Thermo-Kinetic Model to Predict Start of Combustion in Homogeneous Charge Compression Ignition Engine Harshit Gupta and J. M. Malliarjuna Abstract Now-a-days homogeneous charge compression ignition combustion which uses lean homogeneous mixture is becoming very popular in compression ignition engines because of low particulate and smoe emissions. It is hybrid of best features of spar ignition and compression ignition engines. The main challenge with it is controlling start of combustion (SOC). The SOC is dependent on many operating conditions and configuration of the engine. Also, chemical inetics plays an important role on SOC with HCCI combustion. In this study, a single-zone thermo-inetic model has been developed by coupling the thermodynamics and chemical inetics to predict the SOC in HCCI combustion. The model is further used to analyze the effect of various operating conditions on SOC. The predicted results of SOC are also compared with available experimental results. From the results, it is found that there is a good agreement between predicted and experimental results. Therefore, the simulation model developed in this study can be used with confidence to predict the SOC in HCCI combustion. Keywords Homogeneous charge compression ignition, Start of Combustion, Thermo-Kinetic Modeling E I. INTRODUCTION missions from automobiles and fuel economy have always been a matter of concern for the automotive industries and researchers in the area of internal combustion (IC) engines. One of the solutions to decrease emissions and fuel consumption in IC engines is to develop cleaner and more efficient combustion. Now-a-days, homogeneous charge compression ignition (HCCI) is a promising concept for combustion engines to reduce both emission and fuel consumption. It is an alternative technology to conventional spar ignition (SI) and compression ignition (CI) engines. It is characterized by autoignition and combustion of a lean homogeneous mixture. In SI and CI engines, spar and injection timings respectively could be used to control the timing of the combustion. However, ignition timing of HCCI engine is dependent on chemical inetics during the compression stroe; thus causes difficulty in controlling it. In HCCI engine, ignition timing affects operating range, combustion stability, engine performance and emission characteristics. Thus, the challenge to HCCI is related to Harshit Gupta and J M Malliarjuna are with the Indian Institute of Technology Madras, Chennai 600036, India (phone: 91-44-22574698; fax: 91-44-22574652; e-mail: jmmalli@iitm.ac.in). ignition timing. However, main advantage of HCCI combustion is simultaneous reduction of nitric oxides (NOx) and particulates. It is mainly due to instantaneous combustion of lean mixture. In this study, a thermo-inetic model has been developed by coupling thermodynamics, heat transfer, and chemical inetics of combustion chamber fluid. A set of differential equations obtained from the above models is solved in MATLAB simultaneously. At every time step, thermodynamics properties of mixture, concentration of species and rate of change of concentration of species are obtained. From incylinder pressure trace, combustion characteristics are obtained, using which SOC in HCCI combustion is predicted. II. THERMODYNAMICS MODEL Using the principle of conservation of energy and various thermodynamics relations for in-cylinder mixture over control volume (Fig.1), a differential equation is derived using which the thermodynamics state of mixture could be obtained as follows [1]. Fig.1. Control volume of combustion chamber of an engine dv dq dni dn P + hi RT dt + dt dt i= 1 dt i= 1 dt = dt NC N R i= 1 i p i mix Where, P is in-cylinder pressure; V is cylinder volume at any time t; Q is heat transfer between in-cylinder gas and surrounding walls; N i is molar concentration of i th species taing part in combustion reaction; h i and C pi are enthalpy and specific heat respectively of i th species at that temperature; i (1) 272
R is the universal gas constant. From the above equation, it can be said that thermodynamic properties of mixture also depends on rate of change of concentration of species taing part in the combustion. To find out temperature relation, thermodynamics model is coupled with chemical inetic model that determines the concentration of species. III. CHEMICAL KINETIC MODEL The objective of chemical inetic model is to find the concentration of species and rate of change of concentration of species. In this study, n-heptane is chosen as a surrogate fuel for diesel [2]-[3] as the Cetane number and ignition properties of it are closer to diesel. Also, reaction mechanisms for combustion of n-heptane are easily available. The Arrhenius constants for both forward and reverse reactions and thermodynamics properties of each species are from LLNL combustion chemistry group [4]-[5]. Mechanism of n- heptane oxidation consists of 4383 reactions and 544 species. Rate of change of concentration of species in each reaction is obtained from the law of mass action and rate constants are calculated using the Arrhenius relationship depending upon type of reaction [6]. radiation heat transfer and fluid motion during combustion are not present in the HCCI cycle, C 2 is assumed to be zero in the thermo-inetic model. V. START OF COMBUSTION (SOC) The HCCI combustion of diesel fuel exhibits two stages of heat release as shown in Fig.2. First heat release is called as cool flame region occurring due to low temperature reactions (LTR) of intermediate species. Second heat release is due to high temperature reaction of different species. In the HCCI combustion, beginning of main combustion is termed as start of combustion when the ignition occurs. Definition of ignition is the point at which exothermicity of chemical reactions is greater than energy lost to the environment. This definition could be used to determine the SOC from the results. IV. HEAT TRANSFER MODEL Heat transfer process during compression stroe in a HCCI engine is similar to SI or CI engines. Majority of heat transfer is through convective mode. In this study, Woschni s heat transfer correlation is used [1]. Rate of heat transfer is given by, Qt () = ht () AT ( T ) (2) g w Where, A is in-cylinder surface area, T g and T w are gas temperature and average wall temperature; h is convective heat transfer coefficient that is given by (3). -0.2 0.8 0.8-0.53 h = 127.93*B *P *w *T (3) VT = CS + C P P (4) d r w 1 p 2 pv r r ( ) mot Where, S p is mean piston speed, V d is displaced volume of the engine, T r, P r, and V r are reference mixture temperature, pressure and volume respectively and (P-P mot ) is the increase in pressure due to combustion. Second term represents contribution to fluid motion due to change in fluid density during combustion. Coefficients C 1 and C 2 are defined as, C1 = 6.18, C2 = 0 For gas exchange process C1 = 2.28, C2 = 0 For compression stroe (5) C1 = 2.28, C2 = 3.24*10-3 For combustion and expansion As only a closed system is considered in this study, the heat transfer during the gas exchange process is not considered. Original empirical measurements were taen from a diesel engine [7], which had much higher radiative heat transfer components and was accounted for C 2 in (4). As both the high Fig.2. Variation of heat release rate with cran angle From heat release rate curve, the SOC is defined as the cran angle at which the heat release curve crosses zero value of HRR after the cool flame region. The thermo-inetic model gives in-cylinder pressure from inlet valve closure (IVC) to exhaust valve opening (EVO) which can be used to calculate the heat release rate. In this wor, HRR is calculated by the method suggested in [8]. VI. RESULTS AND DISCUSSION The engine specifications considered for the analysis are given Appendix. Model Validation In this study, simulation has been carried for closed part of engine cycle (IVC to EVO) using thermo-inetic model developed. Predicted results obtained so are compared with available experimental data from [9]. Figs. 3 and 4 show variation of in-cylinder pressure and HRR at 2.1 bar BMEP (brae mean effective pressure) of the engine under HCCI combustion. Similar results were observed at other BMEPs also (not shown here). In Figs. 3 and 4, results of both simulation and experiments are shown. 273
reasonably well under different conditions of the engine. TABLE I COMPARISON OF SIMULATED AND EXPERIMENTAL SOC AT VARIOUS LOADS BMEP (bar) Experiment SOC CA atdc Simulated SOC CA atdc 2.1-12 -15 2.6-11 -13 3.15-14 -16 Fig.3. Variation of in-cylinder pressure at BMEP of 2.1 bar Fig.4. Variation of heat release rate at BMEP of 2.1 bar From Fig.3, it can be observed that the predicted values of in-cylinder pressures are in good agreement with those of experiments. Also, in Fig.3, it is observed that, predicted pea pressures are margially higher than those of experimental values. It may be because the single-zone model which assumes uniform and simultaneous combustion of fuel [1]. Similar observations were seen at other BMEPs also. However, in actual engines, combustion is not uniform. Also, the model assumes combustion of all fuel injected into cylinder, whereas in actual engines, not all the fuel injected undergoes combustion. Fig.4 compares HRR obtained from simulation and experiments. Both the simulation and the experimental HRR plots show similar trend of variation with respect to cran angles. Both curve shows presence of cool flame region, negative temperature coeffcient and main combustion. Occurrence of cool flame is due to low temperature reactions of intermediate species, wheras main combustion is due to high temperature reactions of mixture species. Similar trends were observed at 2.5 and 3.1 BMEPs also. In this study, SOC is defined as the cran angle at which the HRR curve after the cool flame region crosses zero heat release rate line. Table I shows the SOC under different BMEP conditions. From the above comparison, it can be concluded that the single-zone thermo-inetic model predicts the SOC VII. PARAMETRIC STUDY To observe effect of various engine parameters on SOC in HCCI combustion, a parametric study has been conducted using thermo-inetic model developed. Intae temperature and pressure, percentage of exhaust gas recirculation (EGR), and equivalence ratio are considered for the analysis. For every operating condition, simulation is carried out using the thermo-inetic model developed. Results are as follows. Effect of Intae Temperature As chemical inetic processes are highly temperature dependant, intae temperature has a significant impact on SOC in HCCI combustion. Figs. 5 to 7 show variation incylinder pressure, HRR and SOC with different intae temperature respectively. With increase in intae temperature, species reaction rate increases as they are temperature dependent, which in turn advances the SOC as shown in Fig.7. Fig.5. Variation of in-cylinder pressure with intae temperatures Fig.6. Variation of HRR with CAD at various intae temperatures 274
Effect of Equivalence Ratio Equivalence ratio affects SOC, combustion duration and power output. Decrease in fuel concentration leads to retarded ignition timing as there is a lesser probability that fuel molecules will collide and react with oxygen molecules. Leaner conditions also increase combustion duration. Figs. 11 to 13 show variation of in-cylinder pressures, HRR and SOC with different equivalence ratios. With increase in equivalence ratio, as expected, SOC advances as shown in Fig.13. Fig.7. Variation of start of combustion with intae temperatures Effect of Intae Pressure Intae pressure changes overall concentration of mixture, thereby increasing rate of chemical inetics. Figs. 8 to 10 show variation in-cylinder pressure, HRR and SOC for different intae pressures. With increase in intae pressure, as expected, the SOC advances as shown in Fig.10. Fig.11. Variation of in-cylinder pressure at equivalence ratio Fig.8. Variation of in-cylinder pressures at various intae pressures Fig.12. Variation of HRR with CAD at equivalence ratio Fig.9. Variation of HRR at various intae pressures Fig.10. Variation of SOC with intae pressure Fig.13. Variation of SOC with equivalence ratio 275
Effect of Exhaust Gas Recirculation EGR affects ignition timing predominantly through charge heating and dilution effects. The charge heating effect, which is more substantial of the two, changes intae temperature of mixture when sufficiently hot exhaust gases are used. An increase in percentage EGR causes dilution of oxygen in airfuel mixture thereby decreasing reaction rate and hence retarding the SOC. Figs. 14 to 16 show variation of incylinder pressure, HRR and SOC with different percentage of EGRs. With increase in percentage of EGR, as expected, SOC retards, as shown in Fig.16. HCCI combustion engine. The results obtained from thermo-inetic model have been compared with the available experimental results. It is found that there is a reasonably good agreement between the predicted and experimental results. The SOC advances with increase in air intae temperature and pressure, and equivalence ratio, whereas it retards with increase in percentage of EGR. APPENDIX ENGINE SPECIFICATIONS AND OPERATING CONDITIONS Bore x Stroe (mm) 80 x 110 Connecting rod length (mm) 231 Compression ratio 16 Speed (rpm) 1500 Fuel flow rate (g/h) 0.805, 0.971, 1.09 Load (g) 2, 2.5, 3 BMEP (bar) 2.1, 2.6, 3.15 Equivalence ratio 0.44, 0.54, 0.614 Volumetric efficiency (%) 88 Fig.14. Variation of in-cylinder pressure at various EGRs Fig.15. Variation of HRR at various EGR REFERENCES [1] J. B. Heywood, Internal Combustion Engine Fundamentals. McGraw Hill, pp 510-514. 1988. [2] Colin O., Pires da Cruz A., Jay S, Detailed chemistry-based auto-ignition model including low temperature phenomena applied to 3D engine calculations, Proc. Combust. Inst. 30 (2) 2649-2656, 2005. [3] Curran H., Gaffuri P., Pitz W. J. and Westbroo C. K. (1998), A comprehensive modeling study of n-heptane oxidation, Combustion and Flame, 114, 149-177. [4] Livermore National Lab, www-cms.llnl.gov/combustion. [5] wb.olin.edu/thermo/sp2004/software/mechanisms/ [6] V. Ganesan, Computer Simulation of Compression-Ignition Engine Processes, University Press, Appendix, 2000. [7] G. Woschni, A Universally Applicable Equation for the Instantaneous Heat Transfer Coefficient in the Internal Combustion Engines, SAE Paper No. 670931, vol 64, pp 3065-3083, 1967. [8] T. K. Hayes, L. D. Savage, and S. C. Soreson, Cylinder pressure data acquisition and heat release analysis on a personal computer. SAE Paper No, 860029, 1986. [9] S. Swami Nathan, J. M. Malliarjuna and A. Ramesh, The Effect of Mixture Preparation in a Diesel HCCI Engine Using Early In-Cylinder Injection During the Suction Stroe, International Journal of Automotive Technology, Vol 8, No.5, pp 543-553, 2007. [10] K. Swan, M. Shahbahti and C. R. Koch, Predicting Start of Combustion Using a Modified Knoc Integral Method for an HCCI Engine. SAE 2006 Transactions Journal of Engines, pages 611-620, 2007. Fig.16. Variation of SOC with EGR VIII. CONCLUSIONS In this study, a thermo-inetic model has been successfully developed and tested to predict star of combustion in diesel 276