Modelling of SI and HCCI Engines on AVL
BOOST and their Comparisons and Improvements
Abstract: The main aim of this Paper is
to develop different types of Engines on Internal combustion software AVL BOOST
and Compare their results and discussing about the Improvements. The different
types of Engines are Spark Ignition(SI) and Homogeneous charge compression
ignition(HCCI). Here we are developing a single cylinder engine of SI and HCCI
working at 1500 rpm. By using the information from a literature paper we have
to calculate the given data for the BOOST software. After this we have to run
multiple simulations by adjusting different parameters to get optimal values. This
paper proves that the AVL BOOST software is good at modelling the engines,
although there is lack of experimental data, so we have to do few assumptions
while inputting data to the BOOST software. Because of this full validation of
the software’s accuracy could not be proven. The improvements were done in
terms of IMEP , NOx , Pmax
compared with the experimental results.
As there are
different varieties of vehicles are coming into the market, the importance of
internal combustion engines is not decreased and as there are lot of
legislations are coming in automobile field which restricts the emissions. So,
keeping this in mind we are trying to work on reducing the emissions and making
improvements of an internal combustion engine using a software called AVL BOOST
in terms of IMep, Pmax, which leads to the increase in its efficiency with less
In this paper, we are doing simulation of different
engines using AVL BOOST software, as developing a prototype is costly and it is
a time taken process. By doing this simulation we can get almost exact results
as an experimental simulation process. After that we can do improvements to the
model to get better IMep and NOx.
1.1.Aims and Objectives
of this paper is modelling and simulation of different types of engines in AVL
BOOST software namely Spark Ignition(SI) and Homogeneous Charge Compression
objectives of this paper are calibrating a particular engine model on a
particular case to get close values of Imep, Pmax, and NOx emissions. After
calibrating we have to validate it with remaining cases sourcing from the
literature review to check with experimental results of those cases. The same
procedure will be followed for both the engine models and then their results
will be compared.
improvements to be made by adjusting different parameters to make the engine
efficient is discussed.
2. Technical Reviews
2.1. AVL BOOST
AVL BOOSTTM is
an internal combustion software used to model different types of engines and
simulate them with all real simulation parameters like pressure, temperature,
volume, air-fuel ratio and some internal cylinder parameters like in cylinder
temperatures and calibration factors and some emission factors is well at a
comparatively low cost than a real life simulation with almost close results.
2.2. SPARK IGNITION(SI) ENGINE
ignition engine uses gasoline as fuel which is pumped into the cylinder with
the help of an injector at stoichiometric air-fuel ratio of 14.7:1. It works on
four strokes namely intake, compression, combustion and exhaust strokes. Here
the air fuel mixture is burned with the help of a spark plug after the air-fuel
mixture was injected into the cylinder during the intake stroke.
2.3. HCCI ENGINE
Homogeneous Charge Compression Ignition(HCCI) shares the properties of both
Spark Ignition and Compression Ignition engines. Here the air-fuel ratio is
different from stoichiometric air-fuel ratio and is based on lambda(?) which is multiplied by the stoichiometric air-fuel
ratio to give the actual value. This
engine also works on four strokes like the Spark Ignition Engine, but the
air-fuel mixture auto ignites under compression without the aid of a Spark
plug. But the efficiency of this engine is higher than the SI and CI engines.
3. BOOST setup of SI and HCCI Models
Figure 1- Model
of SI/HCCI Engine
Figure 1 shows the layout of both the SI and HCCI models with all the
elements like Engine, Cylinder, Connecting Pipes, Plena, Throttle, Injector and
the Junctions. The inlet and exhausts were modelled as system boundaries in
The only difference between SI and HCCI layouts is the diameter of a pipe
after the throttle and throttle angles. In HCCI the throttle remains fully open
whereas in SI It will be changed according to the simulation. Both the models
were set to run for 30 cycles to allow for the convergence of the results in
the simulation1. The fuel is directly injected into
the cylinder of the HCCI model and same Injector is used for SI model also1.
Calculations and Analysis
This step involves the Calculations of input parameters to the AVL BOOST
software using the values in the Literature Review. At first we have to
calculate the Volume of the cylinder, clearance volume and Swept volume from
the given Bore diameter, Length of the Stroke and Length of the Connecting Rod.
Then we have to calculate the cylinder volume at any crank angle using
‘C.V.Ferguson’s’ method2. After this we have to find
Logarithmic values of Cylinder volume at any crank angle and Logarithmic values
of pressure from the experimental results. Then we have to draw a graphs
between Log (P) vs Log (V) for all SI and HCCI cases.
Figure 2- Log p vs Log v graph SI
Figure 3- Log p vs Log v graph HCCI
Figures 2 and 3 shows the Log p vs Log v graphs of SI and
HCCI models for particular cases. From the HCCI graph we can observe that there
is Recompression involved in it which gives some extra pressure for the next
cycle so that the Compression of the air-fuel ratio happens rapidly and the
fuel ignites quickly. This Re-Compression occur due to neither inlet nor
exhaust valve opens at 00 TDC unlike SI engines which has positive
overlap that allows pressure to escape via the exhaust.3
Rate of Heat Release(ROHR)
In the graphs there are some linear equations mentioned on
particular lines(Compression and Combustion lines) which gives us the
polytropic indices (k) which is useful to calculate Rate of Heat Release. This
is obtained from the equation
K*Log(V) = Constant
After calculating this, we have to calculate the Rate of Heat
Release(ROHR) by using ‘H.R.STONE’s method’4.
Figure 4 Rate of Heat Release of SI case 4
Figure 4 shows the calculated Rate of Heat Release of case 4 of
Spark Ignition model and its comparison with the given experimental heat
release data. Here we can observe that there is not much difference between the
calculated and experimental data.
Figure 5 – Rate of Heat Release of HCCI case 1
5 shows the Rate of Heat Release data of case 1 of HCCI model. Here we can find
some small differences between
the calculated and experimental data. That is because of some assumptions
during the recompression which gives different pressure values. We can find
the same difference in the remaining
cases of HCCI model.
Figure 6- ROHR HCCI vs SI
Figure 6 shows us the comparison of heat release data of SI and HCCI
models. Here we can find a extreme difference in heat release between the two
models. That is because of the Recompression in HCCI which burns the fuel at a
rapid rate with more amount of heat release, whereas the SI model burns the
fuel slowly with less heat release.
Mass Fraction Burnt(MFB)
After calculating Rate of Heat Release, we have to calculate Mass
Fraction Burnt(MFB). To calculate Mass Fraction Burnt we are using ‘H.R.
= – (+)k
The above equations gives us the pressure values we need to calculate
Mass Fraction Burnt. ?Pc is the pressure rise
due to combustion and ?Pc* is the
pressure rise due to combustion referred to datum volume4.
Assume that the combustion occurs after N increments which is defined by
pressure rise due to combustion becoming zero4. Then ?Pc* is directly
proportional to the mass fraction burnt4 then,
Mfb = /
Figure 7- MFB Case 1 HCCI
7 shows the comparison of calculated, Experimental and AVL BOOST Mass Fraction
Burned of HCCI model in case 1. If you can observe, the calculated Mass
Fraction Burnt and AVL BOOST are almost same.
Figure 8- MFB (SI VS HCCI)
Figure 8 shows the comparison between Mass Fraction Burned of
SI and HCCI models. Here, we can observe that the fuel is burnt rapidly in HCCI
than SI model which means the time of combustion is very less.
Figure 9- Rate of Pressure rise(RPR) SI vs HCCI
Figure 9 shows the Rate of Pressure Rise due to combustion against the
crank angle degree.
Calibration and Validation of SI and HCCI Models
The valve lift and valve timings for all cases can be found from the
Figure 10- Valve Lift of SI
The figure 10 shows the Valve timings of the SI model. Here if we can
observe that the intake lift came after the exhaust lift, that is because the
AVL BOOST starts the simulation with the combustion stroke which means it
starts at 3600 and there is a valve overlap of 200, which
means both the valves were opened for that 200 time. As per the
Literature review the valve lift is same for all the SI cases.
Figure 11- Valve Lift HCCI
The above figure shows the Valve Timings and Valve Lift of HCCI cases.
For HCCI the Valve Lift is different for different cases.
Calibration and validation of SI Model
The calibration of SI model is
done on the values of IMep, Pmax and Peak Pressure Crank Angle. During
calibration we get close value of IMep, Pmax, Pmax CAD. But while validation
gave close values of IMep when compared with other values. This is because we
can change Throttle angle for all cases untill we get close values.
Calibration and Validation of HCCI Model
The calibration of HCCI model is
done to get close values of IMep, Pmax and Pmax CAD. We got close values during calibration compared with
validation, as we get more differences during validation of other cases. That
is because of the lack of Heat Release Data.
Results and Improvements
Table 1- Results and comparison of Given data vs AVL
Table 1 shows the results of all cases of SI and HCCI on AVL BOOST
simulation and their comparison with the given data from the literature review.
For HCCI case 1 is used for
calibration and the validation is done on the remaining cases. In case of SI case
4 is used for calibration and remaining cases are validated. Other than the
IMep, Pmax and Pmax CAD, the calibration
is also done for the NOx emissions for both SI and HCCI. The error in IMep
during the validation of HCCI model is more I.e., because of the difference in
heat release data from the experimental values whereas the error in validation
of SI cases is very less.
For SI model case 4 has been taken to improve. The Improvements are done
for better IMep and Pmax values and less NOx emissions. By changing the speed
of the engine from 1500rpm to 2000rpm and 2500rpm the NOx emissions will be
reduced and the Pressure values will be increased.
Speed of the Engine
Table 2- Improvements of SI model (case 4 )
Table 2 shows the improvements of IMep, PMax and NOx emissions with
respect to the speed of the engine. By increasing speed from 1500rpm to 2000rpm
the NOx values decreased by 5 percent and at 2500rpm it is 17.54 percent.
In HCCI model case 1 is taken to improve. Like SI model, the same
improvement is carried out in HCCI also but at different speeds.
Speed of the Engine