Simulation of Single Cylinder SI/HCCI Internal Combustion Engine using AVL BOOST

1. Introduction:

Technology has made it possible to design and simulate various engineering models. Based on various formula and conditions the software can predict how a design would perform with great accuracy. One such software is AVL BOOST which is a fully integrated advanced level tool for running simulations on virtual engine. It involves simulation of engine cycle and gas exchange for an entire engine model. It has not only reduced time and cost involved in designing of engine models but also given the user the flexibility to optimize the design by changing parameters or inputs without going into laborious calculations. In short AVL Boost is one of the reliable and efficient tools which give the designer enough confidence to finally have a prototype of the design by drastically reducing the chances of failure [9].

1.1. Aims and Objectives:

 The aim of this report is to build a model of single cylinder SI and HCCI engine in BOOST using the geometries and valve timings provided and to use experimental data to determine combustion profile for the models. The models are calibrated based on in-cylinder pressure trace, IMEP and maximum pressure. The results are then evaluated using experimental data. Engine parameters are changed to make it more efficient and cleaner and an exhaust after treatment is incorporated.

2. Setup for Single Cylinder Engine Model:

2.1. Spark Ignition (SI) Engine:

Combustion is ignited by a spark in SI engines and almost stoichiometric air/fuel ratio is used to allow spark to ignite better and for the flame to propagate better [14].High lift was used in SI mode i.e. cases 4-7 and the fuel used was gasoline which has a stoichiometric air/fuel ratio of 14.7. The actual values of air/fuel ratio were calculated using given values of lambda in each case and the IVO and EVO for each case was input using given data.

2.2. Homogeneous Charge Compression Ignition (HCCI) Engine:

 Auto-ignition takes place in HCCI due to compression, the charge is premixed, and the combustion is lean. The engine is un-throttled which reduces throttling losses and combustion temperatures are low which reduce NOX and the quick combustion makes it comparable with the Otto Cycle [7].Hence the throttle and spark timing was removed for HCCI cases i.e. cases 1-3 and low lift was used. The air fuel ratios were calculated using stoichiometric values and the IVO and EVO was calculated using given data.

2.3. Input Data:

2.3.1. Technical Specification:

Table 1: Basic Geometry

2.3.2. Calculation of Required Input Data:

 Stroke volume was calculated using Eq (1)

$V_d=\raisebox{1ex}{$\pi B^2$}\!\left/ \!\raisebox{-1ex}{$4$}\right. \times L=565630{\mathit{mm}}^3$

      [4] (1)                    

The clearance volume was calculated as:

$V_c=\frac{V_d}{r_C–1}$

= $\frac{565630{\mathit{mm}}^3}{11.5–1}=53869{\mathit{mm}}^3 \left[4\right]$

(2)

$S=a\cos \theta +{\left(L^2–a{\sin \theta }^2\right)}^{\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$2$}\right.}$

, a=44.45mm (3)

$V=V_c+\frac{\pi B^2}{4}\left(L+a–S\right)$

                    [4] (4)

By plotting logarithm of fired cylinder pressure and that of instantaneous volume and by taking the average of the two slopes the polytrophic constant was calculated for all the seven cases. Graphs for HCCI engine case 1and SI engine case 4 are shown in figure 1 and figure 2.Figure 1: Log P vs. log V for case 1 (HCCI)Figure 2: Log P vs. Log V case 4 (SI)

$\frac{\mathit{dQ}}{\mathit{d\theta }}=\frac{\gamma }{\gamma –1}\mathit{ p}\frac{\mathit{dV}}{\mathit{dt}}+\frac{1}{\gamma –1}p\frac{\mathit{dP}}{\mathit{dt}}$

                        [4](5)

γ was used in the above equation to calculate heat release rate. Using the polytrophic constant calculated for all seven cases, the heat release rate was also calculated for each case and the normalized heat release was input into AVL BOOST.

Figure 3: AVL BOOST Setup

The model was setup in BOOST using given data, the engine geometry and intake and exhaust geometry. The cylinder piston area, cylinder head surface area and liner area were calculated using AVL BOOST HELP and the intake and exhaust valve coefficients for both SI and HCCI was calculated using BOOST HELP and literature review.

2.4. Model calibration and validation

2.4.1 SI calibration and validation

In order to calibrate the SI case 4, the throttle angel was changed until the IMEP, Pmax and PCad matched the experimental data. Pressure vs. CAD plot was generated in BOOST which was exported in excel. A plot was also created from the provided experimental data. A comparison was made between the two which showed the pressure profile resembled perfectly and the crank angle at which the pressure peak occurred, the magnitude of the pressure peak and IMEP was same for both.

Figure 4: Pressure vs. Crank angle Curve Case 4 (Experimental Data)

To validate the SI model, the remaining SI cases with their corresponding heat release and air/fuel ratio were fed to the model and only the throttle angel variable was changed. Upon comparison of generated data vs. experimental data it showed that the in-cylinder pressure trace, IMEP, Pmax and crank angle data and IMEP matched, hence validating the model.

SI engines are calibrated based on throttle valve as it controls the amount of air and hence the fuel amount entering the system while maintaining a constant air/fuel ratio, mostly stoichiometric conditions, and thus controls the engine power. By increasing the throttle position more air is allowed to enter which in turn allows more fuel into to the system to maintain stoichiometric conditions and vice versa [22].

2.4.2 HCCI model calibration

To calibrate the HCCI engine the throttle was removed, and low lift was used. The low lift helps trap more residual gases to facilitate combustion and to extend the operating range of the engine [17].The air/fuel ratio and normalized heat release graph for case 1 was put in BOOST and the cylinder wall temperatures were changed until the correct IMEP, Pmax and Pcad were achieved. The wall temperature of ${100}^{0}C$

was used for case 1. Kezhuo Wang (2018) in his CFD simulation of HCCI engine studied influence of cylinder wall temperature on engine performance. In his investigation he varied cylinder wall temperature from about ${30}^{0}C$

to ${180}^{0}C$

and found out that decreasing cylinder wall temperature decreases maximum temperature of the engine cylinder. At temperature lower than ${90}^{0}C$

the engine misfired since heat release rate was quite low giving us the idea of lowering in cylinder temperature to minimum of ${100}^{0}C$

to avoid misfire and to to reduce emissions and increase efficiency for HCCI [20].

The pressure vs CAD was plot in BOOST and was exported to excel to compare the values. The graph below shows the Pmax and Pcad for case 1match the experimental data hence calibrating the model.Figure 5: Pressure vs Crank angle Curve Case 1 (Experimental Data)

To validate the model the data for the remaining HCCI cases was input in BOOST and only the wall temperatures were changed to get the correct values, hence validating the model.

HCCI engines are calibrated based on wall temperatures as HCCI combustion depends on chemical kinetics which is influenced by wall temperatures [21].The wall temperatures effect the charge near the wall and hence effect the combustion duration, ignition rate and heat lost to the walls [5].If the wall temperature is lowered the peak value of heat release rate is significantly decreased. On the other hand, increasing the temperature would delay the rate of pressure rise during combustion [12].The heat transfer happening inside the cylinder is affected by the temperature of its walls which in turn affects the air fuel mixture that enters it thus affecting the combustion process. Decreasing wall temperature results in delay in ignition timing thus extending duration of ignition [20].

 3.Results and Optimization of Engine:

3.1. Results:

Table 2: Experimental data vs AVL Boost Data

The PMax, PMax (CAD) and IMEP from the      BOOST data matches the experimental data for both the SI and HCCI cases with an error of less than 10% thus validating both models.

It was noted that as the throttle angel was reduced in SI engine (part load operation), the pumping losses increased, which could be seen by the PV graphs plotted in excel. This is because as the throttle restricted the amount of air entering the engine, the volumetric efficiency reduced. The intake air pressure dropped below atmospheric pressure thus increasing the cylinder pressure resulting in the piston working against this pressure difference in order to take more air in.

On the other hand, it was noted that the pumping losses in HCCI engine are significantly lower compared to SI engine. This is mainly attributed to the lean combustion used in HCCI engines. Since the ratio of air is greater in lean combustion, this means that the pressure in the intake manifold is high to allow more air into the cylinder thus reducing pumping losses [7].According to the results the greater the air/fuel ratio the higher the volumetric efficiency was.

3.2 Optimization of Engine:

3.2.1 SI Engine Optimization:

Several factors were changed for optimizing the performance of SI engine for case 4. The compression ratio, engine speed, exhaust runner lengths and IVO were varied to get optimum results.

By increasing the compression ratio to 15:1, the BMEP, Brake power, torque and thermal efficiency increased while the Bsfc, NOX and CO2reduced. The improvement is attributed to the better fuel evaporation and better mixing at high compression ratios [15].

The rpm was increased to 2500rpm. The positive valve overlap was also increased by opening IV early. This led to a reduction in NOX and HC emissions, because early intake opening leads to backflow of residual gases into the intake port which is recirculated in the next cycle into the cylinder leading to their combustion. As engine speed was increased the high overlap was beneficial as it increased volumetric efficiency due to ram effect [6].

Changes to the engine geometry were made by reducing the exhaust pipe length to 44mm which led to better efficiency, torque and power. This is due to wave tuning. When the pressure wave in the exhaust manifold is tuned correctly it returns to the cylinder before valve is closed thus producing a negative pressure at the valve opening. This phenomenon is called scavenging which pushes more amount of residual exhaust gas out thus improving efficiency[3][13].

The combined effect of all the parameters can be seen in the table below.

The results show an increase in volumetric efficiency, brake power and BTE. The BSFC decreases as a result of increase in brake power. The increase in brake power also increases the BTE. ${\mathrm{NO}}_x$

and CO2 are reduced.

3.2.2 HCCI Engine Optimization:

Several parameters were changed for optimization of case 1 HCCI. The efficiency in HCCI engine can be increased by optimizing NVO to trap hot residual gases(internal EGR) in the cylinder [23].By closing the exhaust valve early, the hot gases can be used to facilitate the auto-ignition process and reduce combustion timing[24].The hot residual gases heat up the fresh incoming charge thus increasing their temperature and facilitating combustion. Hence EVC and IVO were optimized to get higher volumetric efficiency, higher brake power, lower Bsfc and reduced NOx.

The combustion in an HCCI engine is dependent on the mixture chemistry in the cylinder. By reducing engine speed, the pre-combustion reactions during compression stroke are improved due to more residence time, thus combustion occurs early improving power and efficiency [16]. There is also greater breathing characteristics at reduced rpms which can be attributed to lesser flow friction and enhanced wave dynamics. Hence the rpm was reduced from 1500rpm to 1200rpm which led to an improved volumetric efficiency, thermal efficiency and brake power. The bsfc was also reduced.

Compression ratio has a strong effect on ignition timing and charge temperature in HCCI engines [10]. The compression ratio was also increased to 13.5:1.

The table shows the combined effect of reducing rpm to 1200rpm, increasing the compression ratio to 13.5:1 and optimizing negative overlap.

The results show an increase in volumetric efficiency, brake thermal efficiency and brake power. The Bsfc reduces as a result of increase in brake power. The increase in brake power also increases the BTE. However, there is also an increase in CO levels. This shows that there is a tradeoff between optimum performance and reduced emissions.

4. Model/Software Limitations:

The model has limitations which cannot incorporate certain elements which are present in the operation of engines in real life. Two of which are turbulence and heat loses. Since the Reynolds number is very high inside the internal combustion engines while they are operating, turbulence is developed. Various other complex motions such as swirling flows and tumbling are produced after the introduction of air fuel mixture. As a result of turbulence and complex motions and their interaction with the valve motion, heat transfer inside the engine becomes unsteady and undergoes local changes. Reynolds number increases with increase in engine piston speed and thus turbulence increases which influences the heat transfer inside the engine [12].Effect of turbulence in the real-life 3D engine is quite different and all the heat loses in real life engine cannot be incorporated in the model. Turbulence affects the flame speed by assisting in mixing thus accelerating chemical reactions in SI engine [8].whereas in HCCI engine turbulence affects Rate of Heat Release [19]. Assuming a streamline flow of gasses ignores the changes that happen in reactivity of fuel through actual non-streamline flow. The assumption that the process is isentropic does not account for the loss that would happen due to friction, noise or other heat transfer loses [1]. The software is very much limited to inputs and the design that is made by the user. 

5. Exhaust After treatment:

NOx concentration in the exhaust gas depends on peak value of cyclic temperature and amount of oxygen available inside the combustion chamber. Hence in order to reduce ${\mathrm{NO}}_x$

in the exhaust one can either reduce the peak temperature or reduce available oxygen in the combustion chamber. This can be done by diluting fuel air mixture through addition of substances that are non-combustible before it enters the engine cylinder. Water injection, catalytic converter and Exhaust gas recirculation are among the techniques used for this purpose. Water injection decreases specific fuel consumption, as a result this method cannot be used beyond a certain limit [2]. Catalytic convertor on the other hand reduces ${\mathrm{NO}}_x$

emissions by changing the chemical properties of the exhaust gases. Most of the emissions are eventually converted into carbon dioxide and water vapor [2]. Exhaust gas recirculation is one method that is of more interest since it is effective in reducing harmful gases for both SI and HCCI engines where 10-30 % of engine exhaust gas is recirculated and sent back to the engine inlet manifold. Since the fresh air at inlet is mixed with exhaust gas it reduces oxygen concentration and simultaneously reduces maximum burning temperature thus reducing ${\mathrm{NO}}_x$

[2]. The method is efficient to an extent where ${\mathrm{NO}}_x$

emission can be reduced from 25.4% to 89.6% [11].

There are certain challenges related to usage of EGR and the major one is that it decreases the performance of the engine. Various researchers have come up with different approaches in order to overcome this shortcoming such as EGR hydrogen reforming and treating the stream before it enters the inlet manifold. In case of HCCI engine Lü and coworkers (2005) proposed cooling of EGR in order to prolong combustion time. One most used method to recover the reduced performance is application of turbo charging that avoids self-ignition levels from being reached during the process. By increasing compression ratio through turbo charging ${\mathrm{NO}}_x$

formation increases as peak temperature is increased but at the same time addition of inert gases through EGR reduces it. Thus, in order to reach best results, one is required to optimize value of recycled gas amount and compression ratio [18].

6. Conclusion:

Both models were run successfully and were calibrated and validated with respect to given experimental data by making specific adjustments during initial modelling and simulation. The intent of improving emissions and optimizing the model is evident throughout the project.AVL BOOST proved an efficient tool in effectively simulating the models with good accuracy bearing in mind various limitations that the software has. 

 The software could have been more user friendly if recommendations or suggestions were presented by it especially during model development and calibration. More perfect outcome would have been achieved if phenomena such as turbulence and values of heat loss could have been incorporated in the process making the models closer to real life. Even though there are certain benefits of using 1 D software such as AVL BOOST that are related to simple and fast calculations but developing a working model requires good in-depth knowledge of input parameters. 

7. Project Management:

In order to achieve the given goals, the project was divided into various segments as shown in the Gantt chart in the end. Since it involved modelling, simulation, calibration and validation there were different steps in the process which needed revisiting. One of the most crucial activities was literature review for model development which was efficiently distributed between the team and led to positive discussions till an approach was developed at each stage of the project. Efficient time management and segregation of tasks ensured successful and on time achievement of the stated milestones.

References

[1]    Aceves, S. M., Flowers, D. L., Martinez-Frias, J., Espinosa-Loza, F., Christensen, M., & Johansson, B. (2005). Analysis of the Effect of Geometry Generated Turbulence on HCCI Combustion by Multi-Zone Modeling. Rio de Janeiro: SAE.

[2]    Amritkar, A. B., & Badge, N. (2016). Effect of Exhaust Gas Recirculation (EGR) in Internal Combustion Engine. International Research Journal of Engineering and Technology, 1180-1185.

[3]    Aradhye, O., & Bari, S. (2017). Continuously Varying Exhasut Pipe Length and Diameter to Improve the Performance of a Naturally Aspirated SI Engine. ASME International, 8.

[4]    B.Heywood, J. (2018). Internal Combustion Engine Fundamentals. Ohio: McGraw-Hill Book Company.

[5]    Chang, J., Filipi, Z., Assanis, D., Kuo, T.-W., Najt, P., & Rask, R. (2005). Characterizing the thermal sensitivity of a gasoline homogeneous charge compression ignition engine with measurements of instantaneous wall temperature and heat flux. International Journal of Engine Research, 289-310.

[6]    Choi, K., Lee, H., Hwang, I. G., Myung, C.-L., & Park, S. (2008). Effects of various intake valve timings and spark timings on combustion, cyclic THC and NOX emissions during cold start phase with idle operation in CVVT engine. Journal of Mechanical Science and Technology, 2254-2262.

[7]    Dahl, D. (2012). Gasoline Engine HCCI combustion extending the High Load Limit. Goteborg: Chalmers University of Technology.

[8]    Hynes, J. (1986). Turbulence effects on combustion in spark ignition engines. Leeds: University of Leeds.

[9]    LIST, A. (2018). AVL BOOST™ Combustion and Emissions. Retrieved 11 24, 2018, from www.avl.com: https://www.avl.com/simulation-solutions-for-construction-equipment/-/asset_publisher/gYjUpY19vEA8/content/avl-boost-combustion-and-emissions

[10] Najafabadi, M. I., & Aziz, N. A. (2013). Homogeneous Charge Compression Ignition Combustion: Challenges and Proposed Solutions. Journal of Combustion, 14.

[11] Onawale O, T. (2017). Effect of Exhaust Gas Recirculation on Performance of Petrol Engine. Journal of Engineering and Technology, 14-17.

[12] Park, H. J. (2009). Development of an In-cylinder Heat Transfer Model with Variable Density Effects on Thermal Boundary Layers. Michigan: The university of the Michigan.

[13] Sawant, P., Warstler, M., & Bari, S. (2018). Exhasut Tuning of an Internal Combustion Engine by the Combined Effects of Variable Exhaust pipe Diameter and an Exhasut Valve Timing System. MDPI, 1-16.

[14] Stone, R. (1992). Introduction to Internal Combustion Engines. Middlesex: Macmillan.

[15] T., A., C. O, F., & G. Y. , P. (2012). Influence of compression ratio on the performance characteristics of a spark ignition engine. Advances in Applied Science Research, 1915-1922.

[16] Thring, R. H. (1989). Homogeneous-Charge Compression-Ignition(HCCI) Engines. SAE International, 12.

[17] Uyumaz, A., & ÇINAR, C. (2016). Understanding the Effects of Residual Gas Trapping on Combustion Characteristics, Engine Performance and Operating Range in a HCCI Engine. International Journal of Advances in Science Engineering and Technology, 6-12.

[18] Vianna, J., Reis, A., Oliveira, A., & Fraga, A. (2005). Reduction of Pollutants Emissions on SI Engines – Accomplishments With Efficiency Increase. ABCM , 217-222.

[19] Vressner, A., Hultqvist, A., & Johansson, B. (2007). Study on Combustion Chamber Geometry Effects in an HCCI Engine using High-Speed Cycle-Resolved Chemiluminescence Imaging. SAE International.

[20] Wang, K. (2018). HCCI engine CFD simulations: Influence of intake temperature, cylinder wall temperature and the equivalence ratio on ignition timing. The Ohio State University.

[21] Wilhelmsson, C., Vressner, A., Tunestål, P., Johansson, B., Särner, G., & Aldén, M. (2005). Combustion Chamber Wall Temperature Measurement and Modelling during Transient HCCI Operation. SAE Technical Paper Series, 13.

[22] Xu, C. C., & Cho, M. H. (2017). The study of an Air intake on the Throttle of the Engine by CFD in Spark Ignition Engine. International Journal of Applied Engineering Research, 5263-5266.

[23] Yang, J., Culp, T., & Kenney, T. (2002). Developmant of a Gasoline Engine System Using HCCI Technology- The Concept and the Test Results. SAE International, 16.

[24] Zhao, F., & Asmus, T. W. (2003). Chapter 4 : HCCI Control and Operating Range Extension. In F. Zhao, T. W. Asmus, D. N. Assanis, J. E. Dec, J. A. Eng, & P. M. Najt, Homogeneous Charge Compression Ignition (HCCI) Engines. SAE.