Beltrami 2003

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 6

AFR control in SI engine with neural prediction of cylinder air mass

C. Beltrami', Y. Chamaillard", G. Millerioux', P. Higelin**and G. Bloch'


' : Centre de Recherche en Automatique de Nancy (CRAN, UMR CNRS 7039)
ESSTIN, 2 rue Jean Lamour, 545 19 Vandoeuvre les Nancy Cedex, France.

..

gerard.bloch@esstin.uhp-nancy.fr

Laboratoire de Mecanique et d'Energitique (LME, EA 1206)


ESEM, 8 rue Leonard de Vinci, 45072 Orleans Cedex 2, France.
yann.chamaillard@univ-orleans.fr

Abstract
Accurate Air-Fuel Ratio (AFR) control in a sparkignition engine is a critical point to satisfy pollutant
emission legislation. Using a three-way catalytic
converter with an electronic fuel injection, today's most
effective solution, requires the regulation of the cylinder
AFR in a narrow band around the stoichiometric
conditions during both steady and transient engine
operation to be efficient.
AFR control depends essentially on prediction of the air
mass to be admitted. In this paper, the building of an air
mass predictive neural network is described and its
performances are evaluated. Using this predictor in
addition with transient fuel film compensation for AFR
control allows to drastically reduce the AFR excursions
during fast transients.

Universal Exhaust Gas Oxygen (UEGO) sensor in the


exhaust flow for the AFR measurement to allow a
possible bias to be corrected by feedback.
The control problem becomes more difficult in transient
phases because of the more difficult prediction of the air
mass, the fuel flow dynamics and the inherent delay in
the feedback system. This results in AFR excursions
during fast transients, and so increased pollutant
emissions. A lot of work has been performed in the topic
of air mass prediction [9] [ I l l [I61 and fuel film
dynamics [I] [3] [7] to improve AFR control.

Keywords
air-fuel ratio, event-based control, prediction, neural
network.

1. Introduction
In today's spark ignition engines, three-way catalysts are
used to reduce the exhaust emission of the three main
pollutants that are: unburned hydrocarbons (HC), carbon
monoxide (CO) and nitrogen oxides (NOx). The
optimization of the three-way catalyst efficiency requires
the cylinder Air-Fuel Ratio (AFR) to be kept in a narrow
band which corresponds to the stoichiometric conditions
[6].Figure 1 describes the catalytic conversion efficiency
for the three main pollutants versus the in-cylinder
mixture AFR. Even a small deviation from
stoichiometric conditions can result in a dramatic
degradation of the conversion efficiency.

A modern engine control unit, as the ones commonly


installed on new vehicles, handles this AFR regulation
task very well under steady state conditions [4]. It
provides the injection controller with a prediction of the
air mass to be admitted in the cylinder and uses a

0-7803-7896-2/03/$17.00
02003 IEEE

Air-Fuel Ratio (AFR)

FIG.1: Catalytic converter efficiency


The AFR is defined as the ratio between the air mass and
the fuel mass admitted into the cylinder. These variables
are not accessible for measurement but depend
essentially (though dynamic systems) on throttle angle
and injection duration.
In a first part, a description of the two sub-systems, fuel
and air dynamics, involved in the AFR determination is
provided to understand the process. The AFR control
method is then detailed and the air mass prediction issue
developed. Finally, a solution using a neural air mass
predictor, with a physical model based structure, in
addition to transient fuel film compensation, is proposed
and evaluated.

1404

Proceedings 01 the American Control Conference


Denver, Colorado June 4 6 . 2 0 0 3

. .

2. Engine model

where

The simulations are performed on a non-linear fuelinjected,


mean-value and event-based model.
Computation is performed at each,Top Dead Center. The
engine model includes the engine (fuel flow dynamics,
air mass flow dynamics, combustion and delays inherent
to four-stroke engines), actuators, sensors and a dynamic
model of the load. This model is representative of a
classical four cylinders 1.4 liter engine.

2.1. Fuel dynamics


The fuel flow sub-model describes the fuel transport
from injection location to intake port. A two phase fuel
flow occurs in the intake manifold, with a thin film of
fuel on the manifold walls and droplets transported by
the main stream of air and fuel [I]. It is worth
emphasizing that the fuel film dynamic model is nonlinear in spite of its linear form. Indeed, the parameter
values depend on the engine operation states and are not
constant. Moreover their identification is relatively
complex.
It is assumed that at any time there are uniform
conditions in the intake manifold and that a fraction X of
the injected fuel is deposited on the wall as liquid film.
The evaporation is considered proportional to the mass
of the fuel film. The phenomena can he described by a
model with two time constants [7]:

is the desired fuel mass flow and

m,

dP(0
dt = a Sthr(t) fsV(P(t)/ Putm)

- $ Ne@ PltJ f@(t), Ne(t).Ta)


where

y R Tu Putm

a=---Jp-

Ma Vmun R Tu

with P(t)

W O
fa0

(2)

T"

with mg fuel film mass flow (kgkec),


mp injected fuel mass flow (kg/sec),
m,

(7)

2.2. Air dynamics


The air intake sub-model describes the air mass flow
from the throttle to the cylinder port [ 2 ] . The only input
which can he controlled is the throttle angle that
modifies the intake manifold pressure:

Wt)

m, = m, + m s (3)

1
'
=-(-Gg

+Xm,)
z/
Linear compensation based on the equations above
cannot he achieved to give optimal compensation over
the entire operating range of an engine, especially in
transient condition. This fact has been pointed out in
several earlier publications and is very important in
practical applications [7]. However, the compensation
presented above permits to sensibly reduce the fuel
dynamics effect on AFR and so, to better appreciate the
air estimation effects.
mg

Sthr(t)
1
m, =-[-mfi+(l-x)m,)]

iff
the fuel

film mass flow estimation defined as in ( I ) by:

Patm
Tu
Vmun
and bR.Ma

and

(84

f l = 30 yMuR Ta
Vmun (8h)

manifold pressure (Pa),


effective throttle section (m'),
engine speed (rpm),
Saint-Venant function,
filling function,
atmospheric pressure (Pa),
air temperature (K),
manifold volume (m3),
thermodynamic constants.

fuel vapor mass flow (kdsec),

cylinder port fuel mass flow (kgisec),


x fraction of injected fuel deposited as fuel film,
z, fuel vapor time constant (sec)
and r y fuel film time constant (sec).
mf

In an injection system the vapor time constant r, can


usualIy be neglected with respect to the fuel film time
constant zr and equation ( 2 ) becomes:
m, = ( I - X ) k , (4)
Hence, the global equivalent transfer function that links
the fuel mass injected and the fuel mass admitted can be
described by:
Mf(s) = l + ( l - X ) z p

MjiW

I+rfs

(5)

From (3) and (4), an ideal compensation for the


simplified model can he obtained:
1
mfi =-(mp

1-x

-6,j-I

(6)

The first term of (Sa) corresponds to the entering flow


from the throttle. The second term represents the exiting
flow that is admitted into the cylinder and depends
essentially on manifold pressure and engine speed. The
air dynamics is fairly complex and non-linear, and is a
central problem in AFR regulation

2.3. Event-based model


A closer look at the engine processes shows that the
operations divide the physical processes into four distinct
regimes corresponding to the four events: intake,
compression, power and exhaust, and suggests an eventbased approach according to the crank angle.
As a result, the characteristic behavior of an engine
consists of a combination of two types of dynamics:
time-based and event-based. Event-based dynamics are
described in the crank angle domain. From the engine
control point of view, only one value of AFR exists at
each cycle for each cylinder, and the outputs of an
engine control system are synchronous with crank angle.
Hence, the fundamental sampling period Te, constant in
1405

Proceedings Of the American Control Conference


Denver, Colorado June 46.2003

the crank-angle domain, but varying in the time domain,


corresponds to the intake event occurrence rate and is
defined by:

combined with air estimation by volumetric efficiency


map is the most used.
3.2. AFR control strategy
The general scheme is described in figure 3.

where nC,,, i s the cylinder number and Ne the engine


speed (rpm). In the crank angle domain, the AFR at the
end of the exhaust phase is defined by:

where Muir(k - 3) and Mfueljk - 3j are respectively the


air mass and the fuel mass admitted in the cylinder at the
end of the intake phase. The delay corresponds the
inherent delay in a four-stroke engine between the end of
the intake and the exhaust phase.

3. AFR control
There are two main kinds of AFR control systems: fuel
control and air control [5]. The fuel injection control
system, because of torque control, is the most common.
It regulates the admitted fuel mass according to the
admitted air mass, which is the time integral of the air
mass flow during one engine cycle. AFR conlrol is there
an element of a global torque control strategy that first
controls the air mass flow in order to satisfy the torque
reference and then adjusts the injection duration to be in
desired conditions.

3.1. AFR control equipment


The actuators and sensors (see figure 2) consist of
electronic throttle, electronic injector, electronic ignition,
intake manifold pressure sensor, engine speed sensor and
UEGO sensor for AFR measurement.

Control

reference
Measured
AFR

FIG.

3: AFR Control

A Luenberger observer is built to predict the manifold


pressure. This observer, obtained from (X), after
appropriate linearization and discretization, is iven by:
i ( k + 1) = Ad P(k)+ Bd + Kjkj (P(k)-C i(k)y
(1 1)
where P(kj is the prediction at discrete time k of the
pressure P(k) , Ad, Bd, C are resulting scalar constants,
and K(k) the time varying observer gain.
The admitted air mass is then predicted by a volumetric
efficiency map from the estimated manifold pressure and
engine speed, and, in turn, the fuel mass to be injected is
calculated from this prediction and the engine speed.
Fuel film dynamics compensation is also achieved using
(6).

When the admission occurs and the intake valve closes,


the AFR is already determined. However, its
measurement is only possible after the exhaust phase to
correct the injection duration. Classically, this feedback
uses a Proportional Integral (PI) controller that is able to
correct a bias on the air estimation thanks to its integral
action, but cannot manage the transient errors.

3.3. AFR control issues

1"takt

manifold
pressure

---* Engine speed

I I I
d Control Jignal

------+s m m r signa1

FIG. 2: Basic SI engine

Two methods are commonly used for determining


admitted air mass flow:
air mass sensing, with mass flow meter,
conventional volumetric efficiency method, with
manifold absolute pressure sensor and speed sensor.
In both cases, the goal is to provide an input to an air
induction model that will estimate the air charge. Among
all these methods, AFR measurement by UEGO sensing

As stated in [SI, several sources of AFR excursions have


been identified during engine transients including
manifold air filling, air mass prediction error, fuel flow
dynamics, feedback delay and back-flow.
The feedback delay, the sensor dynamics, slow with
respect to the variation to be detected, and the fuel flow
dynamics lead to inefficiency of the feedback scheme
during transients. Nevertheless, the main problem to be
tackled lies in the estimation of the air mass that will be
admitted in the cylinder. As in port fuel injection engines
the injection system requires time to dispense fuel, the
injection should be completed before the intake valve
opens and the injection controller must get a prediction
of the air mass to be admitted before a direct estimation
is available. The problem is due to the fact that during
transient, operating conditions change between the
instant the estimation is done and the admission phase,
resulting in prediction enors. This prediction is essential
1406

Proceedings Of the American Control Conference


Denver. Colorado June 4-6.2003

in AFR regulation. A solution for air prediction is


proposed in the following section.

4. Neural air mass prediction


4.1. Introduction
As the AFR measurement presents a delay and the sensor
dynamics is slow with respect to the variation to he
detected during transient phases, a feedforward control
seems to be the solution during transients. For such a
scheme, the AFR regulation quality depends essentially
on the prediction of the air mass to be admitted. From the
physical model of the air admission dynamics, given by
(S), the goal is to obtain a discrete event-based model of
the air admission in order to predict the air mass flow to
be admitted in the cylinder. The delay between the angle
reference and the actual throttle position can be used to
develop an air charge anticipation algorithm. Magner and
Jankovic (2002) develop such a solution using a neural
predictor [ I l l . Other works [SI [lo] [I21 already used
neural networks to optimize AFR control.
Because of their ability to represent complex non-linear
mappings with good flexibility and accuracy, neural
networks have become popular to model various
subsystems as discrete black boxes [13] [14].
As parsimonious and flexible universal approximator,
the one hidden layer perceptron with linear output unit is
used here. Its form is given, for single output f , by:

where pj, j = I ; . . , p , are the inputs of the network,


I
wki
and bIk , k = f : . . , n , j = f : . . , p , are the weights

and biases of the hidden layer, the activation function g


is a sigmoid function, chosen here as often as the

4,

hyperbolic tangent,
k = I ; . . , n , and bZ are the
weights and bias of the output neuron or node.
The non linear model (11) can be used as discrete
dynamical predictor of a variable y :

4.2. The neural model


As the variable to be estimated, here the air mass, is not
measured, a simulation model, involving outputs
predicted by the model in the regression vector, is
needed. Hence, a Neural Output Error model WOE) is
used. To predict j ( k ) , the air mass to be admitted at
discrete time k, the following regressors have been
chosen:

Air mass prediction at (k - I ) , $(k - I),


Manifold pressure P(k - I ) and P(k - 2) ,
Engine speed Ne(k - I),
Throttle angle reference Th+@(i), i = k-l.k-b...,k-6 .
This choice is clearly based on physical equations (8),
which involve as dynamical inputs the manifold pressure
Po), effective throttle section Sthr(t) and engine speed
Ne(?). Including P(k - I) and P(k - 2) reflects the
presence of P(t) time derivative in (8). The engine
speed, beyond its role in air admission model, permits to
handle the variable sampling period Te issue. The last
regressors allow the prediction thanks to the delay
present in the throttle actuator which is around 30 ms. At
6000 rpm, the sampling period is 5 ms and 6 samples are
then necessary. The same choice is made in [I I]. The use
of a rapid throttle could reduce the delay and so the
number of regressors.
Training was performed by minimizing the mean
squared error function, with the Levenherg-Marquardt
method implemented in a specific Matlab toolbox [15].
The different signals involved in training the network
should have been scaled to avoid saturation. A hidden
layer of n = 14 neurons (see eq. 12) was selected to
reach a good prediction accuracy.
The training data set was obtained by simulating the
engine on a large range of operation. The torque
reference signal consisted of steps of random length and
size, to which was added up a random step signal with
length and amplitude divided by IO, as shown figure 4.

m=f(m),e8) (13)
where

p(k) = [pl(k)pz(k)...pp(k)P'

is the regression

vector and the parameter vector B is the concatenation


of all the weights w and biases b . Depending on the
choice of the regressors in p@),different models can be
derived.

FIG.4: Engine torque reference (daN.m) vs. time (sec)


The whole engine operation range for different engine
speeds was covered. The speed reference signal was

1407

Proceedings of the American Contml Conference


Denver, Colorado June 4-6, 2003

varying from 1000 to 6000 rpm by step of 1000 rpm. As


the sampling period depends on the engine speed, the
step duration varied with the speed reference to keep the
same learning points number for each level.

5. Results
Two simulation scenarios can be considered for
validation: the engine speed scenario and the torque
scenario. In [4], it is shown that transients in torque are
the most disturbing. So the torque scenario is used here
for comparison.
The main task was to obtain an air mass predictor in
order to enhance AFR control in transient phases. As the
transient fuel dynamics compensation was not the main
problem, results will be compared with the same fuel
admission corrected (in simulation) by the ideal
compensation (6) and (7).
Different simulations have been performed to test the
neural air mass predictor. The torque reference
represented on figure 5 is chosen to generate fast throttle
angle variations and thus rapid transient phases.

Table I - Results
The results show that the neural prediction leads to a
very significant improvement in AFR control thanks tn
its better prediction of the air mass to be admitted. The
neural network interpolates the learning data very well,
hut, for extrapolation, the performances fall down
compared to traditional method (at 6500 rpm for
example).
Results at 3500 rpm are shown in figures 6 and 7, during
only 3 seconds to better illustrate the differences. Figure
6 shows the neural air mass prediction compared to the
real (simulated) air mass. It can he noticed that the
prediction errnr is very weak and the real and predicted
air masses are difficult to distinguish

FIG. 5 : Torque reference (daN.m) vs. time (sec)


That signal is used with different engine speed references
from 1000 to 6500 rpm by 500 rpm step to compare the
results with data similar but different from the learning
set. The simulations have been done with the ideal fuel
film compensation and a PI controller on AFR
measurement (in the AFR controller - figure 3) to avoid

bias.
The test results of the one step ahead neural predictor are
reported in table 1. The engine speed reference value
used with the torque reference is reported in the first
row. The Root Mean Square Error (RMSE) values for
the air mass prediction with a traditional method (Air-t)
(prediction by a volumetric efficiency map from
estimated manifold pressure, given by ( I I), and engine
speed) and for the neural prediction (Air-nn) are
reported in the second and third rows. The last two rows
give the RMSE on AFR control results with traditional
(AFR-t) and neural (AFR-nn) predictions.

1.5

Real air mass


Neural prediction
Predictionerror

8 1

II

8.5

10

FIG. 6 : Predicted and real air mass (mg) vs. time (sec)

As previously mentioned, the good capability of the air


mass neural predictor allows to significantly enhance the
AFR control. Figure 7 shows a comparison between the
AFR excursion with the traditional air mass flow
predictor (AFR-t) and the neural one (AFR-nn).

In all cases the AFR excursions are reduced (by 50%)


especially for high excursions, which are the most
problematic for consumption, agreement and pollutant.
However the static error compensation is rather slower
because the feedback controller has not been redefined.

1408

Proceedingsof the American Conlrol Conference


~ e n v e rcoioraao
.
ne 4-6, zw3

7s

15

95

10

FIG. 7: Comparison ofthe AFR errors vs. time (sec)

6. Conclusion
The need of an accurate prediction of the air mass to be
admitted in the cylinder has been emphasized in the
framework of AFR control. A neural network can be
built and trained to provide a good dynamical air mass
prediction, much better than the prediction based on
classic observer and static volumetric efficiency map.
The neural predictor makes complete use of the delay in
the throttle actuator. For operating points inside the
learning domain, the neural network interpolates very
accurately.
A solution combining this neural air mass one step ahead
predictor and a transient fuel film compensation has been
proposed for AFR control. The results show that the
AFR excursions are drastically reduced on rapid torque
transients if the inputdoutputs of the air admission can
be correctly collected. It appears also that the feedback
controller must be redefined to optimize static error
compensation.
Although the neural dynamical prediction of cylinder air
mass greatly improves the AFR control, further works
must be completed for application to handle the data set
collection and the system non-stationarity over time.

References
1. Arsie, C. Pianese, G . Rizzo, and V. Cioffi. An
adaptive estimator of fuel film dynamics in the
intake port of a spark ignition engine. 3"' IFAC
Workshop Advances in Automotive Control, pp,
293-298, Karlsruhe, Germany, 2001.
P. Anthoine and A. Dauron (1993). Dtpollution des
moteurs a essence: regulation de richesse avec
sonde proportionnelle et actionneur papillon.
Automatiquepour les vihicules, pp. 57-72, Amiens,
France, 1993.
M. Behnia and B.E. Milton. Fundamentals of fuel
film formation and motion in spark ignition engine
induction systems. Energy Conversion. and
Management, 42, 1751-1768,2001.
Y . Chamaillard and C. Pemer. Air-fuel ratio control
by fuzzy logic, preliminary investigation. 3'd IFAC

Workshop Advances in Automotive Control, pp.


221-226, Karlsruhe, Germany, 2001.
[5] C.F. Chang, , N.P. Fekete, and J.D. Powell. Engine
air-fuel ratio control using an event-based observer.
SAE Paper 930766, Detroit, MI, 1993.
[6] R.M. Heck and R.J. Farrauto. Automobile exhaust
catalysts. Applied Catalysis, A : General, 221, 443457,2001.
[7] E. Hendricks, T. Vesterholm, P. Kaidantzis, P.
Rasmussen, and M. Jensen. Nonlinear Transient
Fuel Film Compensation (NTFC). SAE Paper
930767, Detroit, MI, 1993.
[8] R.J. Howlett, S.D.Walters, P.A. Howson, and L.A.
Park. Air-fuel ratio measurement in an internal
combustion engine using a neural network.
Advances in vehicle control and safety, Amiens,
France, 1998.
[9] M. Jankovic and S. Magner. Cylinder air-charge
estimation for advanced intake valve operation in
variable cam timing engines. JSAE, 22, 445-452,
2001.
[IO] N. Li, K. Li and S. Thompson. Employing a new
type of neural network to optimise power plant airfuel ratio. 141h IFAC Triennial World Congress, pp.
333-338, Beijing, China, 1999.
[ I l l S. Magner and M. Jankovic. Delta air charge
anticipation for mass air flow and electronic throttle
control based systems. American Control
Conference, pp. 1407-1412, Anchorage, AK, 2002.
[I21 M. Majors, J.A. Stori, and D.Cho (1994). Neural
network control of automotive fuel injection
systems. IEEE Conirol Systems Magazine, 14(3),
31-36, 1994
[I31 K.S. Narendraand K. Parthasarathy. Identification
and control of dynamical systems using neural
networks. IEEE Trans. on Neural Networks, 1(1),
4-27, 1990.
[I41 M. Norgaard, 0. Ravn, N.K. Poulsen and L.K.
Hansen. Neural networks for modeling and control
of dynamic systems. Springer, 2000.
[I51 M. Norgaard. Neural Network Based System
Identification Toolbox. Technical Report 95-E-773;
Institute of Automation, Technical University of
Denmark, 1995.
1161 R.W. Weeks and J.J. Moskwa. Transient airflow
rate estimation in a natural gas engine using a nonlinear observer. SAE Paper 940759, 1994

1409

Proceedings of the American Control Conterence


Denver. Colorado June 4.62003

You might also like