CADFamily.com: 02/01/2012

The engine studied is a new one : 3.2L, V6, GDI with double
VVT (intake and exhaust).
The engine has been studied in the maximum power
configuration and the performances have been studied at
different engine speed, from 1000 to 7000 [rpm] with step of
500 [rpm].
For the performance predication a commercial 1D fluid-dynamic
solver (Gtpower) has been coupled with the optimisation
algorithm.

The VVA system is not yet present on the engine. For this
reason, the optimisation technique has been used in the project
phase as mean to evaluate engine performance variation.
The aim of the work is the optimisation of engine torque
using Variable Valve Actuation Systems: VVT for intake
and exhaust timing and VVA for the intake valve lift and
opening duration .

The variable valve actuation systems used for the optimisation
work are:
- VVT ( Variable Valve Timing): this system is able to change the
position of valve start opening (see1)
- VVA (Variable Valve Actuation): this system is able to change valve
lift and opening duration

NOTE:The combined use of these two systems allows the users to obtain the required
valve lift characteristics for typical performance objectives.

Before the optimisation phase the model has been tuned
using some experimental data, such as:
1) Pressure waves on the intake and exhaust manifold
2) In cylinder pressure
3) Combustion rate
4) Friction
5) Pressure losses on the intake and exhaust systems
6) Temperature on the catalyst cupel (intake and exhaust)

NOTE: The VVA system is able to create different valve lift and opening
duration. Using modeFrontier, it is possible to consider the valve lift and
opening duration plots as numbers, and so as parameters. Following that way,
calling the files relative to them with numbers (increasing at increasing lift)
the optimisation solver is able to evaluate their influence in a correct
way.

http://www.cadfamily.com/HTML/Tutorial/Optimisation%20of%20engine%20torque%20using%20%20variable%20valve%20actuation_305312.htm

Application of a Quasi-Dimensional Combustion Model

The above values were held fixed for the calculation
of any operating condition of the engine. In addition, they
are very similar to the ones identified for the simulation
of a very different research engine [5]. This practically
means that the model description of the kernel duration,
initial flame development, turbulent flame propagation
and combustion termination, is able to reproduce the
underlying physics in a satisfactory way.
Fig. 5-Fig. 7 report the mean performance
parameters of the engine at WOT. Agreement with the
experimental data is satisfactory all over the engine
speed range. An increased volumetric efficiency can be
observed at about 3000 rpm (Fig. 5). The 1D
schematization of the intake and exhaust pipe network
probably determines an overestimation of the gas-
dynamic tuning at this particular engine speed. The latter
inaccuracy also reflects in the IMEP, BMEP, Power and
BSFC profiles in Fig. 6 and Fig. 7.
A more detailed comparison is presented in Fig. 8-
Fig. 13, in terms of instantaneous pressure cycles at
WOT. The figures also report the pressure cycles
computed with the base version of the model (eq. (2)).
The correction proposed (eq. (16)) enhances, as
expected, the burning rate at high speeds, and
considerably improves the prediction of the in-cylinder
pressure peak, especially in the medium-speed range. In
each operating condition, combustion start, maximum
pressure and expansion phase are well reproduced by
the improved fractal model.
In order to check the model accuracy at part-load
too, some analyses were carried out at fixed rotational
speed (2000 rpm) and for different load levels (WOT and
2.3 bars BMEP). Two different VVT positions (0° and 25°
cam angles) were also analyzed. The results obtained
are summarized in Fig. 14-Fig. 16. The prediction of the
pressure cycle is satisfactory also in these more critical
operating conditions on both high pressure cycle (Fig.
14) and mass exchange phase (Fig. 15). Fig. 15
particularly puts into evidence the strong reduction of the

pumping work achieved through a delayed camshaft
position, reflecting in a relevant BSFC improvement. The
coupled effects of the spark advance, residual fraction
level at intake valve closure, and valve timing really
determine a very different development of the
combustion process, as shown in Fig. 16.
In the tested operating conditions, a maximum EGR
level of about 24% was reached (Fig. 16). A further
validation of the combustion model with a percentage of
residual gases greater than 30-40% is however required
to fully asses the model accuracy.

http://www.cadfamily.com/HTML/Tutorial/Application%20of%20a%20Quasi-Dimensional%20Combustion%20Model_305309.htm

Robust Optimization of ABS Control Parameters

ABS for Automobiles
Anti-lock Braking System
-Prevents wheel lock-up to secure steerability and controllability in
emergency braking situations.
-If wheel rotation exceeds the stability limit, brake pressure will be
decreased to adjust the rotation to be within the stable region.
-Then, brake pressure will be increased again until it exceeds the stability
limit to maintain the rotation status of the stability limit region as long as
possible.

Multi Objective Robust Design Optimization
Perform global exploration by considering the issue as
multi-objective optimization problem of
-Output average value: Maximization (minimization)
-Output standard deviation: Minimization
when probability distribution variation is applied to the input.

Optimization Process
Step1
-Scheduler: FMOGA (RSM Evaluation: 0.8/Linear annealing)
-First generation individuals: 25 DOE (Random)
-Number of generations: 15
-Number of robust sampling: 110 (Actual samples: 10, RSM: 100)
Step2, 3
-Scheduler: FMOGA (RSM Evaluation: 0.6/Fixed)
-First generation individuals: 20 Pareto solutions for previous Step
-Number of generations: 15
-Number of robust sampling: 110 (Actual samples: 10, RSM: 100)
CPU: Pentium 3 (1.0GHz) Dual (RAM: 780MB)
Total calculation time: Approximately 25 hours

http://www.cadfamily.com/HTML/Tutorial/Robust%20Optimization%20of%20ABS%20Control%20Parameters_305320.htm

Integration of AVL Boost with modeFrontier

-The optimisation target is given by the squared
difference between experimental and simulation data,
relatively to the 3 conversion curves (CO, HC and H2)
-Once the LO_optim.bwf model is set and the target
functions are built, the integral value Y_MEAN_INT_DX
for all the three target curves is defined as output variable
(SD_CO, SD_HC, SD_H2)
-The global optimisation objective is given by the
minimisation of global_error = (SD_CO+SD_HC+SD_H2)

-A different approach has been considered:
-After 718 designs calculated by NSGAII (global_error=50),
multi-objective scheduler MOGT is run (minimise SD_CO,
SD_HC, SD_H2 as 3 different objectives)
-MOGT starts from a good configuration (#718) and sets 3
constraints on the values of the 3 objectives, that should
be less than the ones of the starting configuration

-A last approach has been considered:
-Simplex mono-objective algorithm is used from random
DOE (particular efficient in problems that are not highly
non-linear)

-An example of integration of AVL Boost with modeFrontier
has been shown
-The test case was relative to the calibration of 6 kinetic
parameters in a catalytic conversion reaction
-Different optimisation algorithms has been used to minimise
the difference between the experimental and the simulated
conversion curves

http://www.cadfamily.com/HTML/Tutorial/Integration%20of%20AVL%20Boost%20with%20modeFrontier_305322.htm

Coupling AMESim v4_2 with modeFRONTIER

Create the simulation model and assign the
Input Parameters which shall be used as
input variables for modeFRONTIER
-Specify Simple Output and/or Compound Output Parameters to
be processed as Output Variables by modeFRONTIER

-Using the selection by Introspection, this facility lets you specify AMESim parameters by
selecting the parameter names directly in the introspection dialog
-You can start the Introspection process by clicking on the small folder near the parameter name.
-A dialog will immediately report you about the introspection progress

Run modeFRONTIER Node Interface–Assign Assign
Variables Variables
-The Introspection dialog lets you specify the desidered parameter,
selecting it on the Parameters Table.
-Alternatively you can type the parameter name on the Selected
Parameter textfield.
- The Parameters Table contains all the parameters available in the
project.

http://www.cadfamily.com/HTML/Tutorial/Coupling%20%20AMESim%20v4_2%20with%20modeFRONTIER_305323.htm