpowered engine. Journal of Zhejiang University; 2005. [13] Phillips F, Gilbert I, Pirault J-P, Megel M. Scuderi split cycle research engine: overview, architecture and operation. SAE Int J Engines; 2011. p. 450e66. 2011-01-0403. [14] Anderson J. Modern compressible flow: with historical perspective. third ed. 2002. [15] Borgnakke C, Sonntag RE. Fundamentals of thermodynamics. seventh ed. 2008. [16] Liu C-M, You J-J, Sung C-K, Huang C-Y. Modified intake and exhaust system for piston-type compressed air engines. Energy 2015:516e24. [17] Brown TL, Atluri VP, Schmiedeler JP. A low-cost hybrid drivetrain concept based on compressed air energy storage. Appl Energy 2014:477e89. [18] Huang KD, Tzeng S-C. Development of a hybrid pneumatic-power vehicle. Appl Energy 2005:47e59.
L. Experimental analysis regarding the measurement [19] Tes¸ileanu DM, Niculit¸a uncertainty of welded joints with ultrasound testing. Romanian Rev Precis Mech Opt Mecatronics 2013:23e31. [20] Abdelkarim MF, Nasrat LS, Elkhodary SM, Soliman AM, Hassan AM, Mansour sh. Volume resistivity and mechanical behavior of epoxy nanocomposite materials. Eng Technol Appl Sci Res 2015:775e80. [21] Birch K. Estimating uncertainties in testing. Good Pract Guide No 36. 2003. [22] Tai C, Tsao T-C, Levin MB, Barta G, Schechter MM. Using camless valvetrain for air hybrid optimization. 2003-01-0038. SAE Technical Paper; 2003. [23] Trajkovic S, Tunestål P, Johansson B. Investigation of different valve geometries and vavle timing strategies and their effect on regenerative efficiency for a pneumatic hybrid with variable valve actuation. SAE Int J Fuels Lubr; 2008. p. 1206e23. 2008-01-1715. [24] Yu Q, Shi Y, Ca M. Working characteristics of variable intake valve in compressed air engine. Sci World J 2014:1e9.
Contents lists available at ScienceDirect
Energy journal homepage: www.elsevier.com/locate/energy
Performance analysis of a two-stage expansion air engine Chi-Min Liu*, Chin-Lun Huang, Cheng-Kuo Sung, Chih-Yung Huang Department of Power Mechanical Engineering, National Tsing Hua University, Hsinchu, 300, Taiwan, ROC
a r t i c l e i n f o
a b s t r a c t
Article history: Received 26 November 2015 Received in revised form 5 August 2016 Accepted 4 September 2016
This study proposes an air engine consisting of one small and one large cylinder to conduct two-stage expansion in series, in which high-pressure air first expands in the small cylinder and then residualpressure air is transferred to the large cylinder for another expansion, by fully using the high-pressure air and increasing the power output and efficiency of the engine. First, mathematical models of a single-cylinder engine and a two-stage expansion engine were constructed. Second, the relations between the rotational speed and the output power, torque, efficiency, and cylinder pressure were established using MATLAB simulation software for analyzing the air engine in comparison with experimental approaches. The experimental results indicated that the two-stage expansion engine generated up to 1.7 kW of power and 12.42 Nm of torque at air pressure of 12 bar, which was superior to the performance of a single-cylinder engine. By varying the intake and exhaust timing sequences, the relations among the rotational speed, output power, torque, efficiency, and cylinder pressure were investigated. The results showed that early intake in the first cylinder improved power output by 5.3% as the speed increased, whereas early intake and exhaust in the second cylinder increased power output by 7%. © 2016 Published by Elsevier Ltd.
Keywords: Piston-type air-powered engine Two-stage expansion Effects of timing Performance Analysis Improvement
1. Introduction Reducing carbon dioxide (CO2) emissions is a crucial challenge worldwide. A major source of CO2 emissions is the exhaust gas of internal combustion (IC) engines. Studies have explored alternative energy sources that can be used in engines to reduce emissions [1]. Compressed air, as a type of medium, can be easily obtained from the power generation process of renewable energies such as solar energy, wind energy, and tidal energy [2]. In addition, compressed air is a potential alternative to battery electric systems because of its high power density, low cost, and minimal environmental impact [3]. Air engines that use compressed air as fuel have attracted substantial attention and have been investigated for feasibility in vehicles [4]. Compressed air expands in the engine cylinder, driving the piston to output work, and discharges in the form of breathable gas at low temperature. A compressed-air engine, when used as the main engine of a motor vehicle, can produce power with zero CO2 emissions [5]. Compressed air can also be used as an auxiliary energy source to enable an IC engine to operate at the optimal fuel consumption rate
* Corresponding author. E-mail address: [email protected] (C.-M. Liu). http://dx.doi.org/10.1016/j.energy.2016.09.023 0360-5442/© 2016 Published by Elsevier Ltd.
when powering the air compressor, thus improving the IC engine fuel efficiency and reducing CO2 emissions [6]. Compressed-air engines use a conventional IC engine to activate an onboard air compressor that provides compressed air to an onboard air motor, which serves as the main power system for motor vehicles. This approach allows the IC engine to operate at the optimal fuel consumption rate to improve its efficiency and reduce emissions while powering the air compressor [7]. This combination of an IC engine and an air motor can improve fuel efficiency by up to 22% compared with conventional IC engines [8]. A favorable feature of the compressed-air engine is its capability of recovering energy during braking to further improve the engine efficiency by using approaches such as heat recovery [9]. The success of applications using a compressed-air system, especially with regard to its advantage of easy integration with IC engines, has attracted great attention from the energy sector. The applications of a compressedair system can be further extended to a hybrid system with IC engines or electric motors, where it can serve as a secondary power system [10]. As described previously, a simple air-powered engine can provide sufficient power output to drive motor vehicle, and its energy efficiency is more favorable than that of an IC engine [11]. For full use of the air pressure, the intake stroke of an air-powered engine is shorter, which extends the expansion stroke and avoids incomplete
C.-M. Liu et al. / Energy 115 (2016) 140e148
expansion caused by taking in excessive high-pressure air. However, the high-pressure air in a single-cylinder engine cannot be reduced to atmosphere pressure after the expansion stroke. With a higher power output, the air-powered engine requires a higher intake of air pressure, and the residual pressure increases after the expansion stroke, resulting in wastage of energy if air with residual pressure is directly exhausted to the atmosphere. Liu et al. presented the architecture of an air-powered engine consisting of a large cylinder in series with a small cylinder. The exhaust of the small cylinder passed through a heat exchanger and then flowed into the large cylinder. A simulation showed that the two-cylinder design was more favorable than a single-cylinder engine when the intake pressure was 10 bar, rotation speed was 1000 rpm, phase difference between the two cylinders was 180 , and intake stroke exceeded90 [12]. The Scuderi Group designed a hybrid pneumatic engine (HPE) that divides the four strokes of a conventional internal combustion engine cycle over two paired cylinders, i.e., one compression cylinder and one power cylinder, connected by an air tank [13]. However, no experiments or in-depth discussion was reported for this special design. Therefore, this study proposes a method to improve the efficiency of air-powered engines by constructing an engine with two single cylinders in series, one small and one large, to form a two-stage expansion engine. The residualpressure air after the expansion stroke in the small cylinder is transferred to the large one, where it undergoes another expansion and performs work, resulting in two air-powered engine cycles that use high-pressure air more effectively. 2. Thermodynamic model Fig. 1 shows a two-stage expansion engine. The exhaust process in the first cylinder completely overlaps with the intake process in the second cylinder to avoid the compression caused by the exhaust stroke in the first cylinder. Because the second cylinder reuses the gas exhausted by the first cylinder, the first cylinder must produce more output to push the gas into the second cylinder if its volume is smaller than that of the second one. Assuming that the two cylinders have the same pressure while the first cylinder is exhausting gas to the second cylinder, if the first cylinder is larger than the second cylinder, then the negative torque produced by the exhaust in the first cylinder is greater than the positive torque produced by the intake in the second cylinder. Therefore, the first cylinder should be smaller than the second one. Accordingly, the volumes of the first and second cylinders were taken as 50 and 100 cm3, respectively. A thermodynamic model of the air engine is developed using compressible flow [14] and the thermodynamic models [15] of the piston-type air engine. The following assumptions are made [16].
141
First, the tank supplies high-pressure air at a fixed pressure and temperature. Second, the piston-type air engine has no heat exchange with the external environment during the expansion and compression stokes. Third, the piston operates with a simple harmonic motion. Finally, the model ignores the influence of flow resistance, leakage, and friction. 2.1. Intake stroke The Mach number is calculated by using the pressure ratio before and after the intake valve, which can be approximated as the pressure in the high-pressure air source and the pressure inside the cylinder, respectively. Then, the mass flow rate is calculated using the Mach number. The intake amount is determined from the mass flow rate multiplied by time. The cylinder pressure is then calculated by using the equation of state for an ideal gas.
(" Min ¼
Ptank Pc
m_ in ¼ Ptank
Pc ¼
g1 g
#
2 g1
)12
!
rffiffiffiffiffiffiffiffiffiffiffiffiffi
g
RTtank
(1)
AC 1þ
g1 2 2Min
mc R Tc Vc
1þg 2ð1gÞ
(2)
(3)
where Min is the Mach number of inlet air, P is the pressure, the subscripts c and in respectively indicate the cylinder and inlet air, g is the heat capacity, m_ in is the mass flow rate, R is the air constant, A is the valve area, T is the temperature, and C is the discharge coefficient (¼0.75).In the interest of simplicity, approximation through experimental tuning of the discharge coefficients was deemed sufficient [17]. 2.2. Expansion and compression strokes Because it is assumed that gas has no heat exchange with the exterior, the processes of expansion and compression are considered as isentropic processes. The pressure, temperature, and work of the whole process can be calculated by using the isentropic process formula.
Vst Vfi
Pfi ¼ Pc Tfi ¼ Tc
W¼
Pfi Pc
!g (4)
g1 g
1 Pfi Vfi Pst Vst 1g
(5)
(6)
where W is work, the subscripts fi and st respectively indicates the end and start of the isentropic process. 2.3. Exhaust stroke
Fig. 1. Schematic of two-stage expansion engine.
Because there is a greater pressure difference between the cylinder pressure and the atmospheric pressure during the exhaust process, fluid resistance is generated. Therefore, the mass flow rate is calculated by the fluid resistance method.
142
C.-M. Liu et al. / Energy 115 (2016) 140e148
m_ ¼ r* a* A* C
r* ¼ rc T * ¼ Tc
a* ¼
2 1þg
2 1þg
(7)
1
g1
(8)
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi g R T*
(9)
(10)
where * is the throat. During the calculation, the processes of four strokes (intake, expansion, exhaust, and compression) are computed in 40 million time steps. The results are then subjected to iterate in one revolution and then return to the initial condition. By the same computation, the results are substituted into a loop computation to achieve convergence in a stable state, and therefore, the theoretical computing results are closer to the actual operation of the engine. The calculated results of each stroke are entered into a matrix to plot the in-cylinder P- diagram for the engine at various speeds. The integration of the P-V area in the graph can provide the indicative power of the air engine output. In the process of exhaust from the first cylinder to the second cylinder, the Mach number is calculated by using the ratio of the pressure in the first cylinder to that in the second cylinder. The volume of the pipeline connecting the first and second cylinders is disregarded. In the determination of the total flow of air with the resultant Mach number, the first cylinder has a smaller area of exhaust valve with respect to the cross-sectional area, and the flow is limited in the first cylinder at the exhaust valve. Because the air flows from the first cylinder and goes straight into the second cylinder without passing through any connecting pipeline in the simulation, the air does not expand inside the connecting pipes and is not affected by the flow resistance of the pipes. 2.4. Efficiency The efficiency is calculated as follows [18]. The numerator is the output power during the engine operation, and the denominator is the flow work obtained from the specified intake air pressure, P, multiplied by the amount of flow, Q, during operation.
eff ¼
Power 2p t N ¼ PQ DP DPþ1 Q nor
(11)
where t is the torque, N is the revolution per second, DP is the pressure difference at the inlet/outlet of the air engine (the outlet pressure is assumed to be atmospheric pressure in the calculation), and Qnor is the air-flow rate when converted to atmospheric pressure. The two-stage expansion proposed in this study is designed to enhance the efficiency of the engine for using high-pressure gas. Thus, a single-cylinder engine (50 cm3) and a two-stage expansion engine (50 cm3 þ 100 cm3) were selected for comparison. 3. Numerical simulation of two-stage expansion engine As shown in Fig. 2, no noticeable phenomenon of uneven exhaust is found at a low speed of 1000 rpm. Under the same intake pressure at 12 bar, the total pressure volume (P-V) area of the twostage expansion engine is larger than that of the single-cylinder engine, because the gas undergoes a second expansion in the second cylinder. Therefore, the output power of the two-stage expansion engine is higher than that of a single-cylinder engine. During the exhaust stroke, because the single-cylinder engine directly exhausts gas to the atmosphere, the pressure can be reduced to 1 bar. However, because the first cylinder of the twostage expansion engine must exhaust gas to the second cylinder, the exhaust pressure decreases slowly during the exhaust process, resulting in a cylinder pressure of 3 bar at the end of the exhaust stroke. This means that the gas is not completely discharged. As a result, the intake amount of the two-stage expansion engine is less than that of the single-cylinder engine at the next intake. Because of the higher power output of the two-stage expansion engine, the efficiency of the two-stage expansion engine is better. As shown in Figs. 3e5, the power output, torque, and efficiency of the two-stage expansion engine are more favorable than those of a single-cylinder engine when the intake pressure is simulated at 8, 10, and 12 bar. However, at higher speeds, the performance of both engines is similar. Because the gas of the first cylinder has to flow into the second cylinder, it causes lower intake pressure when the back-pressure of the first cylinder becomes excessively high at high speeds. Whereas the exhaust of the single-cylinder engine is discharged directly into the atmosphere, the performance of the
Fig. 2. PeV diagram (at 12 bar and 1000 rpm) for two-stage expansion and a single-cylinder engines.
C.-M. Liu et al. / Energy 115 (2016) 140e148
Fig. 3. Power output diagram (at 8, 10, and 12 bar) for two-stage expansion and singlecylinder engines.
143
During experiments, a high-pressure air system was integrated with a 300 bar air compressor (MACO BAM06, Denver, USA). To achieve various experimental conditions, two pressure regulators (HRP-350-V and SR-290, 3Arrow) arranged in series were used to lower the pressure of the 300 bar compressed air stored in the cylinders in two stages. The compressed air was stored in a 400 L buffer tank to provide a steady airflow rate during experiments. The test bench included a 50 Nm torque transducer (TPS-A-50NM, KYOWA, Tokyo, Japan) combined with an electromagnetic brake (ZKB005AA, Chain Tail Co.) to measure the test engine's output torque and power. The electromagnetic brake applied a load between 0 and 50 Nm. Two pressure transducers (PVL and PVB, KYOWA) and two k-type thermocouples were installed at the inlet and exit of the engine, respectively, to monitor the pressure and temperature variations during experiments. To record the cylinder pressure, a high-pressure sensor (PHL-A-2MP-B, KYOWA) was used; this sensor replaced the spark plug, which was no longer used for the compressed-air operation. A flow meter (TF-4150, Tokyo Keiso Co., Tokyo, Japan) was installed at the engine inlet to record the flow rate. An optical counter (FS-N11N, Keyence, Osaka, Japan), which also functioned as a tachometer, recorded the piston location in the cylinder during the engine operation. All data acquired during experiments were transferred to a data acquisition unit (GL900, GRAPHTEC, Yokohama, Japan) for recording and further analysis. The power output was calculated as the product of the measured torque and rotational speeds. Fig. 6 shows a schematic overview of the test bench for the compressed-air engine. 4.1. Measurement uncertainty
Fig. 4. Torque diagram (at 8, 10, and 12 bar) for two-stage expansion and singlecylinder engines.
The uncertainty of the measurement system is analyzed by calculating the uncertainty from the instruments and sensors integrated in the system, and the errors can be evaluated by balancing the uncertainties of each of them [19]. When estimating the measurement uncertainty, there are factors that must be considered, including: (1) the operator proficiency, (2) instrument precision, and (3) environment [20]. First, the average value of the experimental data is calculated by summing them and dividing by the number of results ‘n’ [21].
X¼
X1 þ X2 þ … þ Xn n
(12)
where X is the arithmetic mean of the n results considered. Second, this average value is subtracted from each data to find the deviation from each measured result.
d1 ¼ X1 X; d2 ¼ X2 X; …dn ¼ Xn X
Fig. 5. Efficiency diagram (at 8, 10, and 12 bar) for two-stage expansion and singlecylinder engines.
single-cylinder engine is better at high speeds. 4. Experimental setup In this study, series-connected 50 and 100 cm3 four-stroke IC engines supplied by a Taiwanese motorcycle company, KYMCO, were used. By modifying the design of the engine, the four-stroke IC engine was changed to a two-stroke air engine. The modified air engine used high-pressure air as the driving source, and the output performance and fluid characteristics of the air engine were measured.
(13)
where dn is the deviation. Finally, the estimated standard deviation for n measurements is calculated as follows:
sffiffiffiffiffiffiffiffiffiffiffiffi P 2 dn s¼ n1
(14)
where s is the standard deviation. The measurement results in the experiment are affected by factors such as the operator proficiency, instrument precision, and other changes in the environment. As a result, the measured values deviate from the actual values. The operator proficiency is caused by random and unpredictable factors that cannot be completely eliminated, and errors can be reduced only by increasing the number of measurements. The environmental conditions are controlled in the same way as the ambient conditions in the
144
C.-M. Liu et al. / Energy 115 (2016) 140e148
Fig. 6. Experimental setup for the compressed-air engine test bench.
laboratory. The instrument precision is estimated by the instrument specifications from the manufacturer's instructions from a probability distribution correction, which are the accuracy of the torque transducer, flow meter, pressure transducer, and optical counter (as a tachometer for rotational speed). The measurement uncertainty can be obtained by calculating the aforementioned factors. As shown in Fig. 7, the experimental results obtained under 12 bar and continuous rotation at 1300 rpm from 100 cycles are recorded to determine the accuracy of the measurements as the average and deviation from the torque and efficiency per lap in the experiment.
The calculated results indicate that the average values and measurement uncertainties of the torque and efficiency are10.368 ± 0.314 Nm and 70.689 ± 2.702%, respectively. During the experiment, the instrument precision, error of engine mounting, and vibration during engine operation are factors influencing the measurement values. 5. Results and discussion The experimental investigations of a two-stage expansion air engine have been performed using a 50 and 100 cm3 four-stroke IC engine for inlet pressures varying from 8 to 12 bar. The experimental results of a single-cylinder air engine have also been measured and plotted side-by side for comparison. 5.1. Experimental comparison of two-stage expansion and singlecylinder engines
Fig. 7. Torque and efficiency of each cycle.
The P-V area comparisons of the two-stage expansion and single-cylinder engines are shown in Figs. 8 and 9. As shown in Fig. 8, the P-V area of the two-stage expansion engine is larger than that of the single-cylinder engine, according to the experimental results under 12 bar intake pressure and 1000-rpm speed. However, Fig. 9 indicates that the obvious rough intake results in reduced output power as the two-stage expansion engine operates at 12 bar intake pressure and 2500 rpm speed, which may be attributed to the excessively high back-pressure of the first cylinder. In addition, the differences of the pressure drop between the experimental data (in Figs. 8 and 9) and the simulation data (in
C.-M. Liu et al. / Energy 115 (2016) 140e148
Fig. 8. PeV diagram (at 12 bar and 1000 rpm) for two-stage expansion and singlecylinder engines.
145
Fig. 11. Torque diagram (at 8, 10, and 12 bar) for two-stage expansion and singlecylinder engines.
generated when the back-pressure of the first cylinder becomes excessively high at high speeds. However, the exhaust of the singlecylinder engine is discharged directly into the atmosphere; thus, the performance of the single-cylinder engine is more favorable at high speeds. As shown in Fig. 12, the efficiency of the two-stage expansion engine is higher than that of the single-cylinder engine at 2000 rpm or less, and the efficiency of the two-stage expansion engine is lower than that of the single-cylinder engine when the speed is higher than 2000 rpm because of the increased backpressure of the exhaust of the first cylinder and the decreased intake pressure in the second cylinder. 5.2. Effects of timing
Fig. 2) are attributed to the volume of the connecting manifold between the first and the second cylinders, which was disregarded in the simulation for ease of calculation. The volume of the connecting manifold is large compared to that of the cylinders in the two-stage expansion air engine; therefore, the pressure drop during the expansion process is small. As shown in Figs. 10 and 11, the performance of the two-stage expansion engine is less favorable than that of the single-cylinder engine after ~2500 rpm. Because the gas from the first cylinder must flow into the second cylinder, lower intake pressure is
As observed from the simulation and experiment of the twostage expansion engine, the pressure of either the first or second cylinder tends to increase relatively slowly, with inadequate intake during the intake process, resulting in rough exhaust and high back-pressure during the exhaust process. Tai et al. used four fullyflexible cam-less valves (two intakes and two exhausts) for each cylinder, where one intake valve is switchable and connects either the intake manifold or the air tank to the cylinder through a threeway valve. Their air-hybrid engine improved fuel efficiency by 64% for city driving and 12% for highway driving [22]. Trajkovic et al. examined the intake and exhaust systems using the conventional crankshaft-driven cam system [23]. To ensure smooth running and fast response, the airflow was controlled using a simple cam mechanism in the compressed air system. For continuous running
Fig. 10. Power output diagram (at 8, 10, and 12 bar) for two-stage expansion and single-cylinder engines.
Fig. 12. Efficiency diagram (at 8, 10, and 12 bar) for two-stage expansion and singlecylinder engines.
Fig. 9. PeV diagram (at 12 bar and 2500 rpm) for two-stage expansion and singlecylinder engines.
146
C.-M. Liu et al. / Energy 115 (2016) 140e148
operation and fast response, the air flow is simply controlled by a proper cam design mechanism in many corresponding compressed-air engine designs [24]. Therefore, to improve the overall efficiency, timing adjustments are necessary to ensure that the pressure can rapidly increase during the intake process and that the back-pressure in the cylinder can be quickly eliminated during the exhaust process.
5.2.1. Early intake in the first cylinder As determined from the simulation diagrams in Figs. 13 and 14, the greater the degree of intake, the higher is the efficiency at higher speeds. If the degree of intake is increased earlier, the gas in the cylinder would be compressed by the upward piston during intake. Therefore, the pressure would increase rapidly, and the overall efficiency would increase. An excessively large angle of advance would cause the overall power to decrease because the intake process takes place while the piston starts moving up, resulting in an increase in the negative power. Owing to insufficient intake time during the stroke, the piston begins to move down without maximum pressure inside the cylinder. However, if the intake begins earlier at the point where the piston approximates the TDC, then the pressure can reach its maximum earlier during the down stroke of the piston, thus increasing the power. This improvement becomes more obvious with an increase in the rotation speed. As shown in Figs. 15 and 16, early intake in the first cylinder can increase the total power output to 1.79 kW (without early intake: 1.7 kW; 5.3% increase).
Fig. 15. Power and torque output for an early intake of 0 and 20 with 12 bar air supply pressure.
Fig. 16. Output efficiency for an early intake of 0 and 20 with 12 bar air supply pressure.
Fig. 13. Power output diagram (at 12 bar) for different degrees in the first cylinder early intake.
5.2.2. Early intake and exhaust in the second cylinder As shown in Fig. 9, the pressure in the second cylinder increases slowly during the intake process, and the back-pressure increases during the exhaust process at high speeds. Therefore, advancing the intake timing in the second cylinder can improve the intake process; on the other hand, advancing the exhaust timing can reduce the backpressure. As shown in the simulation diagrams in Figs. 17 and 18, early intake and exhaust in the second cylinder can increase its power output and efficiency during the exhaust process from the first to
Fig. 14. Efficiency diagram (at 12 bar) for different degrees in the first cylinder early intake.
Fig. 17. Power output diagram (at 12 bar) for various degrees of early intake and exhaust in the second cylinder.
C.-M. Liu et al. / Energy 115 (2016) 140e148
147
in the second cylinder may result in a long compression stroke, and the overall performance will be poor. Early intake and exhaust in the second cylinder can increase the total power output to 1.82 kW (without early intake and exhaust: 1.7 kW; 7% increase). 6. Conclusions
Fig. 18. Efficiency diagram (at 12 bar) for various degrees of early intake and exhaust in the second cylinder.
Fig. 19. Power and torque output for an early intake and exhaust of 0 and 20 with 12 bar air supply pressure.
the second cylinders. However, if the intake is too early, the negative output caused by the second cylinder increases. In this case, the piston compresses during intake, reducing the rate of increase of the overall power output. As shown in Figs. 19 and 20, early intake and exhaust in the second cylinder is intended to increase the pressure during the intake process and reduce the back-pressure during the exhaust process, both caused by rough exhaust. Early intake and exhaust are more favorable when the speed is increased; because early exhaust
Fig. 20. Efficiency for an early intake and exhaust of 0 and 20 with 12 bar air supply pressure.
In a comparison of two-stage expansion and single-cylinder engines, the simulations and experiments showed that the twostage expansion engine is more favorable than the single-cylinder engine at low speeds. However, at high speeds, because the gas in the first cylinder cannot smoothly exhaust to the second cylinder, the power output of the two-stage expansion engine becomes similar to that of the single-cylinder engine. Through experiments, it became obvious that at high speeds, because the gas in the first cylinder cannot smoothly exhaust to the second cylinder, the power output and efficiency of the two-stage expansion engine are less favorable than those in the single-cylinder engine. Therefore, the two-stage expansion engine is more favorable for applications with low speed and high torque. The two-stage expansion engine can generate up to 1.7 kW of power and 12.42 Nm of torque at air pressure of 12 bar, which is more favorable than the performance of a single-cylinder engine. Through different intake and exhaust timing sequences, early intake in the first cylinder is found to improve the power output by 5.3% as the speed increases, whereas early intake and exhaust in the second cylinder can increase the power output by 7%. Timing adjustments are needed to address the challenge of rough intake and exhaust, which are obvious at high speeds. The results of both the simulation and the experiment indicate that the higher the speed, the greater is the improvement. Therefore, the camshaft valve can be changed to a solenoid valve, and the valve timing can be adjusted in concert with the speed to achieve the greatest efficiency. Furthermore, the two-stage expansion engine can be employed as an auxiliary system in conventional gasoline engines during idle-stop/start and regenerative braking. Acknowledgments The authors would like to thank Taiwanese motorcycle manufacturer KYMCO for providing a 50 cm3 and a 100 cm3 IC engines used in this study. References [1] Drury E, Denholm P, Sioshansi R. The value of compressed air energy storage in energy and reserve markets. Energy 2011:4959e73. [2] Foley A, Lobera ID. Impacts of compressed air energy storage plant on an electricity market with a large renewable energy portfolio. Energy 2013: 85e94. [3] Agarwalla DK, Sethi S. Estimation of run time parameters of compressed air engine prototype. Int J Enhanc Res Sci Technol Eng 2014:108e12. [4] Huang C-Y, Hu C-K, Yu C-J, Sung C-K. Experimental investigation on the performance of a compressed-air driven piston engine. Energies 2013: 1731e45. [5] Papson A, Creutzig F, Schipper L. Compressed air vehicles: drive-cycle analysis of vehicle performance, environmental impacts, and economic costs. J Transp Res Board 2010:67e74. [6] Higelin P, Charlet A, Chamaillard Y. Thermodynamic simulation of a hybrid pneumatic-combustion engine concept. Int J Appl Thermodyn 2002:1e11. [7] Schechter MM. New cycles for automobile engines. 1999-01-0623. SAE Technical Paper; 1999. [8] Fazeli A, Khajepour A, Devaud C, Azad NL. A new air hybrid engine using throttle control. 2009-01-1319. SAE Technical Paper; 2009. [9] Li D, Xu H, Wang L, Fan Z, Dou W, Yu X. Simulation and analysis of a hybrid pneumatic engine based on in-cylinder waste heat recovery. 2014-01-2355. SAE Technical Paper; 2014. [10] Yu Q, Cai M, Shi Y, Yuan C. Dimensionless study on efficiency and speed characteristics of a compressed air engine. J Energy Resour Technol 2015:1e9. [11] Kumar V, Takkar J, Chitransh M, Kumar N, Banka U, Gupta U. Development of an advanced compressed air engine kit for small engine. 2014-01-1666. SAE
148
C.-M. Liu et al. / Energy 115 (2016) 140e148
Technical Paper; 2014. [12] Liu H, Chen Y, Tao Gl. Modeling and simulation of two-stage expansion airpowered engine. Journal of Zhejiang University; 2005. [13] Phillips F, Gilbert I, Pirault J-P, Megel M. Scuderi split cycle research engine: overview, architecture and operation. SAE Int J Engines; 2011. p. 450e66. 2011-01-0403. [14] Anderson J. Modern compressible flow: with historical perspective. third ed. 2002. [15] Borgnakke C, Sonntag RE. Fundamentals of thermodynamics. seventh ed. 2008. [16] Liu C-M, You J-J, Sung C-K, Huang C-Y. Modified intake and exhaust system for piston-type compressed air engines. Energy 2015:516e24. [17] Brown TL, Atluri VP, Schmiedeler JP. A low-cost hybrid drivetrain concept based on compressed air energy storage. Appl Energy 2014:477e89. [18] Huang KD, Tzeng S-C. Development of a hybrid pneumatic-power vehicle. Appl Energy 2005:47e59.
L. Experimental analysis regarding the measurement [19] Tes¸ileanu DM, Niculit¸a uncertainty of welded joints with ultrasound testing. Romanian Rev Precis Mech Opt Mecatronics 2013:23e31. [20] Abdelkarim MF, Nasrat LS, Elkhodary SM, Soliman AM, Hassan AM, Mansour sh. Volume resistivity and mechanical behavior of epoxy nanocomposite materials. Eng Technol Appl Sci Res 2015:775e80. [21] Birch K. Estimating uncertainties in testing. Good Pract Guide No 36. 2003. [22] Tai C, Tsao T-C, Levin MB, Barta G, Schechter MM. Using camless valvetrain for air hybrid optimization. 2003-01-0038. SAE Technical Paper; 2003. [23] Trajkovic S, Tunestål P, Johansson B. Investigation of different valve geometries and vavle timing strategies and their effect on regenerative efficiency for a pneumatic hybrid with variable valve actuation. SAE Int J Fuels Lubr; 2008. p. 1206e23. 2008-01-1715. [24] Yu Q, Shi Y, Ca M. Working characteristics of variable intake valve in compressed air engine. Sci World J 2014:1e9.