Data processing: vehicles – navigation – and relative location – Vehicle control – guidance – operation – or indication – With indicator or control of power plant
Reexamination Certificate
1999-09-08
2001-07-03
Yuen, Henry C. (Department: 3747)
Data processing: vehicles, navigation, and relative location
Vehicle control, guidance, operation, or indication
With indicator or control of power plant
Reexamination Certificate
active
06256575
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a process for controlling an internal combustion engine and, more specifically, to a process using a model of the engine to define the commands to be applied in order to obtain a desired result.
Modern techniques for controlling internal combustion engines of motor vehicles place increasing reliance on mathematical modeling of the engines to produce more robust control processes capable of more accurately taking account of requirements imposed by drivers of the motor vehicles and of the constraints imposed by pollution-control regulations. Mathematical models enable output variables y of an engine to be estimated using a set of commands u that is input to the engine. For example, the model depicted in block form in
FIG. 1
, enables a set of commands u such as the set formed by the throttle position TPS, the position EGRV of an exhaust gas recirculation valve, the ignition advance IGA, the amount INJ of the fuel injected, etc., to be used to estimate output variables y of the engine such as the torque TQ supplied, the amount of air drawn in MAF, the richness LAM of the exhaust gases, and their recirculation rate EGR. These output variables cannot always be measured directly and economically on an actual engine.
FIG. 2
depicts, in simplified form, a known process for controlling an internal combustion engine. Such a process is known, for example, from French patent application 9700648 filed by the assignee of the present application. A set-point torque TQ_SP is determined by evaluating the depression of a throttle pedal. A set of commands u to be applied to means
11
for adjusting the engine
1
is determined by applying the set-point torque TQ_SP to an inverse model M
−1
of the engine. Other set points LAM_SP and EGR_SP that might be present can also be applied to the inverse model. However, an inverse model of this kind needs to be obtained by an analytical inversion of the direct model M shown in
FIG. 1
, which, in the case of multi-variable models (multiple inputs and multiple outputs) is an extremely complex operation. Furthermore, the various coefficients of the analytical equations that form the direct model are generally obtained experimentally by identifying the model with the actual engine. The coefficients are stored in numerous mapping tables. When the direct model is inverted, these tables have to be inverted and, given the non-linearity of the coefficients, this often leads to indeterminacy or inaccuracies which are prejudicial to the effectiveness of the control process. Furthermore, any modification to the direct model, even to just one coefficient, necessitates a further complete inversion of the model. Thus development and optimization is extremely lengthy and expensive.
SUMMARY OF THE INVENTION
The object of the present invention is therefore to propose an internal combustion engine management process which, while retaining the advantages of the processes of the prior art, does not display the difficulties associated with inverting the used models.
With the foregoing and other objects in view there is provided, in accordance with the invention, a process for controlling an internal combustion engine, which includes: determining a set of set-point values of output variables of an engine; implementing a mathematical loop including a direct model of the engine and a correction matrix for correcting inputs of the direct model as a function of outputs of the direct model; determining an inverse model of the engine by iteratively using the loop; and obtaining a set of commands to be applied to means for adjusting the engine by applying the set of the set-point values to the inverse model of the engine.
In accordance with an added feature of the invention, the process includes: supplying a set of initial commands to the direct model of the engine; obtaining a difference vector by comparing an estimate of output variables supplied by the direct model with the set of set-point values; processing the difference vector with the correction matrix to obtain a correction vector; and summing the correction vector with the set of initial commands to obtain a set of commands for application to the means for adjusting the engine.
In accordance with addition features of the invention, a set of set-point values of output variables of the engine is determined. These set-point values are applied to an inverse model of the engine and a set of commands are obtained that can be applied to means for adjusting the engine. The inverse model of the engine is produced from an iterative exploitation of a loop with a direct model of the engine and a correction matrix for correcting the inputs of the direct model as a function of its outputs.
According to the process that is the subject of the invention, a set of initial commands is supplied to the direct model of the engine. The estimate of the output variables which is supplied by the model is compared with the set of set-point values and a difference vector is obtained from this comparison. The difference vector is processed using the correction matrix to obtain a correction vector for correcting the commands, and the correction vector is summed with the set of initial commands to obtain the set of commands to apply to adjusting means.
In accordance with further features of the invention, the above steps are repeated with a predetermined temporal recurrence, using, on each iteration, the set of commands obtained in the previous iteration as the set of the initial commands. In a second implementation of the process, a series of iterations is triggered each time the engine reaches top dead center. The successive correction vectors are integrated, and the set of commands is applied to the adjusting means only when the relative difference vector is below a predetermined threshold. In this last implementation, each time the engine reaches top dead center, the set of commands applied at the previous top dead center is used as the set of initial commands.
In accordance with further added features of the invention, the correction matrix is obtained by the inversion or pseudo-inversion of the Jacobian matrix of the partial derivatives of the output variables of the engine with respect to the commands. Advantageously, the influence of slowly-varying parameters is neglected when calculating the coefficients of the Jacobian matrix, and the direct model receives a state vector representing the current operating conditions of the engine. The state vector can include parameters that are measurements of current operating conditions.
In accordance with a further additional feature of the invention, a first alternative form is provided wherein the correction matrix is determined experimentally during engine development testing and its coefficients are stored in a table as a function of the operating conditions of the engine.
In accordance with a concomitant feature of the invention, there is provided a second alternative form wherein the Jacobian matrix is determined during each iteration from an estimate of the partial derivatives which is drawn from the calculation of the variation of the outputs of the direct model as a function of a unit variation in one of its inputs about the current operating point.
Other features which are considered as characteristic for the invention are set forth in the appended claims.
Although the invention is illustrated and described herein as embodied in a process for controlling an internal combustion engine, it is nevertheless not intended to be limited to the details shown, since various modifications and structural changes may be made therein without departing from the spirit of the invention and within the scope and range of equivalents of the claims.
The construction and method of operation of the invention, however, together with additional objects and advantages thereof will be best understood from the following description of specific embodiments when read in connection with the accompanying drawings.
REFERENCES:
pat
Greenberg Laurence A.
Lerner Herbert L.
Siemens Automotive S.A.
Stemer Werner H.
Vo Hieu T.
LandOfFree
Process for controlling an internal combustion engine does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Process for controlling an internal combustion engine, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Process for controlling an internal combustion engine will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2526984