Research Article
An optimization of free-piston engine generator combustion using variable piston motion
Advances in Mechanical Engineering 2017, Vol. 9(9) 1–10 Ó The Author(s) 2017 DOI: 10.1177/1687814017720878 journals.sagepub.com/home/ade
Jing Xu, Chenheng Yuan, Yituan He and Rong Wang
Abstract The free-piston engine generator has been a research focus lately due to its crankless structure and unconstrained piston motion, which brings about some weaknesses for the combustion process. This study aims to explore the optimization possibility of combustion using its free-piston motion. Three potential piston motion trajectories were obtained by the free-piston engine generator dynamic model and then served as the boundary condition to establish the combustion model. Through the simulations and comparisons of combustion processes in different piston motions, a manner to optimize the free-piston engine generator combustion performance, that is, by adjusting the piston motion trajectory, was proposed and verified effectively. The results show that under the situation of the slowest velocity and smallest acceleration around the top dead center, the piston moves with a smallest change in in-cylinder gas volume but greatest heat release around the top dead center, during which the free-piston engine generator gains the greatest peak values of incylinder pressure, temperature, and the rate of heat release and pressure rise in the three, naturally leading to the highest NO emission and soot emission; the late combustion duration and heat released in the late combustion are smaller than the other two cases. But the fastest piston motion around the top dead center shortens the duration of combustion and heat release and brings about greater soot emission than the case of conventional sinusoidal trajectory. Keywords Free-piston diesel engine, piston motion, combustion, heat release, optimization
Date received: 27 November 2016; accepted: 22 June 2017 Academic Editor: Jiin-Yuh Jang
Introduction Increasing investigations aiming at transportation energy usage of higher efficiency and cleanliness have aroused comprehensive public attention up to now, with concerns about the severe energy crisis and the rigorous emission regulations. The free-piston engine generator (FPEG), previously prospered as an emerging alternative of low pollutants emission in extensive discussions, has attracted waves of research interests in auto industry.1–3 Without the restriction of the crankconnecting rod mechanism in conventional internal combustion engines, the piston motion in FPEG is free between its two endpoints, converting the piston’s kinetic energy directly into electricity power and
bringing in variable stroke length and compression ratio.3 Besides, its performance advantages, such as multi-fuel possibilities, fewer motion parts, and combustion optimization flexibility, make the FPEG be a promising power device that may promote the energy
College of Traffic & Transportation, Chongqing Jiaotong University, Chongqing, China Corresponding author: Chenheng Yuan, College of Traffic & Transportation, Chongqing Jiaotong University, No. 66, Xuefu Road, Nan’an District, Chongqing 400074, China. Email:
[email protected]
Creative Commons CC-BY: This article is distributed under the terms of the Creative Commons Attribution 4.0 License (http://www.creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/ open-access-at-sage).
2 conversion efficiency and consequently ease the burden of resources and environment upon human.4–6 However, there exist certain barriers on the way to commercialized production of the FPEG. Due to the elimination of flywheel, the unconventional engine has no choice but to assemble more complex starting mechanism and suffer severe misfire.7,8 Otherwise, for the cycle-based energy balance, the engine mostly employs the two-stroke cylinder structure, and the changeable compression ratio does make great difference in the scavenging characteristic and further engine performance. Mao et al.9 indicated that the high scavenging and trapping efficiency can be possibly achieved by a combination of a high effective stroke length to bore ratio and long valve overlapping distance with a low supercharging pressure. Goldsborough and Blarigan10 found that a combination of stratified uniflow scavenging scheme and low-temperature/pressure charge will best optimize the engine efficiency and emissions characteristics. In addition, some researches have indicated that the free-piston motion leads to a slower combustion and heat release and lower indicated thermal efficiency than the conventional engine because of its more intense late combustion. Yuan and colleagues11,12 showed that due to the faster motion around the top dead center (TDC), the ignition delay in FPEG is shorter than a conventional diesel engine, and the FPEG has a more significant pre-mixed combustion, a longer late combustion and a shorter scavenging duration. The similar conclusion was found for a spark-ignition gasoline FPEG by Beijing Institute of Technology.13 They indicated that the gasoline FPEG has a slight disadvantage of indicated thermal efficiency compared to a corresponding traditional gasoline engine due to its slower heat release speed around the TDC. Researchers from Chongqing Jiaotong University have validated that compared with conventional hydrogen engine, a lower combustion efficiency was also occurred in the FPEG fueled by hydrogen.14 Besides, the operation stability of the system deteriorates due to the different TDC positions in different cycles, and the extreme acceleration of piston around the dead centers is an obstacle for reliable and rapid fuel injection,15 which attracted extensive attention of specialists and scholars. Hu et al.16 and Zhao et al.17,18 studied the formation mechanism and factors of the cyclic vibration in single-piston hydraulic free-piston engine. The above researches indicated that the FPEG combustion cannot show advantage in efficiency. However, its ‘‘free’’ characteristic of the piston motion does generate alterable piston motion trajectories, giving a way of strengthening the engine performance, that is, to optimize the variable piston trajectories.19,20 So, this study developed a realizable controlling way, named as
Advances in Mechanical Engineering ‘‘variable piston motion,’’ to obtain the optimal trajectory that would farthest satisfy the demand of power and economical performance. In detail, the in-cylinder mixture motion and chemical reaction control the combustion, heat release, and emission, which are determined by the piston motion on a deeper level. And through altering the load on the generator, various trajectories can be gained in reality, making the potential piston motion trajectories that come from the dynamic model and simulate the practical ones be feasible and of more convenience. This article analyzed and explored the optimization possibility of FPEG combustion using its variable piston motion trajectories. The diesel FPEG prototype proposed in previous researches3,11 was used for this research. Meanwhile, a coupled combustion model was presented to describe the effect of different motion trajectories on the combustion.
Modeling and method With the help of MATLAB, three potential motion trajectories of FPEG were obtained using the dynamic model established in our previous researches,11,14 giving an insight into the dynamic characteristics, on the basis of which displacement curves of piston motion were then transformed into data files that can be read by the threedimensional (3D) software—AVL-FIRE. Adopting a way of numerical interpolation, the software located the piston position in cylinder and the dynamic mesh model was set up accordingly, to accomplish the simulation of the combustion and heat release in the FPEG. The coupled simulation of the dynamic and combustion models investigates the precise affecting laws that the movement feature of the free piston exerts on combustion process.
Variable piston motion Getting rid of the conventional crankshaft connecting rod system, the piston of FPEG is driven back and forth between the two endpoints, the motion of which is only affected by the forces from the in-cylinder gas and external load, without any lateral force. Under stable operations, the force-bearing status of the assembly can be indicated by formula (1) and Figure 1. In which, psl and psr are the gas forces from the left and right scavenging pumps, respectively, which are the same numerically but in opposite directions; pl and pr are the in-cylinder gas pressure forces on the top surfaces of left and right piston, respectively; Ff and Fe are the friction force and the electric external load force, respectively. m
dx2 = S ðpl pr Þ Ff Fe dt2
ð1Þ
Xu et al.
3
Figure 1. FPEG assembly and forces acting on its piston.
where m is the mass of the pistons and connecting rod, x is the position of the pistons, t is time, and S is the top surface area of the piston. Because of the variable external electric load, random piston motion trajectories are theoretically available, giving a chance to achieve alterable cylinder compression ratio and adjustable instantaneous piston velocity accordingly.21,22 However, it is because of the ‘‘free’’ characteristic of the FPEG that there exists a severe challenge to accurately control the combustion and heat release process. As a result, the in-cylinder abnormal combustion phenomenon, especially the misfire, significantly degraded the engine performances. In addition, when the piston moves around the dead centers with a tremendous acceleration, namely, a sharply falling or soaring velocity, it indeed brings obstacles to the precise control of the fuel injection. Taking in this sense, the FPEG does not show any advantage in fuel economy compared with the conventional engine. According to the above discussions, there is an obvious interplay between the dynamic movement of piston and the in-cylinder combustion and heat release process. So, a novel adjusting method aiming at the promotion of combustion and the optimization of economy and emissions was proposed in this study to manage the dynamic motion. Through altering the external electric load, three different piston motion trajectories representing three moving conditions (A, B, and C), respectively, as shown in Figure 2, were gained according to the dynamic model. Figure 2 features the comparisons of dynamic characteristics under three piston motion situations for the FPEG prototype. The main parameters of the prototype are listed in Table 1. The motion situation A, B, and C declare three kinds of potential motion conditions for the FPEG, in accordance with the aforementioned adjustability of its piston motion. In those three situations, accelerations at the TDC or bottom dead center (BDC) reach values of 6 1.11, 6 2.26, and 6 0.66 km/s2 in turn, meaning that the velocity of B ranks first and C remains in the last around the dead centers, and the velocities all reach zero at the dead centers. The reason is that there exists constantly changing transient resultant force acting on the piston assembly,
with the various in-cylinder gas pressures and temperatures and the different trajectories as the root cause. After passing the TDC, the velocity of the situation B acquires the most rapid growth but the C gets the slowest one, indicating that the B holds the maximum incylinder gas volume in this combustion process but the C maintains the minimum one with equal crank angle in this area, which can be testified by the curves of the piston displacement. As a result, with same engine speed, the piston remains near the TDC for a longer duration in the situation C, where the in-cylinder gas pressure and temperature are higher than other piston positions and more possible for the production of NO and soot. On the contrary, the rapidly enlarging in-cylinder gas volume may deteriorate the indicated thermal efficiency and fuel economy of the FPEG in situation B.
Modeling of combustion In essence, the diesel FPEG is still a two-stroke compression ignition diesel engine, although it has special configuration and operates freely with linear movement. Therefore, most sub-models developed for conventional compression ignition engine are also available to the FPEG. In the assumption that the time scales of chemical reaction appear much smaller than that of turbulence, which govern the time scales of combustion, the computational fluid dynamics (CFD) software provides a suitable combustion model—Eddy Breakup Model—to perform the combustion simulation, and the model considers that the reactant is contained by various eddy masses, whose breaking and blending govern the reaction rate, with the minimum concentration of reactant as the third controlling factor.23–25 Moreover, the Extended Zeldovich model and the Lund Flamelet model, respectively, serve to fulfill the calculations of NO and soot production.26–28 Other sub-models applied in current simulation are listed in Table 2 and introduced in detail by many researches.29–33 However, it is well known that crank angle is generally employed as the parameter to present the performances of engine. And the 3D software applied in this study accepts the boundary conditions input in the form of crank angle more conveniently. But without
4
Advances in Mechanical Engineering
Figure 2. Comparisons of dynamic characteristics in three piston motion conditions: (a) positions of piston with different crank angles, (b) velocities of piston with different crank angles, (c) accelerations of piston with different crank angles, and (d) velocities of piston with different positions of piston.
Table 1. Main geometric parameters of the FPEG. Parameters
Value
Bore (mm) Nominal effective stroke length (mm) Nominal stroke length (mm) Nominal compression ratio Mass of the translator (kg) Bowl volume (m3)
60 57 90 16 5.2 5.65 3 1026
FPEG: free-piston engine generator.
Table 2. Sub-models of CFD model in combustion simulation. Sub-models
Simulation model
Turbulence Evaporation Breakup Turbulent dispersion Wall interaction Combustion model NO Soot
k-zeta-f Dukowicz Wave Gosman and Ioannides Walljet1 Eddy Breakup Extended Zeldovich Lund Flamelet
CFD: computational fluid dynamics.
the crank-connecting rod mechanism, the kinetic calculation of the FPEG is time-based. So, the equivalent crank angle (ECA) was adopted in this article to perform the transformation, which appears to be a time and operation frequency–dependent function,34 as follows ECA = ðt t0 Þ f 360
ð2Þ
where t0 is the starting time and f is the operation frequency. As a crucial factor to the CFD model of the FPEG, the grid distribution of the mesh model established for the combustion chamber is not fixed when the crank angle varies. Under different piston positions, the module contains different number of cell layers, cells, or faces, which need to be renewed and confirmed in time.14 Figure 3 illustrates the dynamic mesh model applied in this article.
Initialization of the boundary condition Before the numerical calculation, there is a necessity to define the geometric parameter and initial boundary condition of the model, which are shown in Table 3.
Xu et al.
5
The controllable piston motion trajectory is utilized to optimize the combustion of FPEG due to the fact that the changed trajectories can give a different boundary condition for the injection and combustion. The comparisons indicate that the initial temperature, pressure, exhaust gas recirculation (EGR), and so on are set uniform, and with same injection timing at displacement of 0.042 m, the ECAs of above piston motion in Figure 4 are different, being 160.2°CA, 165.6°CA, and 154.8°CA, respectively. In addition, the starting, ending, and the duration of simulation in the three differ from one another, in which the duration of them lasts 212°CA, 224°CA, and 204°CA, respectively, for a fundamental reason of the alterable piston motion trajectories.
the pressures of A, B, and C continue to rise before they reach their maximums, in which the pressure of C ranks first and the B last, then they peak at 190.5°CA, 189.5°CA, and 191.5°CA, respectively, with the values of 8.14, 6.66, and 8.97 MPa in turn. And in this duration, the rate of pressure rise of the three shows an upward tendency in general, in which the peak value of C is 0.55 MPa/°CA and considerably larger than the other two. The reason lies in the in-cylinder combustion states. In detailed, with the uniform injection position of 3.0 mm to the TDC, injection fuel mass of 8.0 mg, and injection duration of 0.3 ms, it is certain that the injection is completely before the TDC. And the injection timing of the three are 160.2°CA, 165.6°CA, and 154.8°CA, respectively, as shown in Figure 7. Consequently, the C has compressed gas of highest temperature (in Figure 7) and pressure and the most appropriate environment to form combustible mixture at that time and further obtains the highest peak values of in-cylinder gas pressure, temperature, and the rate of pressure rise, followed by the A and B. Furthermore, in a certain region after the TDC, the piston moves downward rapidly. Although the mixture continues to burn and release heat, the in-cylinder gas temperature and pressure decline quickly, due to the rapid increase in the gas volume, as shown in Figure 8. And the B also has the most rapid decrease in the rate of pressure rise, due to its fastest velocity and greatest acceleration when it leaves the TDC, which may shorten the duration of isometric combustion.
Validation of modeling and method
Comparison of combustion heat release process
The modeling and method were supported by an experimental result of the FPEG prototype. The experimental prototype has been introduced in our previous publication,35 and the main specifications of the prototype are listed in Table 2. Figure 5 shows the experimental pressure of the prototype operating with a motion trajectory. Meanwhile, the combustion of the prototype was modeled and simulated using this motion trajectory and the above models and method. The comparison of measured and simulated pressure was also plotted in the figure. It is clearly found that the simulated and measured in-cylinder pressure are in good agreement with each other, and the greatest differences of the results appears in the peak values, which is also within acceptable range. So, the results proved the correctness of the modeling and method.
Figure 9 compares the rate of heat release of three piston motions with the same injection schedule. It can be found that the C starts to release heat earliest and the B latest, and the C achieves the greatest peak value of rate of heat release, followed by the B and A successively. After reaching the peak value, the gas volume is in the process of continually increasing, when the incylinder gas pressure and temperature quickly fall, leading to a doubtless rapid drop in the rate of heat release. According to Figure 4 and Table 3, with the same injection location, the B gains the shortest ignition delay period, bringing in the least combustible mixture that releases least heat after reaching the tipping point of multi-point ignition. Figure 10 demonstrates the comparison of the accumulated heat release, and it is clearly revealed that the starting of heat release of the B takes place latest in the three but the ending of heat release in three situations almost takes place in the meantime. So, the duration of heat release is shortened by the faster piston motion around the TDC. Table 4 shows the parameters in the combustion and heat release process. Before the piston reaching the TDC, the B has a fastest piston motion around the
Figure 3. Dynamic mesh model of simulation.
Results and discussion Comparisons of pressure and temperature Figure 6 shows the comparisons of in-cylinder gas pressure and the rate of pressure rise. It can be seen that
6
Advances in Mechanical Engineering
Table 3. Initial boundary conditions for combustion. Initial parameters
Trajectory A
Trajectory B
Trajectory C
Boost pressure (bar) Boost temperature (K) Initial EGR Starting of compression (°CA) Ending of expansion (°CA) Injection fuel mass (mg) Injection timing (°CA) Injection position to the TDC (mm) Injection duration (ms) Number of nozzle holes
1.2 300.0 0.0 74 286 8.0 160.2 3.0 0.3 4
1.2 300.0 0.0 68 292 8.0 165.6 3.0 0.3 4
1.2 300.0 0.0 78 282 8.0 154.8 3.0 0.3 4
EGR: exhaust gas recirculation; TDC: top dead center.
Figure 4. Injection positions in three situations.
temperature, there is least combustible mixture formed in the ignition delay under the situation B. And in the early expansion stroke, the piston of the B also moves fastest, so the B gains a most rapid increase in the gas volume. That is why the CA50 (the crank angle at 50% of accumulated heat release), on behalf of the center of heat release, and the CA95 of the B (the crank angle at 95% of accumulated heat release), representing the ending of the heat release, are both latest, and there is most heat released in the late combustion. In addition, though with a latest fuel injection, the B has the shortest heat release duration, and the peak value of rate of heat release is almost same with the other two, the reason for which also lies in the fastest piston motion. That is to say, the fastest piston motion contributes to a strong squish flow in cylinder of the B, which quickens the evaporation and atomization of fuel and the combustion and the heat release further. By comparison, the A has a slower combustion and heat release and the C has the slowest one.
Comparisons of the emissions
Figure 5. Comparison of simulation and experiment.
TDC, which causes a shortest ignition delay in the three, so the heat released before the TDC is only 6.2 J. Combined with the lowest in-cylinder pressure and
Figures 11 and 12 compare the emissions of the FPEG in three piston motions. It is clear that the NO starts to form after the TDC in the three situations, and there is greatest gross of NO in the C but smallest in the B. This is because there is a rapid burning phase duration of highest temperature, pressure, and rate of pressure rise in the C, bringing about a large amount of NO, while the situations in the B appear to be opposite. Figure 12 shows that there exists most and earliest soot in the C, while the amount of soot in the B exceeds that in the A, which is because of the fastest piston motion that gives rise to a larger majority of incomplete combustion in the situation B. That is, the shortest heat release duration in B does mean a fastest combustion, in which there is a greater mass of rich mixture that cannot be completely oxidized than the situation A. Of
Xu et al.
7
Figure 6. Comparison of cylinder pressure: (a) cylinder pressure and (b) rate of pressure rise.
Figure 7. Comparison of in-cylinder temperature.
Figure 9. The rate of in-cylinder heat release.
Figure 10. Accumulated heat release. Figure 8. Comparison of in-cylinder gas volume.
Conclusion course, the vast majority of these incomplete oxidation products is oxidized in the later combustion process.
This study presented a simulation on the effect of variable motion on the combustion of a diesel FPEG and
8
Advances in Mechanical Engineering
Table 4. Heat release parameters of three piston motions. Heat release performances
Trajectory A
Trajectory B
Trajectory C
Starting of heat release (°CA) Ending of heat release (°CA) Duration of heat release (°CA) Ignition delay (°CA) Peak value of rate of heat release (J/°CA) Released heat before TDC (J) CA50 (°CA) CA95 (°CA) Duration of late combustion (°CA) Released heat in late combustion (J)
175 216 41 14.8 28.7 32.2 186 197.5 7 55.5
178 217 39 13.6 29.1 6.2 187.5 198.5 9 97.9
172 218 46 18.8 29.3 50.5 185 196 5 33.5
TDC: top dead center.
the engine performances, of which a optimization method that is to adjust the motion trajectories of piston was prospered. Furthermore, a coupling model consisting of a dynamic one-dimensional (1D) model and a thermal 3D model was applied to conduct the numerical calculation of the FPEG. The main analysis results indicate the following: 1.
2. Figure 11. Comparison of NO emission.
3.
Figure 12. Comparison of soot emission.
explored the optimization possibility of combustion using the variable motion. According to the above, the variation in the piston motion trajectory certainly affects the combustion and heat release process, further
4.
The combustion, heat release, and emissions of FPEG are exactly influenced by the piston motion through affecting the in-cylinder gas flow and the environment of chemical reaction, which can be optimized by adjusting the piston motion trajectory. The quantity of combustible mixture is mainly depend on the length of the ignition delay duration and the in-cylinder environment in the FPEG, because of the rapid variation of piston velocity around the TDC. With a longest ignition delay, the C produces most mixture in this period, owing to its highest temperature and pressure in the three. But there is least mixture generated in the ignition delay under the situation B, whose piston moves fastest around the TDC in the three. The fastest formation of mixture with high temperature and pressure of the C does promote the shortening of the late combustion to a certain duration, making for the accomplishment of the major combustion and heat release near the TDC. Compared with the C, situations in the ignition delay of the B have a negative effect on the late combustion process. The B performs the highest velocity and acceleration in the three around the TDC, causing the center of heat release farthest away from the TDC and a smallest peak value of the rate of the heat release. Comparatively speaking, there exists a quickest change in the gas volume at the pre-mixing combustion phase in the B, meaning
Xu et al.
5.
6.
that the isometric process is of least percent of the whole combustion compared to the other two. Because of the fastest velocity and acceleration leaving the TDC, there is much mixture remained to the late combustion, the duration of which reaches a length of 9°CA, releasing 28.8% of the total heat but leading to the shortest combustion duration in B. And the late combustion duration in the C is only of 5°CA, to the benefit of fuel economy in the three. The highest temperature and pressure in the pre-mixing combustion phase of the C give rise to the greatest mass of NO and soot, while the fastest piston motion of the B contributes to an obvious increase in the soot emission but has little effect on its NO emission.
Declaration of conflicting interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Scientific Research and Innovation Fund for Postgraduate Student of Chongqing Municipality (Grant number: CYS16177) and the Basic Natural Science and Frontier Technology Research Program of the Chongqing Municipal Science and Technology Commission (Grant number: cstc2016jcyjA0221).
References 1. Xu Z and Chang S. Prototype testing and analysis of a novel internal combustion linear generator integrated power system. Appl Energ 2010; 87: 1342–1348. 2. Mikalsen R and Roskilly AP. A review of free-piston engine history and applications. Appl Therm Eng 2007; 27: 2339–2352. 3. Yuan C, Xu J, Feng H, et al. Friction characteristics of piston rings in a free-piston engine generator. Int J Engine Res. Epub ahead of print 12 December 2016. DOI: 10.1177/1468087416683076. 4. Hung N and Lim O. A study of a two-stroke free piston linear engine using numerical analysis. J Mech Sci Technol 2014; 28: 1545–1557. 5. Zare S and Tavakolpour-Saleh AR. Frequency-based design of a free piston Stirling engine using genetic algorithm. Energy 2016; 109: 466–480. 6. Preetham BS and Weiss L. Investigations of a new free piston expander engine cycle. Energy 2016; 106: 535–545. 7. Li Y, Zuo Z, Feng H, et al. Parameters matching requirements for diesel free piston linear alternator start-up. Adv Mech Eng 2015; 7. DOI: 10.1177/1687814015574408.
9 8. Yuan C, Feng H, He Y, et al. Motion characteristics and mechanisms of a resonance starting process in a freepiston diesel engine generator. Proc IMechE, Part A: J Power and Energy 2016; 230: 206–218. 9. Mao J, Zuo Z and Li W. Multi-dimensional scavenging analysis of a free-piston linear alternator based on numerical simulation. Appl Energ 2011; 88: 1140–1152. 10. Goldsborough S and Blarigan P. Optimizing the scavenging system for a two-stroke cycle, free piston engine for high efficiency and low emissions: a computational approach. SAE paper 2003-01-001, 2003. 11. Yuan C, Feng H, He Y, et al. Combustion characteristics analysis of a free-piston engine generator coupling with dynamic and scavenging. Energy 2016; 102: 637–649. 12. Feng H, Guo C, Yuan C, et al. Research on combustion process of a free piston diesel linear generator. Appl Energ 2016; 161: 395–403. 13. Miao Y, Zuo Z, Feng H, et al. Research on the combustion characteristics of a free-piston gasoline engine linear generator during the stable generating process. Energies 2016; 9: 655. 14. Yuan C, Xu J and He Y. Performance characteristics analysis of a hydrogen fueled free-piston engine generator. Int J Hydrogen Energ 2016; 41: 3259–3271. 15. Wu W, Hu J and Yuan S. Semi-analytical modelling of a hydraulic free-piston engine. Appl Energ 2014; 120: 75–84. 16. Hu J, Wu W, Yuan S, et al. Fuel combustion under asymmetric piston motion: tested results. Energy 2013; 55: 209–215. 17. Zhao Z, Wang S, Zhang S, et al. Thermodynamic and energy saving benefits of hydraulic. Energy 2016; 102: 650–659. 18. Zhao Z, Wu D, Zhang Z, et al. Experimental investigation of the cycle-to-cycle variations in combustion process of a hydraulic free-piston engine. Energy 2014; 78: 257–265. 19. Zhang C and Sun Z. Using variable piston trajectory to reduce engine-out emissions. Appl Energ 2016; 170: 403–414. 20. Li K, Ali S and Sun Z. Active motion control of a hydraulic free-piston engine. IEEE-ASME T Mech 2014; 19: 1148–1159. 21. Kim J, Bae C and Kim G. The operation characteristics of a liquefied petroleum gas (LPG) spark-ignition free piston engine. Fuel 2016; 183: 304–313. 22. Woo Y and Lee Y. Free piston engine generator: technology review and an experimental evaluation with hydrogen fuel. Int J Automot Techn 2014; 15: 229–235. 23. Hung N and Lim O. A review of free-piston linear engines. Appl Energ 2016; 178: 78–97. 24. Hirotatsu W, Yoshikazu S, Yohsuke M, et al. Spray combustion simulation including soot and NO formation. Energ Convers Manage 2007; 48: 2077–2089. 25. Yasuda K, Yamasaki Y, Kaneko S, et al. Diesel combustion model for on-board application. Int J Engine Res 2016; 17: 748–765. 26. Kosmadakis G, Rakopoulos D and Rakopoulos C. Methane/hydrogen fueling a spark-ignition engine for studying NO, CO and HC emissions with a research CFD code. Fuel 2016; 185: 903–915.
10 27. Lira-Teco J, Rivera F, Farias-Moguel O, et al. Comparison of experimental and CFD mass transfer coefficient of three commercial turbulence promoters. Fuel 2016; 167: 337–346. 28. Chintala V and Subramanian KA. CFD analysis on effect of localized in-cylinder temperature on nitric oxide (NO) emission in a compression ignition engine under hydrogendiesel dual-fuel mode. Energy 2016; 116: 470–488. 29. Sadabadi K, Shahbakhti M, Bharath A, et al. Modeling of combustion phasing of a reactivity-controlled compression ignition engine for control applications. Int J Engine Res 2016; 17: 421–435. 30. Li Y, Jia M, Liu Y, et al. Numerical study on the combustion and emission characteristics of a methanol/diesel reactivity controlled compression ignition (RCCI) engine. Appl Energ 2013; 106: 184–197.
Advances in Mechanical Engineering 31. Choi M, Song J and Park S. Modeling of the fuel injection and combustion process in a CNG direct injection engine. Fuel 2016; 179: 168–178. 32. Barari G, Sarathy S and Vasu S. Improved combustion kinetic model and HCCI engine simulations of diisopropyl ketone ignition. Fuel 2016; 164: 141–150. 33. Rezaei R, Dinkelacker F, Tilch B, et al. Phenomenological modeling of combustion and NOx emissions using detailed tabulated chemistry methods in diesel engines. Int J Engine Res 2016; 17: 846–856. 34. Mao JL, Zuo ZX and Feng HH. Parameters coupling designation of diesel free-piston linear alternator. Appl Energ 2011; 88: 4577–4589. 35. Yuan C, Feng H and He Y. An experimental research on the combustion and heat release characteristics of a freepiston diesel engine generator. Fuel 2017; 188: 390–400.
