EP2232035B1 - METHOD FOR PRODUCING A SYNCHRONIZATION SIGNAL for an INTERNAL COMBUSTION ENGINE - Google Patents

Description

The present invention relates to a method for generating a synchronization signal, representative of the unfolding of the operating cycle of a four-stroke internal combustion engine, of the multicylinder type, the expansion phases of each cylinder, during which the combustions take place. , occur at different angular positions of the crankshaft rotation motion as is the case with odd-numbered four-stroke engines.

The invention more specifically relates to a method for generating a signal for locating a predetermined time of the cycle such as the passage to the High Admission Dead Point or to the Low Admission Dead Point.

The performance of an engine as well as the control of the emission of pollutants are related to the various control processes governing the operation of the engine. These methods, for example fuel injection or ignition, require precise knowledge of the current thermodynamic cycle in the engine cylinders.

The document FR 2,441,829 , proposes a way: to detect information on the thermodynamic cycle of the cylinders by locating, on a target secured to the crankshaft, the angular position zones corresponding to a specific phase of the stroke of the different pistons. The target consists of a disc having locating elements disposed along its periphery, such as teeth of different lengths. A fixed receiving member detects these locating elements and generates electrical pulses to produce a signal identifying the passage to the top dead center position of a determined piston.

Such a tracking device is however insufficient. Indeed, for a four-stroke internal combustion engine, the crankshaft performs two complete turns (or 720 ° angle), before a given piston is found in the same operating position in the engine cycle. As a result, from the only observation of the rotation of the target attached to the crankshaft, it is not possible, in principle, to provide information on each cylinder without an indeterminacy of two engine times in the cycle (the identification of the top dead center position covering both the Admission phase and the Relaxation phase).

The precise determination of the position of each cylinder in the cycle can not be deduced from the only observation of the position of the crankshaft, the search for additional information is necessary to know if the cylinder is in the first or the second half of the engine cycle (Admission and Compression phases during the first crankshaft revolution, Relaxation then Escape phases during the second lap).

In order to obtain such additional information, it is known to use secondary locating elements carried by an emitting disk which rotates half as fast as the crankshaft. For this purpose, it is possible to arrange this emitter disk on the camshaft or on any other shaft which is driven via a gearbox 1/2 from the crankshaft.

Combining the signals from the crankshaft sensor and the camshaft sensor allows the system to accurately detect the High Dead Point in the Admission phase of a reference cylinder.

However, such angular tracking systems using both a crankshaft sensor and a camshaft sensor, are relatively bulky, expensive, and difficult to mount.

In order to remedy these drawbacks, the publication FR 2,749,885 proposes a simple and effective method of locating which does not require any specific position sensor apart from that which serves to locate the angular position of the crankshaft.

This method uses a timing signal that is generated from the combustion conditions in each of the four-cycle and four-cylinder fitter cylinders, and information transmitted by the crankshaft sensor.

For this, at least one factor governing the combustion in a given reference cylinder is modified so as to cause an alteration controlled combustion. This alteration of the combustion in the reference cylinder is then detected thanks to a value Cg developed from the information from the crankshaft position sensor of the engine, thus making it possible to synchronize the passages at the Top Dead Center Admission of the engine cylinders with the Top Dead Center signal of the crankshaft sensor.

However, this invention requires a degradation of the combustion of the engine altering its operation and increasing pollutant emissions.

The present invention therefore aims to overcome the drawbacks of known tracking systems in the case of a four-stroke engine having an odd number of cylinders, by providing an improved registration method, requiring no specific position sensor outside of that which serves to locate the angular position of the crankshaft, and does not affect the operation of the engine, for example of the type described in the document DE-A1-102005043129 .

• fonctionnement du moteur durant une période donnée avec un allumage des cylindres à chaque tour des cylindres, de manière à produire une combustion systématique du carburant injecté,
• calcul du signal représentatif
• comparaison du signal représentatif (Cg, Bn) d'un cylindre lors d'un premier tour du cycle (Cg1_1, ${\mathit{B}}_{\mathit{n}}^{\mathit{Comb}}$
  
  ) avec le signal représentatif du cylindre lors d'un second tour du cycle (Cg1_2, ${\mathit{B}}_{\mathit{n}}^{\mathit{ech}}$
  
  ), de manière à connaître la phase du premier tour.
• réinitialisation du signal de synchronisation si l'analyse de la comparaison indique que le phasage du signal de synchronisation était erroné.

To this end, the invention proposes a method for producing a synchronization signal of an odd-numbered four-stroke internal combustion engine from cylinders by an electronic control system, the synchronization signal enabling the identification of a predetermined instant in the engine. thermodynamic cycle of each of the cylinders of the engine being determined from a signal identifying a determined position of each cylinder, and a signal representative of a magnitude representative of the kinematics of the crankshaft generated by each combustion, both generated at from the information of a crankshaft position sensor of the engine, the method comprising the following steps:

• operating the engine for a given period of time with ignition of the cylinders at each revolution of the cylinders, so as to produce a systematic combustion of the injected fuel,
• calculation of the representative signal
• comparing the representative signal (Cg, Bn) of a cylinder during a first revolution of the cycle (Cg1_1, ${\mathit{B}}_{\mathit{not}}^{\mathit{Comb}}$
  
  ) with the signal representative of the cylinder during a second round of the cycle (Cg1_2, ${\mathit{B}}_{\mathit{not}}^{\mathit{ech}}$
  
  ), so as to know the phase of the first round.
• resetting the synchronization signal if the analysis of the comparison indicates that the phasing of the synchronization signal was wrong.

According to other features of the invention, the representative signal may be a representation of the gas pair or a harmonic representation of the tooth duration.

• La figure 1 est un schéma du dispositif de contrôle moteur intégrant le procédé de l'invention ;
• La figure 2 est un schéma détaillant les étapes d'un procédé d'injection utilisant le signal de synchronisation de l'invention :
• La figure 3 représente le phasage du signal de synchronisation NOCYL selon l'invention.

Other characteristics and advantages of the invention will appear on reading the description which will now be made with reference to the appended drawings, in which:

• The figure 1 is a diagram of the engine control device incorporating the method of the invention;
• The figure 2 is a diagram detailing the steps of an injection method using the synchronization signal of the invention:
• The figure 3 represents the phasing of the synchronization signal NOCYL according to the invention.

The references used throughout the description are the same to designate identical or similar elements, as to their function, regardless of the variant embodiment of the invention.

Referring to the figures, an engine control system implementing the method for generating the synchronization signal object of the present invention has been shown. Only the constituent parts necessary for understanding the invention have been shown. Moreover, in this example, the synchronization signal makes it possible to control an injection system, but the use of the signal is not limited and the synchronization signal can be used to control other elements or processes of the engine.

A four-stroke internal combustion engine for a motor vehicle comprises three cylinders (C1, C2, C3) each comprising a fuel injection device of the multipoint type with control electronics whereby each cylinder is supplied with fuel from a specific electro-injector 5.

The opening of each electro-injector 5 is controlled by the electronic engine control system 7, which adjusts the amount of fuel injected and the injection time (Inj) in the cycle according to the operating conditions of the engine, so that precisely slaving the richness of the air-fuel fuel mixture admitted into the cylinders to a predetermined target value.

The electronic engine control system 7 conventionally comprises a microprocessor (CPU), random access memories (RAM), read only memories (ROM), as well as analog-digital converters (A / D), and various input and output interfaces. exits,

The microprocessor includes electronic circuitry and appropriate software (10, 222, 223, 224) for processing signals from suitable sensors, determining engine states, and implementing predefined operations to generate control signals at the desired destination. in particular injectors so as to better manage the combustion conditions in the engine cylinders.

The electronic engine control system 7 is more particularly intended to operate a fuel injection of separately triggering each injection 5 so that the fuel injection is completed before opening the corresponding intake valve or valves.

Among the input signals of the microprocessor include those addressed by a crankshaft sensor 22. This sensor 22, of the type for example magneto-reluctant, is fixedly mounted on the motor housing to be positioned in front of a measuring ring 12 integral with the steering wheel. inertia attached to one end of the crankshaft. This ring 12 is provided at its periphery with a succession of identical teeth and troughs with the exception of a tooth which has been removed so as to define an absolute reference point making it possible to deduce the moment of passage to the Top Dead Center. a given cylinder of reference, in this case the cylinder C1.

The sensor 22 delivers a signal Dn corresponding to the running of the teeth of the ring gear 12, which signal after treatment by a processing device 10 makes it possible to generate a PMH signal every 120 ° of crankshaft rotation making it possible to locate the passages at Top Dead Center alternately. cylinders C1 (reference 0 °) then C2 (reference 120 °) and finally C3 (reference 240 °) if the combustion order of the engine is C1-C3-C2 as in this example.

It should be noted that for this type of four-stroke engine and three cylinders, and more generally for all four-stroke engines and an odd number of cylinders, the cylinders, here C1, C2 and C3, pass to the position Dead spot High in distinct angular positions.

The device 10 for processing the signal Dn emitted by the sensor 22 also makes it possible to measure the running time of the teeth of the ring 12, and thus to obtain the running speed and the instantaneous rotation speed N of the motors.

The signal Dn is further processed by the device 10 to produce a signal (Cg, Bn) which is a representation of a magnitude representative of the kinematics of the crankshaft. For example, this quantity can define a representation of the estimated gas torque generated by each of the combustions.

The value of the signal Cg for each combustion of the gas mixture in the engine cylinders is obtained in particular from the analysis of the signal Dn delivered by the fixed sensor 22 observing the toothed wheel 12 integral with the crankshaft.

The signal is not used directly, for example by taking the instantaneous speed of rotation, because the presence of measurement noise or the lack of realization of the teeth would generate significant errors due to inaccuracies of the signal and make the process less robust . Thus a harmonic decomposition analysis makes it possible to eliminate these defects.

dans laquelle:

• [Cgaz,0]u est le couple gaz moyen produit par au moins une combustion dans un cylindre u au cours d'un cycle de combustion,
• β_k,i est une fonction de Δt_k et/ou de ω_k respectivement durée et vitesse de passage du motif Dk en face du capteur,
• α _k,i est un coefficient de pondération de la durée associée au motif Dk, dépendant au moins d'un paramètre dé fonctionnement du moteur,
• α _0,i est une variable dépendant au moins d'un paramètre de fonctionnement du moteur,
• δ_i est un coefficient de pondération,
• i est un indice qui comptabilise les combinaisons linèaires de fonctions,
• qu et ru désignent respectivement le numéro du premier motif et le numéro du dernier motif perçus par le capteur de position au cours de la combustion du cylindre u, ou du dernier motif virtuel élaboré à partir du signal du capteur, définissant la fenêtre angulaire d'analyse du couple du moteur associé à la combustion du cylindre u.

A method for producing such a Cg signal is described in particular in the patents EP0532420 or WO9829718 in which the signal Cg is a representation of the gas torque. In general, the estimation of the couple medium gas produced by at least one combustion in the cylinder "u" of the engine comprising p cylinders is given by a relation of the type: $${\left[{\mathrm{VS}}_{\mathit{gas},0}\right]}_u={\displaystyle \underset{i}{\Sigma }}{\delta }_i\left[{\displaystyle \underset{k=q_u}{\overset{r_u}{\Sigma }}}{\alpha }_{\mathit{k},\mathit{i}}{\beta }_{\mathrm{k},\mathrm{i}}+{\alpha }_{0,i}\right]$$

in which:

• [Cgaz, 0] u is the average gas torque produced by at least one combustion in a cylinder u during a combustion cycle,
• β _{k, i} is a function of Δt _k and / or ω _k respectively duration and speed of passage of the pattern Dk in front of the sensor,
• α _{k, i} is a weighting coefficient of the duration associated with the pattern Dk, depending at least on an operating parameter of the motor,
• α _{0, i} is a variable depending at least on an engine operating parameter,
• δ _i is a weighting coefficient,
• i is an index that counts the linear combinations of functions,
• that ru and denote respectively the number of the first pattern and the number of the last pattern perceived by the position sensor during the combustion of the cylinder u, or the last virtual pattern developed from the sensor signal, defining the angular window of analysis of the engine torque associated with the combustion of the cylinder u.

By applying specific values to certain coefficients of the relation of Cg above, other quantities representative of the kinematics of the crankshaft can be defined as harmonic components representative of the speed or the duration of travel of the teeth of the crown. The tooth duration is the duration measured between two teeth of the target.

For example, using a representation of the estimated average gas torque, for a four-stroke engine and three cylinders C1, C2 and C3, the combustion time of the cylinder C1 is between 0 ° and 480 °, the torque estimate is will do by observing the speed of rotation between 0 and 240 ° or a angular window generally encompassing 0-180 ° of the combustion phase of the cylinder C1.

On the same principle, the combustion time associated with the cylinder C3 is between 240 ° and 420 °, the observation will be made on the angular range between 240 ° and 480 °.

For the cylinder C2, the combustion time is between 480 ° and 660 °, the observation range of the engine speed will be around 480 and 720 °.

The principle of the method of generating the synchronization signal is then as follows.

The location of the predetermined moment in the unfolding of the engine cycle for phasing the injection of each of the cylinders, which in the example illustrated is the passage to the High Admission Dead Point, or any other time that can serve as a reference, is operated from a synchronization signal NOCYL synchronized with the signal PMH providing the identification of the passage to the top dead center of each cylinder in a circuit 224,

Several types of NOCYL synchronization signal can be used which alone or in association with a signal from a counter of the number of High Dead Point cylinders passing in front of the position sensor to determine the phase of the combustion cycle for each cylinder.

In this example, the NOCYL signal represented at figure 3 , does not require comparison with other signals and it provides all the predetermined moments of the stints in the course of the engine cycle for phasing the injection or ignition of each of the cylinders for all the cylinders of the motor.

Indeed, the signal NOCYL provides the passage to the High Admission Dead Point and the passage to the Top Dead Center of all your cylinders at the time of the change of value. A single signal is then sufficient to synchronize all the actuators of the motor control.

The PMH signal indicates each passage to the Top Dead Center of the engine cylinders by generating a leading or falling edge. The signal NOCYL is arbitrarily initialized to 0 at the first detection of the transition to the Top Dead Center of the reference cylinder (C1 in this example) which is therefore considered arbitrarily as a High Admission Dead Point, then it is incremented. The NOCYL signal is constructed by incrementing a modulo 6 counter each time the High Dead Point of the PMH signal is passed.

Thus, when the signal NOCYL goes to the value 0 or 3, it means that the PMH of the cylinder C1 has just been detected respectively during the admission or expansion phase.

When the signal NOCYL goes to the wager 1 or 4, it means that the PMH of the cylinder C2 has just been detected respectively in phase of admission or relaxation.

When the signal NOCYL goes to the value 2 or 5, it means that the PMH cylinder C3 has just been detected respectively in the admission phase or relaxation.

Whatever the embodiment of the NOCYL signal, the latter provides a unique reference for all motor cycles that allows a phasing system 222 to synchronize any motor control process (ignition, injection, control actuators ...) .

Given the arbitrary choice made during the initialization of the NOCYL signal, two cases occur: either the NOCYL signal is well phased, the Top Dead Point reference used to initialize the signal actually corresponding to a High Admission Dead Point for the reference cylinder C1, the NOCYL signal is poorly phased, the High Reference Dead Point then corresponding to a High Dead Point for the reference cylinder C1.

In a first embodiment of the invention, when the engine is in the starting phase for example, the estimation of the torque Cg by the torque estimator described above makes it possible to know if the synchronization is the right one. Indeed, if the injection and ignition are poorly phased, the engine can not produce torque because combustion is done during admission. A processing unit 223 compares the estimated value Cg with a reference or setpoint value that is conventionally developed by the engine control during the start-up phase to estimate whether the phasing is correct. The condition to check is then: $$\left|\mathit{Cg}\right|\ge \mathit{CC}-\xi$$

ξ is a positive torque value that can be constant mapped according to the engine control parameters to ensure the robustness of the criterion E1 by limiting the consideration of signal noise.

If this condition is not satisfied, the phasing is not correct, the strategy then considers that the Top Dead Point initially detected was the High Dead Point, the NOCYL signal is then reset in the circuit 224 by an Init signal of C3 signal processing circuit 223. The phasing is now correct, the condition E1 must then be verified. If this is not the case, there is a failure in the injection or ignition system or at the engine level.

However, this process is not necessarily compatible with a so-called "lost spark" strategy generally used when starting the engine, which consists of switching on the cylinders at each engine revolution (High Dead Point Admission and Relaxation) during the start-up phase of the engine. a gasoline engine, unlike the sequential ignition once per thermodynamic cycle, to be sure that the combustion will be done and thus avoid the risk of injecting fuel that will not be burned and ensure at the same time a quick start. Thus, as long as the recognition of the thermodynamic cycle is not effective, the ignition of the engine is controlled in "lost spark" mode before restoring a sequential ignition

A second embodiment of the invention makes it possible to operate in the case of a startup with lost spark.

In this embodiment, as long as the synchronization is not established, the ignition of the engine is controlled in "lost spark" mode. previously described to ensure the starting and operation of the engine including when the phasing engine is not identified.

The torque estimation is done by observing the acyclic speed or instantaneous rotation times of the crankshaft of the engine over an angular range directly related to the supposed phasing of the engine which theoretically covers the combustion phases of the three cylinders.

In this example, the combustion time of the cylinder C1 is between 0 ° and 180 °, the estimation of the torque will be done by observing the speed of rotation between. 0 and 240 ° or an angular window generally encompassing 0-180 ° of the combustion phase of the cylinder C1.

On the same principle, the combustion time associated with cylinder C3 is between 240 ° and 420 °, the observation of acyclism will be on the angular range between 240 ° and 480 °.

For the cylinder C2, the combustion time is between 480 ° and 660 °, the observation range of the engine speed will be around 480 and 720 °.

If the phasing of the combustion sequences is not identified, the observation of the estimated torque Cg of the combustion of the cylinder C1 will be one revolution later, ie between 360 ° and 600 ° instead of the range 0-240 °.

Therefore, it is no longer the combustion of the cylinder C1 which is observed but the end of the combustion of the cylinder C3 and the beginning of the combustion of the cylinder C2, and the estimated torque Cg in this example for these combustion moments of the cylinder C2 and C3 is negative.

Thus, when the synchronization is not correct, the value of the estimated torque Cg is here negative instead of positive. Therefore , if Cg ≥ 0 (E2) synchronization is good, on the contrary if Cg ≤ 0 (E3) the synchronization is bad and the NOCYL signal is then reset as in the first embodiment.

Variants of the second embodiment can also be envisaged.

A first variant of the second embodiment consists in estimating the gas torque Cg for all the engine revolutions. The torque of cylinder C1 thus estimated in the first round Cg1_1 is recorded and compared with a new observation of the pair Cg1_2 of the cylinder C1 in the next turn.

• Cg1_1 > Cg1_2 (E4) si le tour 1 correspond au cylindre C1 en détente et le tour 2 à l'admission.
• Cg1_1 < Cg1_2 (E5) si le tour 2 correspond au cylindre C1 en détente et le tour 1 à l'admission.

The comparison of the pairs Cg1_1 and C1_2 makes it possible to determine the good synchronization according to the following property:

• Cg1_1> Cg1_2 (E4) if turn 1 corresponds to cylinder C1 in expansion and turn 2 to intake.
• Cg1_1 <Cg1_2 (E5) if turn 2 corresponds to cylinder C1 in expansion and turn 1 to intake.

In each of these two cases, the synchronization can thus be performed according to the result of the comparison.

• Cg > Cc - Delta(Régime, Cc) (E6) si la synchronisation est correcte
• Cg < Cc - Delta(Régime, Cc) (E7) si la synchronisation n'est pas correcte

A second variant consists in comparing the value of the estimated torque Cg for a given cylinder with respect to a set torque value Cc according to the following relationship:

• Cg> Cc - Delta (Speed, Cc) (E6) if the synchronization is correct
• Cg <Cc - Delta (Speed, Cc) (E7) if the synchronization is not correct

Indeed, if the synchronization is not correct, the signal NOCYL does not correspond to the thermodynamic cycle of each cylinder, the value of the estimated torque Cg is then significantly less than the torque value of Setpoint Cc and vice versa.

The Delta offset is a torque value that can be a constant or derived from a map dependent on the Engine Speed and / or Torque, and which makes it possible to freeze a threshold necessary for the comparison in order to exclude any risk of false synchronization due to noise in the signals,

The basic methods and variants of the first and second embodiments can be reliable by limiting errors due to disturbances or noise of the signals for example by summing the torque estimates Cg.

The relation (E1) of the first embodiment then becomes: $${\displaystyle \underset{1}{\overset{\mathit{NbreCycles}}{<figure-callout\; id="1"\; label="\Sigma "\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">\Sigma </figure-callout>}}}\mathit{Cg}\ge {\displaystyle \underset{1}{\overset{\mathit{NbreCycles}}{<figure-callout\; id="1"\; label="\Sigma "\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">\Sigma </figure-callout>}}}\left(\mathit{CC}-\xi \right)$$

si la synchronisation est correcte $${\displaystyle \sum _1^{\mathit{NbreCycles}}}\mathit{Cg}\le 0$$

si la synchronisation est incorrecteThe relations (E2) and (E3) of the basic method of the second embodiment become: $${\displaystyle \underset{1}{\overset{\mathit{NbreCycles}}{<figure-callout\; id="1"\; label="\Sigma "\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">\Sigma </figure-callout>}}}\mathit{Cg}\ge 0$$

if the synchronization is correct $${\displaystyle \underset{}{\overset{}{\Sigma }}}$$$${\displaystyle \underset{1}{\overset{\mathit{NbreCycles}}{}}}\mathit{Cg}\le 0$$

if the synchronization is incorrect

$${\displaystyle \sum _1^{\mathit{NbreCycles}}}\mathit{Cg}1\_1\le {\displaystyle \sum _1^{\mathit{NbreCycles}}}\left(\mathit{Cg}1\_2\right)$$

The relations (E4) and (E5) of the first variant of the second embodiment of the invention become: $${\displaystyle \underset{1}{\overset{\mathit{NbreCycles}}{<figure-callout\; id="1"\; label="\Sigma "\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">\Sigma </figure-callout>}}}\mathit{Cg}1\_1\ge {\displaystyle \underset{1}{\overset{\mathit{NbreCycles}}{<figure-callout\; id="1"\; label="\Sigma "\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">\Sigma </figure-callout>}}}\left(\mathit{Cg}1\_2\right)$$

$${\displaystyle \underset{1}{\overset{\mathit{NbreCycles}}{<figure-callout\; id="1"\; label="\Sigma "\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">\Sigma </figure-callout>}}}\mathit{Cg}1\_1\le {\displaystyle \underset{1}{\overset{\mathit{NbreCycles}}{<figure-callout\; id="1"\; label="\Sigma "\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">\Sigma </figure-callout>}}}\left(\mathit{Cg}1\_2\right)$$

$${\displaystyle \sum _1^{\mathit{NbreCycles}}}\mathit{Cg}\le {\displaystyle \sum _1^{\mathit{NbreCycles}}}\left(\mathit{Cc}-\mathit{Delta}\left(N,\mathit{Cc}\right)\right)$$

The relations (E6) and (E7) of the second variant of the second embodiment of the invention become: $${\displaystyle \underset{1}{\overset{\mathit{NbreCycles}}{<figure-callout\; id="1"\; label="\Sigma "\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">\Sigma </figure-callout>}}}\mathit{Cg}\ge {\displaystyle \underset{1}{\overset{\mathit{NbreCycles}}{<figure-callout\; id="1"\; label="\Sigma "\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">\Sigma </figure-callout>}}}\left(\mathit{CC}-\mathit{Delta}\left(\mathrm{<figure-callout\; id="1"\; label="NOT\; CC\; \Sigma "\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">NOT\; </figure-callout>},\mathit{CC}\right)\right)$$

$${\displaystyle \underset{}{\overset{}{\Sigma }}}$$$${\displaystyle \underset{1}{\overset{\mathit{NbreCycles}}{}}}\mathit{Cg}\le {\displaystyle \underset{1}{\overset{\mathit{NbreCycles}}{<figure-callout\; id="1"\; label="\Sigma "\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">\Sigma </figure-callout>}}}\left(\mathit{CC}-\mathit{Delta}\left(\mathrm{NOT},\mathit{CC}\right)\right)$$

avec 0 < α < 1For each of the two embodiments the error limitation can also be carried out by filtering, for example carried out according to a filtering of the first or second order or any other filter making it possible to filter the noise of the measurements and estimates and thus rendering the result of the comparisons more robust. As an example we will give a filter of the first discrete order F defined by: $${\mathrm{F}}_{\mathrm{not}}\left(X_B\right)=\alpha {\mathrm{X}}_{\mathrm{not}}+\left(1\_\alpha \right){\mathrm{F}}_{\mathrm{not}-\mathrm{1}}$$

with 0 <α <1

The relation (E1) of the first embodiment then becomes: $$F\left(\mathit{Cg}\right)\ge F\left(\mathit{CC}-\xi \right)$$

$$F\left(\mathit{Cg}\right)\le 0$$

The relations (E2) and (E3) of the basic method of the second embodiment become: $$F\left(\mathit{Cg}\right)\ge 0$$

$$F\left(\mathit{Cg}\right)\le 0$$

$$F\left(\mathit{Cg}1\_1\right)\le F\left(\mathit{Cg}1\_2\right)$$

The relations (E4) and (E5) of the first variant of the second embodiment of the invention become: $$F\left(\mathit{Cg}1\_1\right)\ge F\left(\mathit{Cg}1\_2\right)$$

$$F\left(\mathit{Cg}1\_1\right)\le F\left(\mathit{Cg}1\_2\right)$$

$$F\left(\mathit{Cg}\right)\le F\left(\mathit{Cc}-\mathit{Delta}\left(N,\mathit{Cc}\right)\right)$$

The relations (E6) and (E7) of the second variant of the second embodiment of the invention become: $$F\left(\mathit{Cg}\right)\ge F\left(\mathit{CC}-\mathit{Delta}\left(\mathrm{NOT},\mathit{CC}\right)\right)$$

$$F\left(\mathit{Cg}\right)\le F\left(\mathit{CC}-\mathit{Delta}\left(\mathrm{NOT},\mathit{CC}\right)\right)$$

As previously described a representation of a magnitude representative of the kinematics of the crankshaft other than the flexible can be used. Thus a third embodiment of the invention uses a harmonic analysis representative of either the tooth duration or the instantaneous rotation speed of the crankshaft.

Thus, this embodiment consists in studying the harmonic component of order n of the rotational speed or preferably of the tooth duration established from the signal Dn. The harmonic component of Bn, will here be calculated using the cosine harmonic function, but the method can be adapted for any other harmonic function for example using a trapezoid function or other more complex function. This component Bn is a simplified representation based on the relationship of the estimated torque (Cg) above, taking appropriate coefficients.

For this example, the harmonic component of the tooth duration can be established by the following relation: $${\mathrm{B}}_{\mathrm{not}}={\displaystyle \underset{i=0}{\overset{\mathrm{not}-1}{\Sigma }}}{\mathrm{<figure-callout\; id="2"\; label="d\; i\; cos"\; filenames="imgf0002.png"\; state="\{\{state\}\}">d\; </figure-callout>}}_i\cos \left(\frac{2i\pi }{\mathrm{not}}\right)$$

entre les amplitudes harmoniques calculées pour les deux phasages possibles du cylindre étudié, calculés sur le Point Mort Haut supposé combustion, et sur le Point Mort Haut supposé échappement, permet d'établir trois cas.The calculation of the difference ${\mathit{B}}_{\mathit{not}}^{\mathit{Handset}}-{\mathit{B}}_{\mathit{not}}^{\mathit{ech}}$

between the harmonic amplitudes calculated for the two possible phasages of the cylinder studied, calculated on the Top Dead Center assumed combustion, and on the Top Dead Center assumed exhaust, allows to establish three cases.

est supérieur à une valeur maximale, alors le moteur est bien phasé ;Yes ${\mathit{B}}_{\mathit{not}}^{\mathit{Handset}}-{\mathit{B}}_{\mathit{not}}^{\mathit{ech}}$

is greater than a maximum value, then the motor is well phased;

est inférieur à une valeur minimale, alors le moteur est mal phasé ;Yes ${\mathit{B}}_{\mathit{not}}^{\mathit{Handset}}-{\mathit{B}}_{\mathit{not}}^{\mathit{ech}}$

is less than a minimum value, then the motor is poorly phased;

est compris entre ces deux seuils. Il y a indétermination, et le calcul doit être à nouveau effectué.Yes ${\mathit{B}}_{\mathit{not}}^{\mathit{Handset}}-{\mathit{B}}_{\mathit{not}}^{\mathit{ech}}$

is between these two thresholds. There is indeterminacy, and the calculation must be done again.

In each of these cases, the synchronization can thus be performed according to the result of the comparison as in the previous embodiments

This embodiment is robust to target defects, since it compares two quantities potentially biased in the same way by the target defect, the calculations of Bn on the two PMH of the cylinder originating from durations measured on the same portions. angular of the target of the crown. Indeed the harmonic component established in the first round will be decomposed into a sum of the harmonic component representative of the thermodynamic cycle during the first round and a harmonic component representative of the target defects. The harmonic component established in the second round will be decomposed into a sum of the harmonic component representative of the thermodynamic cycle during the second round and a harmonic component representative of the target defects that is the same as in the first round. Therefore, a comparison of the harmonic component established in the first round and that established in the second round will eliminate the representative component of the target defects.

Regardless of the detection mode, in the event of a bad synchronization, the NOCYL signal is reset by changing the synchronization hypothesis (offset of one revolution of the synchronization). This reset can be done on the PMH of the reference cylinder or on any PMH of any cylinder. The synchronization must then be confirmed again by a method according to one of the previously described embodiments before establishing the normal operation of the engine with sequential ignition.

The invention advantageously makes it possible to synchronize the thermodynamic cycle of each cylinder without modifying operating parameters of the engine and without altering the operation of the engine.

Claims (4)

1. Method for producing a synchronization signal (NOCYL) for a four-stroke internal combustion engine with an odd number of cylinders (C1, C2, C3) via an electronic control system (7), the synchronization signal (NOCYL) used to identify a predetermined instant in the thermodynamic cycle of each of the cylinders of the engine being determined from a TDC signal identifying a determined position of each cylinder, and a signal (Cg, Bn) representative of a quantity representative of the kinematics of the crankshaft generated by each of the combustions, both generated from information from an engine crankshaft position sensor (22), characterized in that it comprises the following steps:

   - operation of the engine for a given period with ignition in the cylinders on each revolution of the cylinders, so as to produce a systematic combustion of the injected fuel,

   - calculation of the representative signal (Cg, Bn)

   - comparison of the representative signal (Cg, Bn) of a cylinder in a first revolution of the cycle (Cg1_1, B_n ^Comb ) with the representative signal of the cylinder in a second revolution of the cycle (Cg1_2, B_n ^ech ), so as to ascertain the phase of the first revolution,

   - resetting of the synchronization signal (NOCYL) if the analysis of the comparison indicates that the phasing of the synchronization signal was incorrect.
2. Method for producing a synchronization signal (NOCYL) according to Claim 1, characterized in that the representative signal (Cg) is a representation of the gas torque.
3. Method for producing a synchronization signal (NOCYL) according to Claim 2, characterized in that the estimation of the gas torque produced by at least one combustion in the cylinder "u" of the engine comprising p cylinders is given by a relation of the type: $${\left[C_{\mathrm{gas},0}\right]}_u={\displaystyle \sum _i}{\delta }_i\left[{\displaystyle \sum _{k=q_u}^{r_u}}{\alpha }_{\mathrm{k},\mathrm{i}}{\beta }_{\mathrm{k},\mathrm{i}}+{\alpha }_{0,i}\right]$$

   

   
   in which:

   [Cgas,0)u is the average gas torque produced by at least one combustion in a cylinder u during a combustion cycle,

   β _{k, i} is a function of Δl_k and/or of ω_k, respectively the duration and speed of passage of the pattern Dk across the face of the sensor,

   α_k,i is a weighting coefficient for the duration associated with the pattern Dk, at least dependent on an engine operating parameter,

   α_{O, i} is a variant dependent at least on an engine operating parameter,

   δ_i is a weighting coefficient,

   i is an index which counts the linear combinations of functions,

   qu and ru respectively designate the number of the first pattern and the number of the last pattern observed by the position sensor during the combustion in the cylinder u, or of the last virtual pattern generated from the signal from the sensor, defining the angular window for analyzing the torque of the engine associated with the combustion in the cylinder u.
4. Method for producing a synchronization signal (NOCYL) according to Claim 1, characterized in that the representative signal (Bn) is a harmonic representation of the tooth duration.

Images