FR2877995A1 - METHOD FOR MANAGING AN INTERNAL COMBUSTION ENGINE - Google Patents

Abstract

Method for managing an internal combustion engine (100) according to which a position of the shaft (1) of the internal combustion engine (100) is determined using a sensor (120) and the signal is calibrated supplied by the sensor (120). A first phase relationship is determined between a preferred position of the shaft (1) and a signal characterizing the evolution of combustion in the internal combustion engine (100), a second phase relationship is determined between a position of the signal sensor (120) giving the preferred position of the shaft (1) and the signal characterizing the evolution of the combustion, and the position of the shaft (1) supplied by the sensor (120) is corrected as a function of the comparison, in particular of the difference, between the first phase relation and the second phase relation.

Description

Field of the invention

  The present invention relates to a method for managing an internal combustion engine in which a position of the internal combustion engine shaft is determined by means of a sensor and the signal supplied by the sensor is calibrated.

  STATE OF THE ART Methods for managing an internal combustion engine are already known, according to which the position of the motor shaft is determined by means of a sensor whose signal is calibrated. For example, document EP 0 661 433 B1 describes, for example, the use of an inductive sensor in an internal combustion engine to detect the teeth of the starter ring gear and to deduce a rotation speed therefrom. It is also planned to use a marking to characterize the top dead center of at least one cylinder. Due to the installation tolerances of the starter ring gear, there is a difference between the position of a reference pulse generated by the mark or marking and the actual top dead center of the crankshaft. The corrective angle which gives the deviation between the reference pulse and the actual top dead center is recorded as the correction quantity. During operation, memory correction values are then taken into account in order to calculate the control signals.

  DESCRIPTION AND ADVANTAGES OF THE INVENTION The present invention relates to a method for managing an internal combustion engine of the type defined above, characterized in that a first phase relation is determined between a preferred position of the shaft and a signal characterizing the evolution of the combustion in the internal combustion engine, a second phase relationship is determined between a position of the sensor signal giving the preferred position of the shaft, and the signal characterizing the evolution of the combustion, and corrects the position of the shaft provided by the sensor according to the comparison, in particular of a difference, between the first phase relation and the second phase relation.

  The method for managing an internal combustion engine according to the invention has the advantage over the state of the art of enabling, with the aid of the signal characterizing the evolution of combustion in the internal combustion engine, to calibrate in a particularly simple and reliable way the position of the sensor signal indicating the preferred position of the shaft without requiring a second shaft of the internal combustion engine and an associated sensor.

  It is particularly advantageous to determine the first phase relation between a predefined position of the signal characterizing the evolution of the combustion and the preferred position of the shaft. This makes it possible to very simply determine the first phase relation and this in a definite manner. The first phase relation can then be used as a reference phase relation if it is obtained as a phase relation with respect to the effective preferential position of the shaft using a perfect sensor that is to say without defects.

  It is further advantageous to choose as a predefined position of the signal characterizing the evolution of the combustion, a position on a sidewall of the signal in the compression phase. This ensures that the first phase relation thus obtained and the second phase relation will be as accurate as possible because the signal characterizing the evolution of the combustion in such a compression phase is in synchronism with the angle of the motor shaft. internal combustion. To determine the angle as precisely as possible, it is particularly interesting to exploit the flank of the signal characterizing the combustion.

  It is also advantageous to determine the second phase relationship between the predefined position of the signal characterizing the evolution of the combustion and the position of the sensor signal indicating the preferred position of the shaft. Thus, the second phase relation makes it possible to use the signal characterizing the evolution of the combustion to determine in a simple and precise manner the respective position of the sensor signal associated with the preferred position of the shaft without having to characterize particularly that position of the sensor signal or the position of a sound wheel associated with the sensor, for example by deleting a tooth or widening or narrowing a tooth if the tone wheel is made in the form of a toothed wheel.

  It is also advantageous to average the signal characterizing the evolution of the combustion over several periods. This improves the accuracy of the calibration.

  Advantageously, the operating cycle of a cylinder of the internal combustion engine is chosen as the period.

  It is also advantageous for the signal characterizing the evolution of the combustion to be filtered in particular by means of a low-pass filter. This eliminates the disturbances of the signal characterizing the evolution of the combustion, which makes it possible to determine more precisely the first phase relation and the second phase relation.

  It is particularly advantageous to use as a signal characterizing the evolution of the combustion, a signal provided by a structural noise sensor and in particular the low frequency component of this signal, or a cylinder pressure or strain gage signal. .

  Drawings The present invention will be described below in more detail with the aid of an exemplary embodiment shown in the accompanying drawings, in which: FIG. 1 is a very simplified schematic view of an internal combustion engine equipped with FIG. 2a shows the curve of an amplitude signal provided by a structural noise sensor as a function of the crankshaft angle; FIG. 2b shows the curve of the amplitude of a signal. generator for crankshaft top dead center versus crank angle, - figure 3 shows a flowchart for determining a first and a second phase relationship, - figure 4 shows a flowchart for correcting the generator signal for the crankshaft angle of the internal combustion engine. DESCRIPTION OF THE EMBODIMENT The method according to the invention will be described below by way of example applied to a diesel engine without the described method being limited to this application; it can also be transposed to other types of internal combustion engine. In this case, the corresponding items must be exchanged. Thus, for example, in the case of a spark-ignition internal combustion engine, the moment of ignition will have a meaning similar to that of the start of the injection in the case of a non-ignition engine. ordered.

  The reference 100 designates an internal combustion engine including a diesel engine. The crankshaft 1 of the internal combustion engine carries the ring gear 110 for the starter. This ring is detected by an inductive sensor 120 which supplies a control apparatus 130, in particular an electronic control unit for a diesel engine and a test apparatus, by supplying signals thereto.

  The internal combustion engine receives fuel dosed by a fuel pump 140 driven by a pump drive shaft carrying an incremental wheel 150. Increments of the incremental wheel 150 are detected by an increment sensor 160. increment 160 then provides a signal to the EDC controller 130.

  The drive shaft of the pump is driven by the camshaft 170 of the internal combustion engine or this camshaft serves as the drive shaft of the pump. The camshaft 170 and the crankshaft are connected by a driving means 180 including a toothed belt or a chain. The drive shaft of the pump can be shifted with respect to the camshaft 170 via an injection adjusting member 190.

  Based on the signals received from different sensors that enter different operating parameters, the EDC controller 130 controls the fuel pump 140 and the injection adjuster 190 by providing them with control signals.

  This installation works mainly as follows: Starting from different sensor signals, the EDC 130 controller calculates a value for the fuel dose to be injected as well as the start of the injection. With these values, the controller 130 determines control pulses to control the injection adjuster 190 setting the start of the injection. The controller 130 also provides control pulses for controlling a regulator that adjusts the dose determining member at the fuel pump 140.

  The inductive sensor 120 detects the teeth of the starter ring gear 110. The control unit 130 thus knows the position of the crankshaft 1. The top dead center of the crankshaft 1 during the compression phase of one of the cylinders of the combustion engine internal 100 is a preferred position of the crankshaft 1. To optimally dose the fuel, it is necessary that the beginning of the injection is made with respect to this top dead center of the crankshaft 1. For this reason, the ring gear 110 of the starter has a mark or marking that the inductive sensor 120 recognizes. The pulse generated by this reference is usually referred to as a reference pulse. Since a cylinder operating cycle extends over a crankshaft angle of 720, i.e. two rotations of the crankshaft, the crankshaft position alone is not sufficient to determine whether the reference pulse generated by the reference of the starter gear ring 110 characterizes the top dead center of the crankshaft 1 in the compression phase of the cylinder or its high dead point in the expulsion phase of the cylinder. As with each work cycle, the camshaft performs only one revolution, the cylinder compression time can be determined by using a phase generator on the camshaft sounding wheel. As the voice wheel of the camshaft, it is possible to use the incremental wheel 150 and the corresponding phase sensor of the incremental sensor 160. Thus, only the reference pulses of the inductive sensor 120 which characterize the top dead center will be determined. crankshaft 1 in the compression phase or during the compression time of the cylinder. These pulses are shown in Figure 2b).

  Because of the installation tolerances, there is a difference between the position of the reference pulse and the top dead center of the crankshaft.

  In order to have a precise fuel dosage, this offset must be taken into account. This is why it is intended to determine correction values, to record them in a memory and to take them into account when calculating the control signals.

  This determination of the correction values is preferably done at the end of the assembly line and / or in certain intervals. This means that the method for determining the correction values is carried out by the control unit 130 and / or by an external installation, at the end of the manufacture of the internal combustion engine of the vehicle or during a repair. , during a maintenance intervention or in certain intervals.

  According to FIG. 1, the internal combustion engine 100 comprises for example four cylinders 5, 10, 15, 20 driving the crankshaft 1. The first cylinder carries the reference 5, the second cylinder, the reference 10, the third cylinder, the reference 15 and the fourth cylinder, the reference 20. As shown in broken line in Figure 1, a structural noise sensor 25 is installed between the second cylinder 10 and the third cylinder 15. This sensor captures structural noise or noise. of the body of the internal combustion engine 100 and provides a measurement signal corresponding to the control apparatus 130. This measurement signal will hereinafter also be referred to as the structure noise signal. It is characteristic of the evolution of the combustion of the internal combustion engine 100. The order of the cylinders 5, 10, 15, 20 is for example the following: first cylinder 5 - third cylinder 15 - fourth cylinder 20 - second cylinder 10. The structure noise signal indicated in Figure 1 by the letter K then changes according to the crankshaft angle KW, for example according to the diagram of Figure 2a).

  Figure 2a) particularly shows the low frequency component of the structure signal. For the internal combustion engine 100 operating in pulled mode, the amplitude of the structure signal K is represented as a function of the crankshaft angle KW. The negative peaks of the signal generated by the pressure phases of the cylinders 5, 10, 15, 20 during their compression time thus appear significantly. The negative peaks of lesser amplitude of the signal correspond to the compression time of the first signal 5 and the fourth cylinder 20; these negative signal peaks alternate according to the crankshaft angle KW. The negative peaks of the high amplitude signal come from the compression phases of the compression time of the second cylinder 10 and the third cylinder 15; these signals also alternate depending on the angle of the crankshaft. To explain this situation, in Figure 2a), each negative signal tip is identified by the cylinder it generates; the first cylinder 5 carries the reference Zyl. 1; the second cylinder 10 has the reference Zyl. 2; the third cylinder 15 has the reference Zyl. 3 and the fourth cylinder 20, the reference Zyl. 4. The negative signal peaks generated by the second cylinder 10 and the third cylinder 15 have a greater amplitude than the signal peaks generated by the first cylinder 5 and the fourth cylinder 20. The reason is that the noise sensor of structure 25 is installed between the second cylinder 10 and the third cylinder 15 and thus the structure noise generated by these two cylinders are more strongly reflected than the structural noise generated by the first cylinder 5 and the fourth cylinder 20. According to the polarity of the structure 25 noise sensor, the signal it provides can also be inverted that is to say have an opposite algebraic sign and in this case must adapt its operation.

  FIG. 2b) represents the signal of the inductive sensor 120 bearing the reference I in FIG. 1. The curve of the amplitude of the signal of the inductive sensor 120 according to FIG. 2b) is also represented as a function of the crankshaft angle KW. Due to the reference carried by the starter gear ring 110 which can be made for example by the removal of a tooth, it is possible to detect from the signal I of the inductive sensor 120, the top dead center of the crankshaft 1 as the preferred position of the crankshaft 1; in the present example, this top dead center corresponds to the top dead center of the piston of the first cylinder 5 in the compression time. FIG. 2b) only shows the reference pulses described above of the signal supplied by the inductive sensor 120.

  As it appears, these reference pulses occur for the same crank angles as those which correspond to FIG. 2a) to a negative tip of the structure noise signal K of the first cylinder 5. Thus, FIGS. 2a) and FIG. 2b) show that the phase between the peaks of the structure noise signal K and the signal I of the inductive sensor 120 is constant; in the example described, this phase is equal to zero because the reference pulses of the signal I of the inductive sensor 120 relative to the crank angle coincide with the negative peaks of the structure noise signal K of the first cylinder 5. But very generally, there can be any phase shift between the signal I provided by the inductive sensor 120 and the structure noise signal K; this phase shift will always be constant. In a very general manner, the reference pulses of the signal I of the inductive sensor 120 are always connected in a fixed relation, that is to say a constant phase shift, to the structure noise signal K. Starting from the structure noise signal K it is thus proposed to deduce the top dead center of the crankshaft as a function of the phase relation obtained between the structure noise signal K and the top dead center of the crankshaft 1.

  Thus, for example, in the context of an application operation of the internal combustion engine 100, using, for example, the starter gear ring 110, which is customary with the reference mark described for the top dead center of the crankshaft 1, the evolution of the signal of the inductive sensor 120 according to FIG. 2b) will be recorded and this signal will be compared with the structure noise signal K supplied by the structure noise sensor 25 according to FIG. 2a) in the control apparatus 130. For this, a position of the structure noise signal K and the phase relationship between this predefined position of the structure noise signal and the top dead center of the crankshaft 1 according to the reference pulses of FIG. 2b) are fixedly predefined. . Thus, according to the example of FIGS. 2a) and 2b), there will be, as predefined position of the structure noise signal K, a position on a sidewall of the structure noise signal K in the compression phase of the internal combustion engine 100c. that is to say during the compression time of one of the cylinders 5, 10, 15, 20 selected. The advantage is that the structure noise signal K which is a low frequency signal is particular to the compression time, in synchronism with the angle of the crankshaft 1. To determine the angle as accurately as possible for the the operation of the structure noise signal K, it is preferable to use a signal flank. For example, such a signal flank may be selected in the high pressure phase for which there will be about 50% of the amplitude of the structure noise signal K. This amplitude is also referred to as the structural noise signal excursion K FIG. 2a) indicates the predefined position of the structure noise signal K by the letter V. According to FIG. 2a), the predefined position V of the structure noise signal K is each time on the negative side of the signal characterizing a high pressure phase in the compression time of the third cylinder 15; this position is substantially at about 50% of the signal excursion as indicated. FIG. 2b) shows that the phase relationship between the predefined position V of the structure noise signal K and the reference pulse closest to the signal I of the inductive sensor 120 according to FIG. 2b) corresponds to a constant phase difference. A. When the internal combustion engine 100 is operating normally, that is to say, once the adaptation is complete, then it is possible to use a starter ring gear 110 without a reference mark for the top dead center of the crankshaft 1 it is that is, without having to insert a tooth to achieve this mark. The top dead center of the crankshaft 1 is then determined as follows: the predefined position V of the body noise signal K is first determined in the control apparatus 130. Starting from this predefined position V, the difference in phase A obtained by adaptation for the predefined position V which gives the new phase corresponding to the top dead center of the crankshaft 1. In addition, it takes the position of the crankshaft 1 in the usual way using the inductive sensor 120 using the pulses the signal of the inductive sensor 120, not shown in Figure 2b) and corresponding to the other teeth of the ring gear 110 of the starter. The top dead center of the crankshaft 1 is determined by the control apparatus 130 as being the position of the crankshaft 1 gripped by the inductive sensor 120 and which corresponds to the base relation with respect to the structure noise signal K obtained in the period adaptation. This means that the pulse of the signal I supplied by the inductive sensor 120 is considered to be that of the top dead center of the crankshaft 1, the pulse being shifted from the predefined position V of the structure noise signal K of the phase difference A obtained in the adaptation period. Thus, to determine the top dead center of the crankshaft 1, we continue to use the high accuracy of the measurement made by the inductive sensor 120 based on the detection of the teeth of the ring gear 110 of the starter. This ring gear 110 may be made for example in the form of a toothed wheel having 60 teeth with 6 degrees of crank angle per tooth and interval. For the structure noise signal K, then just a precision of + / -1/2 tooth is required, that is to say + / -3 degrees of crankshaft angle KW; in fact, the structure noise signal K is merely intended to guarantee the choice of the correct tooth of the starter tooth ring 110 as a reference point for the top dead center of the crankshaft 1. The accuracy itself of the determination of the top dead center of the crankshaft 1 is thus guaranteed by the accuracy of the installation of the starter ring gear 110 on the crankshaft 1 of the internal combustion engine 100. The structure noise signal K itself can be separated from the parasites by filtering; in particular, a low-pass filter will be used. In this way, the determination of the predefined position V and also that of the phase difference A for each operating cycle will be more precise.

  In addition to the detection of the top dead center of the crankshaft 1, the invention also provides for calibrating the top dead center of the crankshaft 1 obtained with the aid of the signal from the inductive sensor 120 either as described above or so usual by marking or removing a tooth of the ring gear 110 of the starter. According to the invention, a first phase relation is determined between the top dead center of the crankshaft 1 and the structure noise signal K. Then, a second phase relation is determined between the position of the signal I of the inductive sensor 120 indicating the top dead center of the crankshaft 1 and the structural noise signal K. The position of the crankshaft 1 determined by the inductive sensor 120 will be corrected according to the difference between the first phase relation and the second phase relation. In practice, the operation is as follows: the first phase relationship is determined between a predefined position of the structure noise signal K which may correspond to the predefined V position described above and the neutral position high effective crankshaft 1. This can be done for example during the adaptation phase already described of the internal combustion engine 100 using a particular sensor. This means that to determine the top dead center of the crankshaft 1, a perfect starter ring gear 110 and a perfect inductive sensor 120 are used; to locate the top dead center of the crankshaft 1, the starter gear ring 110 can as already described be marked by the removal of a tooth. Thus, the reference pulses of the signal delivered by the inductive sensor represent the actual position of the top dead center of the crankshaft 1. Then, as the first phase relation, a first phase difference A 1 is determined between the actual top dead center of the crankshaft. 1 and the predefined position, close V of the structural noise signal K. During a calibration phase which can be done on the one hand also during the adaptation of the internal combustion engine 100 and on the other hand also during the subsequent normal operation of the internal combustion engine 100, determining the second phase relationship between the predefined position V of the structure noise signal K and the position of the signal I indicating the top dead center of the crankshaft 1 of the inductive sensor 120 actually used. for normal operation of the internal combustion engine 100; this inductive sensor no longer works in a perfectly correct manner without defects for example because of manufacturing tolerances or aging. The same applies to the starter ring gear 110 used for the normal operation of the internal combustion engine 100. Thus, by using the same predefined position V of the structure noise signal K also for the starter ring gear 110 used in normal operation and the inductive sensor 120 used in normal operation of the internal combustion engine 100, is determined as a second phase relationship, a second phase difference A2 between the predefined position V of the structure noise signal K and the position the signal provided by the inductive sensor 120 used then and which gives the top dead center of the crankshaft 1. The difference between the first phase difference A 1 and the second phase difference A2 then gives a correction value for the phase relationship between the predefined position V of the structure noise signal and the position of the signal indicating the top dead center of the crankshaft 1 ie by the inductive sensor 120 used for the normal operation of the internal combustion engine 100. With this corrective value, the respective position of the crankshaft 1 supplied by the inductive sensor 120 is corrected.

  To improve the calibration method described above, it is advantageous to install the structural noise sensor 25 on a cylinder of the internal combustion engine 100 which is as close as possible to the sound wheel; in this example, it will be installed as close as possible to the starter ring gear 110. The structure noise signal K is even better correlated with the signal I of the inductive sensor 120. To improve the accuracy, it is possible to determine the Structure noise signal K during several operating cycles. In a very general manner, the structure noise signal K can also be determined for this purpose during several periods of the structure noise signal K. A period of the structure noise signal K corresponds to a cycle of operation of a cylinder of the internal combustion engine 100. As already described above, it is also possible to improve the accuracy and eliminate the disturbances of the structure noise signal K for calibration, in particular by filtering with a low-pass filter or by a bandpass filter. For better reproducibility, the calibration method described above can be limited to certain predefined operating points or operating conditions of the internal combustion engine 100.

  FIG. 3 shows a flow chart describing, for example, the determination of the first phase relation. For this, the particular sensor used is a perfect sensor that is to say, a ring gear starter 110 faultless and a perfect inductive sensor that is to say, an inductive sensor 120 without defects. The program executes during an adaptation of the internal combustion engine 100. After the start of the program, at the program point 200, as described, the control apparatus 130 averages the structure noise signal K over several cycles of functioning as for example ten cycles of operation. Then we go to program point 205.

  At program point 205, from the average structure noise signal, control 130 determines the predefined position V as has been described. Then, we go to program point 210. At program point 210, command 130 determines in the manner described

  using the signal provided by the perfect inductive sensor 120, the actual angle of the crankshaft for the top dead center of the crankshaft 1. For this, the starter gear can include the indicated marking. Then we go to program point 215.

  At program point 215, the control 130 forms, as described, the first phase difference Δ 1 between the predefined position V of the structure noise signal K and the actual position of the top dead center of the crankshaft 1. we leave the program.

  The flowchart of Fig. 3 may be correspondingly executed also to determine the second phase relation both during adaptation of the internal combustion engine 100 and during the next normal operation of the internal combustion engine 100. In this case no longer uses the perfect starter ring gear or the perfect inductive sensor, but a series specimen marred by, for example, manufacturing tolerances and aging increasing with the running time of the internal combustion engine 100. the position of the signal of the inductive sensor 120 which corresponds in a similar way to the program point 210 of FIG. 3 for the top dead center of the crankshaft 1 does not necessarily correspond or in general not to the actual position of the top dead center of the crankshaft 1. In a program step similar to point 215 of the program, the second phase difference A2 is determined as it was described. For the remainder, the second phase relation is determined as described for the flow-ment of the flowchart of FIG. 3.

  FIG. 4 shows a flowchart traversed during the adaptation of the internal combustion engine 100, in particular at the end of the adaptation in normal operation of the internal combustion engine 100 by using the starter gear 110 which will then no longer be in general without fault and the inductive sensor 120 which will not be generally flawless either. The flowchart of FIG. 4 thus represents the actual angular correction operation for determining the position of the crankshaft 1 from the signal of the inductive sensor 120. After the start of the program, the program point 300 the control 130 determines the current current angle position of the crankshaft 1 with the aid of the signal I of the inductive sensor 120. Then, the program point 305 is passed.

  At program point 305, control 130 corrects the current determined angle of crankshaft 1 by the difference (A1 - A2). This difference corresponds to the phase difference between the actual top dead center of the crankshaft 1 and the top dead center of the crankshaft 1 obtained by the inductive sensor 120 which is no longer a perfect sensor. It should be noted that at program point 305, the correction is made with the algebraic sign. This is then the case if the phase difference (Al-A2) is added to the position of the crankshaft 1 supplied by the inductive sensor 120 and which corresponds to the crankshaft angle 1 measured. Then we leave the program.

  Instead of the structural noise signal, it is also possible to use a cylinder pressure signal as a signal characterizing the evolution of the combustion of the internal combustion engine. This cylinder pressure signal can be determined by means of one or more cylinder pressure sensors installed in one or more of the cylinders of the internal combustion engine 100. Thus, it is possible to exploit the cylinder pressure signal. chronological evolution of the pressure signal of the first cylinder 5 analogously to the structure noise signal as described by the control apparatus 130 to determine the phase relationship with respect to the preferred position of the crankshaft 1 which is then for example again the top dead center of the crankshaft 1. Correspondingly, from the cylinder pressure signal can be deduced the preferred position of the crankshaft 1. The cylinder pressure signal has a maximum at the top dead center Crankshaft 1. It is thus possible to select a corresponding predefined position of the cylinder pressure signal by referring to a plane of the cylinder pressure signal with a synchronizer. angular ism.

  The signal of the inductive sensor 120 is not considered here as a signal characterizing the evolution of the combustion in the internal combustion engine 100 because unlike the structure noise signal U or the cylinder pressure signal, it does not allow to subdivide the different times or phases of combustion. A signal characterizing the evolution of the combustion of the internal combustion engine 100 allows on the other hand such subdivision at least the recognition of a practical combustion phase. Thus, the structure noise signal and the cylinder pressure signal at least make it possible to recognize unequivocally a compression phase of a cylinder opposite the signal supplied by the inductive sensor 120.

  As a signal characterizing the evolution of the combustion, it is also possible to use also the signal supplied by a strain gauge installed so as to detect at least one of the cylinders of the internal combustion engine 100 by the length variations caused by the compression. and expansion.

Claims (9)

  1) A method of managing an internal combustion engine (100) in which a position of the shaft (1) of the internal combustion engine (100) is determined by means of a sensor (120) and is calibrated the signal supplied by the sensor (120), characterized in that a first phase relation is determined between a preferred position of the shaft (1) and a signal characterizing the evolution of the combustion in the internal combustion engine ( 100), a second phase relationship is determined between a position of the sensor signal (120) giving the preferred position of the shaft (1), and the signal characterizing the evolution of the combustion, and the position of the sensor is corrected. (1) provided by the sensor (120) as a function of the comparison, in particular of a difference, between the first phase relation and the second phase relation.   2) Method according to claim 1, characterized in that it determines the first phase relationship between a predefined position of the signal characterizing the evolution of the combustion and the preferred position of the shaft (1).   3) Process according to claim 2, characterized in that as a predefined position of the signal characterizing the evolution of the combustion, a position is selected on a signal flank in the compression phase.   4) A method according to claim 3, characterized in that as predefined position of the signal characterizing the evolution of the combustion, a position is chosen on the side of the signal which corresponds substantially to half the excursion of the signal in the phase of corresponding compression.   5) Method according to any one of claims 2 to 4, characterized in that one determines the second phase relation between the predefined position of the signal characterizing the evolution of the combustion and the position of the sensor signal (120) indicating the preferred position of the shaft (1).   6) Process according to claim 1, characterized in that the signal characterizing the evolution of the combustion is averaged over several periods.   7) A method according to claim 6, characterized in that as a period is chosen an operating cycle of a cylinder (5, 10, 15, 20) of the internal combustion engine (100).   8) Process according to claim 1, characterized in that the signal characterizing the evolution of combustion is filtered, in particular by means of a low-pass filter.   9) Process according to Claim 1, characterized in that a signal of a structural noise sensor, in particular its low frequency component or a cylinder pressure signal or a signal, is selected as signal characterizing the evolution of the combustion. oven-ni by a strain gauge.