JP3392152B2 - Method and apparatus for determining rotational fluctuation values of an internal combustion engine - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for determining a rotational fluctuation value (rotational unevenness value) of an internal combustion engine. The rotation fluctuation value is used to determine whether the internal combustion engine has a misfire.

In the following, the concept of "ignition stroke" is often used. This is a four-stroke engine stroke in which the compressed air / fuel mixture is essentially burned after ignition. Therefore, the concept of "ignition stroke" generally refers to the above stroke, regardless of whether ignition is actually performed or whether there is a misfire.

Similarly, the concept of "top dead center before the ignition stroke of the cylinder" is often used. This is the top dead center of the cylinder piston which the piston occupies before the ignition stroke.

[0004]

2. Description of the Related Art In the prior art, two different principles for determining the rotational fluctuation value of an internal combustion engine have been known for more than 10 years. According to the first principle, the time intervals of two crankshaft angular sectors per ignition stroke are measured and subtracted from each other, whereas according to the second principle one time interval of each such ignition stroke is present. Only that time is measured and that time interval is subtracted from the time intervals that belong to another ignition stroke.

In the following, the concept of linear acceleration or deceleration is often used. This is as follows. In other words, when the time interval of the crankshaft angular sector is illustrated with respect to time, a linear characteristic with a negative slope is obtained due to the linear acceleration, and a linear characteristic with a positive slope is obtained due to the linear deceleration. can get.

The method shown in the first principle is, for example, DE-A.
-2507138 (US-A-4044234). Two crankshaft angular sectors are defined for each cylinder, one located before top dead center before the ignition stroke and the other after top dead center. For each crankshaft angular sector, the sector time interval through a given crankshaft angular sector is measured. A first difference is formed from the two sector time intervals belonging to the ignition stroke n, whereby the basic term of the rotational fluctuation is determined. A second difference is formed from the two sector time intervals of adjacent ignition strokes, whereby a rotational fluctuation correction term is determined. This is used to correct the difference in the fundamental terms caused by acceleration or deceleration. The correction of the basic term is performed by subtracting the correction term, thereby obtaining the rotation fluctuation value. If there is no misfire, this rotation fluctuation value is zero if the rotation speed is constant or the acceleration or deceleration is linear. On the other hand, if the rotational fluctuation value exceeds a predetermined threshold value, this indicates a misfire.

A method based on the second principle is described in DE-A-3917978 showing another prior art. In this principle, only one crankshaft angular sector is defined for each cylinder. A first difference is formed from the two sector time intervals of two different ignition strokes, by which the fundamental term of the rotational fluctuation is determined. The second difference is obtained from the two sector time intervals for the two ignition strokes, and these two ignition strokes are separated from each other by the same amount as the ignition stroke having the sector time interval used to form the first difference. ing. This second difference becomes a correction term for the rotational fluctuation. The basic term is newly corrected by subtracting the correction term, thereby obtaining the rotation fluctuation value. In this second method, the sector time intervals used for forming the difference will be provided and chosen so as to reflect the misfire as strongly as possible in the formed difference. However, the linear acceleration and deceleration are compensated by forming the difference twice.

[0008]

The above two principles have a drawback in that the effects of non-linear acceleration or deceleration cannot be well compensated. Therefore, an object of the present invention is to provide a method and a device for determining the rotational fluctuation value of an internal combustion engine, which can determine the value of the rotational fluctuation value so that the effects of nonlinear acceleration and deceleration can be compensated as well as possible. Is to provide.

[0009]

According to the present invention described below, a correction method is applied to the above-described method for determining a rotation fluctuation value. The correction method is the median (median)
To form a correction term. When obtaining the median value, instead of adding many values and dividing by the number of values, the values are arranged according to size, and the median value indicates the median value. This method of finding the median has been found to be very suitable for compensating for acceleration and deceleration effects of a larger class than the linear case. According to the embodiment using this principle, it is possible to reliably detect so-called bank misfire, that is, that all cylinders in one of the two cylinder banks do not ignite.

By using the above-mentioned principle, it is not necessary to form the difference of the time interval difference in order to calculate the rotation fluctuation value, and it is possible to directly use the time interval difference. In that case, one time interval is not a directly measured time interval, but a processed, preferably time interval obtained by determining the median. However, it has been found that an average value can be formed. Therefore, in another method according to the invention, averaging is used when determining the rotational variation of the internal combustion engine.

The first method of the present invention can be used for the above-mentioned two principles of forming a difference in time interval differences. It is done as follows.

That is, at least one crankshaft angular sector is defined for each cylinder with respect to the top dead center before the ignition stroke of each cylinder, and a sector that passes through a predetermined crankshaft angular sector for each ignition stroke. are timed interval, for each ignition stroke n, first difference from the two sectors time interval to determine the fundamental term of the rotation fluctuation is formed, also respectively around the ignition stroke n of interest Two sectors for the ignition stroke
A correction term for the rotational fluctuation is obtained by calculating a plurality of time interval differences and forming a median value of the calculated differences .
And the basic term is corrected by subtraction of the correction term,
Thereby, the rotation fluctuation value is obtained.

The second method of the present invention is to apply the principle of median value formation to a method in which the difference between sector time intervals is not processed, but processing is performed directly using sector time intervals. This is done as follows.

That is, a similar crankshaft angle sector is defined for each cylinder with respect to the top dead center before the ignition stroke of each cylinder, and a sector time interval for passing through a predetermined crankshaft angle sector for each cylinder. section for ignition stroke but is determined for each ignition stroke n, respectively centered on the ignition stroke n of interest
A correction term is formed by obtaining a plurality of data time intervals and obtaining the median value of the obtained sector time intervals, and subtracting the correction term from the sector time interval of the ignition stroke n of interest, thereby The rotation fluctuation value is obtained.

The third and fourth methods of the present invention utilize the idea of the method described above and modify it to form an average value instead of determining a median value. Of course, averaging has been found to be less effective in compensating for the effects brought about by non-linear accelerations or decelerations in determining rotational fluctuation values, but it is an improvement over the prior art.

The third method of the present invention is carried out in substantially the same manner as the second method of the present invention, except for the following. That instead of obtaining a median value, which corresponds to the ignition stroke each around the ignition stroke n of interest
A plurality of sector time intervals are obtained, and an average value of the obtained plurality of sector time intervals is formed.

The fourth method of the present invention has the following features. That is, for each cylinder, two crankshaft angular sectors are defined, one crankshaft angular sector each being located before top dead center before the ignition stroke of the cylinder and the other one located after top dead center, and For each cylinder, the time interval through which a given crankshaft angular sector is passed is measured, and for each ignition stroke n, the first from the two sector time intervals belonging to the ignition stroke to determine the basic term of the rotational fluctuation. the difference is formed, also when the two sectors for the ignition stroke each around the ignition stroke n of interest
A plurality of differences between the intervals are obtained, and a correction term for the rotational fluctuation is obtained by forming an average value of the obtained plurality of differences , and the basic term is corrected by subtracting the correction term,
Thereby, the rotation fluctuation value is obtained.

The dependent claims show preferred embodiments of the first two methods of the invention for determining the median. In order to better compensate for acceleration and deceleration, preferably to determine the median, the sector time intervals (or sectors) of an equal number of ignition strokes before and after the ignition stroke n of interest.
Time difference) is used. For more reliable detection of bank misfire, preferably first and last sector time
Interval (or the difference between the sector time interval), Ru is formed by forming a mean value of two consecutive sectors time interval (or the difference between the sector time interval). Also the first session
The time interval (or difference in sector time intervals) is
One sector time interval (or difference between sector time intervals)
Skips to the next sector time interval (or sector time interval)
Gap) and the third to last sector time interval
(Or the difference in sector time interval) is calculated between the next sector times.
Interval (or sector time interval difference)
The sector time interval (or the difference between the sector time intervals) is set to follow.

The apparatus of the present invention is configured to carry out the method of the present invention.

In other words, the first device of the present invention is arranged such that at least one crankshaft angle sector is defined for each cylinder with respect to the top dead center before the ignition stroke of each cylinder. Means for measuring the sector time interval passing through the sector, and forming a first difference from the two sector time intervals for each ignition stroke n to obtain a basic term of the rotational fluctuation and forming a correction term of the rotational fluctuation. And means for correcting the basic term by subtracting this correction term, thereby determining a rotation fluctuation correction term configured to obtain a rotation fluctuation value, and determining the rotation fluctuation correction term. means, two sectors for ignition stroke each around the ignition stroke n of interest
Obtains a plurality of differences in time interval, configured to determine a correction term by forming the median of a plurality of differences that this determined.

Also, the second device of the present invention is such that each crankshaft angular sector passes through a predetermined crankshaft angular sector defined for each cylinder with respect to top dead center before the ignition stroke of each cylinder. and means for measuring the sector time interval, provided with means for determining the correction term of the rotation fluctuation, this means, a plurality of sectors time interval of the ignition stroke each around the ignition stroke n of interest
Calculate the median of the calculated sector time intervals
The correction term is obtained by forming the correction term, and the correction term is subtracted from the sector time interval of the ignition stroke n, thereby obtaining the rotation fluctuation value.

In the third device of the present invention, each crankshaft angular sector passes through a predetermined crankshaft angular sector defined for each cylinder with respect to the top dead center before the ignition stroke of each cylinder. and means for measuring the sector time interval, provided with means for determining the correction term of the rotation fluctuation, this means, during the sector time for ignition stroke around the ignition stroke n respectively interest
A plurality of intervals are obtained, a correction term is obtained by forming an average value of the obtained plurality of sector time intervals, and the correction term is subtracted from the sector time interval of the ignition stroke n,
Thereby, the rotation fluctuation value is obtained.

In the fourth device of the present invention, one crankshaft angular sector for each cylinder is located before the top dead center before the ignition stroke of each cylinder, and the other is located after the top dead center. Means for measuring the time interval over which a given two crankshaft angular sectors pass through a defined crankshaft angular sector and for each ignition stroke n a first difference is formed from the two sector time intervals. To obtain the basic term of the rotational fluctuation, and to obtain the correction term of the rotational fluctuation, and to correct the basic term by subtracting the correction term to obtain the rotational fluctuation correction value. and a determining means, means for determining the correction term of the rotation fluctuation, points are centered around the ignition stroke n of interest
Find multiple differences between two sector time intervals for a fire stroke
Therefore, the correction term is obtained by forming an average value of the obtained plurality of differences .

[0024]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS In the method shown in the flow chart of FIG. 1, the crankshaft angle is measured in step s1 after starting the method in which the ignition of the internal combustion engine is turned on in time. In step s2, it is determined whether the measured angle coincides with the start of the crankshaft angle sector of the ignition stroke n + m. If so, the start time tAn + m of the sector is detected in step s3. After that, the process returns to step s1. If, at the next step s2, it becomes clear that the above-mentioned agreement does not exist, the process proceeds to step s4, during which it is determined whether the end of sector has been reached for the ignition stroke n + m. . If not, return to step s1 again,
In other cases, the process proceeds to step s5, and the end time tEn + m of the sector is detected. Next, in step s6, the time interval Tn + m of the crankshaft angular sector for the ignition stroke n + m is calculated from the values at the start time and the end time.

The count value n is continuously given for the ignition stroke. For example, in the cylinder 1 (with respect to the ignition sequence of the cylinders), the 1000th ignition stroke occurs,
If four cylinders are provided and a total of seven time intervals are required to detect the rotational variation of cylinder 1 in ignition stroke n (1000 in this case), ignition strokes 997 to 1003 You have to use up to. In this case, n = 1000 and m = 3. However, the rotation fluctuation value of the ignition stroke 1001 generated in the cylinder 2 may also be seen. For this ignition stroke n = 1001 and the time interval T1000 + 3 is the time interval T1001 + 2 for this cylinder.

The above assumes that only one sector time interval is measured per ignition stroke n.
However, as mentioned at the outset, the principle by which two sector time intervals are measured per ignition stroke n is also known. In that case, for each of the two sector time intervals, it must be determined when each start and end, and the associated time points are detected and the two sector time intervals are calculated accordingly. Must.

When all the sector time intervals necessary for calculating the rotation fluctuation value for the ignition stroke n have been obtained, the rotation fluctuation value LU is calculated in step s7. An example of the calculation method is shown in FIG.

After the calculation of the rotational fluctuation value in step s7, it is determined in step s8 whether the end of processing has been reached, for example, whether the ignition of the internal combustion engine has been cut off. If so, the process is terminated, otherwise the above process is repeated from step s1.

The illustrated process flow does not relate to how the rotational fluctuation value is evaluated. The evaluation is usually performed by comparing the rotation fluctuation value with a threshold value. However, there are some very complicated evaluation methods. For that, reference can be made, for example, to the above-mentioned publication DE-A-3917978. Evaluation methods are not the subject of the present invention. In the present invention, when calculating the rotational fluctuation value, the problem is to compensate as much as possible the effects brought about by acceleration and deceleration.

In the method of measuring only one sector time interval per ignition stroke, there is considerable room for selection of crankshaft angular sectors. But it is 720 ° / Z
It cannot be longer than (Z = cylinder). This is because otherwise it is not possible to determine which cylinder the misfire is based on. Preferably part of the crankshaft angular sector used is at least in the ignition stroke. Preferably, within an angle range of ± 720 ° / 2Z,
This is a symmetrical position centered on the top dead center before the ignition stroke of the cylinder.

To calculate the rotational fluctuation value, a method using two crankshaft angular sectors is used, one before the top dead center before the ignition stroke of each cylinder and the other behind the top dead center. If so, crankshaft angular sectors of approximately the same size can be used, or two sector time intervals are formed, the sector lengths being approximately equal when the internal combustion engine is operating without misfire. Can also be tuned to each other. The latter case is particularly effective. This is because pattern recognition is possible in that case. This kind of processing method has been described in the previous application (DE-40
28131).

FIG. 2 lists the time interval differences for the method of measuring only one time interval per ignition stroke. The difference Dn for the ignition stroke n is calculated from the time interval Tn + 1 of the ignition stroke n + 1 and the time interval Tn of the ignition stroke n. This applies for the preferred sector positions symmetrical with respect to top dead center as described above.

On the other hand, FIG. 3 lists how to calculate the difference in time intervals for a method of measuring two sector time intervals per ignition stroke n. The difference Dn of the ignition stroke n is formed from the time interval TKn measured in the compression stroke before the ignition stroke n and the time interval TEn measured in the expansion stroke, ie in the ignition stroke itself.

According to FIG. 4 (a), the calculation of the rotation fluctuation value LU in step s7 of the flow chart shown in FIG. 1 is performed by obtaining the median value (median).
With respect to the equation of FIG. 4 (a) used, FIGS. 2 and 3 are used. The cube of the quotient T described in the formula is the cube of Tn or the cube of TKn depending on whether the difference D described in the formula is calculated according to FIG. 2 or according to FIG.
It can be calculated as a power. Instead of one time cubed, for example, a product of three different times, eg T
It is also possible to use n-1 * Tn * Tn + 1. However, it is also possible to divide by one secondary or primary time term and take that into account accordingly in the evaluation. The publication DE-A-3917, which has already been explained many times about it.
978 can be referred to. The operation of the method illustrated by FIGS. 4 (a), 2 and 3 will be described below with reference to FIGS. 5 (a) to 5 (c).

Before going into the detailed description of FIGS. 5A to 5C, general matters of FIGS. 5A to 5F will be described. All these figures consist of two figures, the upper figure showing the time intervals with respect to the number of ignition strokes, and the lower figures each showing the difference in successive time intervals with respect to the ignition stroke. The measured value of the time interval for each ignition stroke n and the difference value calculated for this respective ignition stroke are marked "X" in all figures. An adjacent time interval used for calculation of the rotation fluctuation value LU or its difference is indicated by “x”.

FIG. 5A shows the ideal engine rotation when the rotation speed is constant. All time intervals T are of the same length, so all differences are zero.
As is apparent from the figure, in that case, the median formed by arranging the values used for the calculation according to the sizes and taking out the middle of the order as the median is also zero. Since the two terms of the difference equation 4a are zero, the rotation fluctuation value is also LUn = 0.

FIG. 5B shows the case where the deceleration is linear. The measured time intervals continuously increase with the same difference according to the ignition stroke. The fact that the differences are equal means that
It is clear from (b). The median of the differences exactly matches the difference Dn applied to the ignition stroke n, since all differences are equal. Therefore, also in this case, LUn = 0 holds.

FIG. 5C shows the case where the deceleration is non-linear. The measured time interval T increases with each ignition stroke, and in particular the magnitude of the difference increases, as is apparent from the lower diagram of FIG. 5 (c). When determining the median value shown in the equation of FIG. 4 (a), two difference values before the ignition stroke n and two difference values after the ignition stroke n are used, and one of the two difference values is obtained from the difference value of the ignition stroke n. The difference value of the ignition strokes n is in the middle, since the other two are smaller, so that the median of these five difference values is the same as the difference value Tn of the ignition strokes n. Therefore, also in this case, LUn = 0 holds. If an average value is formed, such a result cannot be obtained. As the difference gradually increases, the average value of the five difference values does not match the third of the five values, that is, the value Dn, and becomes larger than that as shown by the dotted line in FIG. 5 (c). I will end up.

When detecting the rotational fluctuation using the equation of FIG. 4 (a), a so-called bank misfire exists, that is, a problem occurs when all cylinders in one of the two cylinder banks have misfires. . In that case, the time interval T is alternately increased as shown in FIG.
It becomes low. Accordingly, the difference D alternately takes a positive value and a negative value. No misfire is detected, since the difference value of the ignition stroke n is again equal to the median value, so that the difference between the difference value Dn and the median value is likewise zero.

This defect is expressed by the equation of FIG. 4 (a) as shown in FIG. 4 (b).
Can be easily removed by changing it as described in the equation (1) or the equation in FIG. 4 (c). In that case, the case shown in FIG. 4 (b) is shown in FIG. 5 (e), and the case shown in FIG. 4 (c) is shown in FIG. 5 (f).

In the case of the equation shown in FIG. 4 (b) and FIG. 2 or 3, the value formed by the average value of the difference between two time intervals before and after to find the median is the outer value. Used. The average value mentioned above is zero, since the differences are alternately positive and negative with the same magnitude. Since this average value lies between the above-mentioned positive and negative values, the median value is zero and therefore deviates from the difference value Dn of the ignition stroke n. Therefore, a rotation fluctuation value LUn different from zero is formed.

It should be noted that this result is better than that obtained when forming the average value. The formation of the average value is shown in FIG.
It is explained by (d). When the average value is formed from the five values for forming the median value shown in FIG. 5D, a value slightly lower than zero is obtained. Because three of the five values are below zero and the two values are above zero by the same amount. Therefore, when the values shown in FIG. 5D are used, it is more advantageous to form the average value than to find the median value. However, as shown and described in FIG. With a slight modification, median formation can again be favored and the evaluation method can be adapted to its properties.

The examples shown in FIGS. 4C and 5F are more effective than the methods shown in FIGS. 4B and 5E. That is, instead of using the difference value of the ignition stroke n to form the median value, the difference between the previous two ignition strokes and the difference between the two subsequent ignition strokes is used, but the values on both outer sides are equal to one difference. Has been skipped and continues to lead to other differences. The four values used thereby lie on the positive side, only the difference with respect to the ignition stroke n lies on the negative side. Thereby the median is in agreement with the positive difference value. Since this value is subtracted from the negative difference value, a value twice as large as the difference Dn is obtained for the rotation fluctuation value LUn. This gives a very favorable S / N ratio.

As is clear from FIGS. 5 (e) and 5 (f),
The above mentioned advantages are obtained if only one of the two outer medians is formed by averaging or by jumping one difference.

The above embodiments relate to calculating the rotation fluctuation value by forming the difference of the sector time interval difference. In that case, the difference between the time intervals is taken as the individual ignition stroke n.
It was not important whether it was formed from the two time intervals of, or from different ignition strokes n.
Forming the difference from the time interval uses all conventional methods. However, as shown in the equation of FIG. 4 (d), for example, the median is determined by the time interval Tn of the ignition stroke n.
It has been found that it is most suitable for calculating the rotation fluctuation value by a simple difference consisting of the median centered around the time interval. It is directly apparent from the above-mentioned diagram of FIG. 5 (c) that this calculation can compensate not only for a certain effect but also for linear and nonlinear acceleration and deceleration.

Irrespective of the type of acceleration or deceleration, the sector time interval Tn of the ignition stroke n is the center time of the time intervals Tn-2 to Tn + 2 of a plurality of consecutive ignition strokes centered on the ignition stroke n. It is an interval. The median value thereby corresponds in all cases to the sector time interval Tn of the ignition stroke n itself. Therefore, the rotation fluctuation value LUn is zero. When bank misfire is detected, the same problem as described with reference to the lower diagram of FIG. However, this problem is also solved by choosing different outer time intervals corresponding to what was described with respect to the lower figures, as can be seen from the upper figures of FIGS. 5 (e) and (f). . Therefore, it is preferable to determine the median value at the time interval T.
It is performed based on n-3, Tn-1, Tn, Tn + 1, Tn + 3.

It should be noted that FIG. 4D shows that the median value centered on Tn is obtained. in this case,
The median value can be obtained in the same manner as described for forming the difference using FIGS. 4 (a) to 4 (c). That is, the formation of the median centered on Tn can be done using five consecutive and directly measured time intervals, or three directly measured median time intervals and two averaged outer times. , Or 5 directly measured time intervals, of which the outer time intervals are performed with one time interval being the other time interval.

Calculating the rotational fluctuation value using the time interval and the median of the time intervals is much easier than calculating the rotational fluctuation value from the median of the time interval difference and the time interval difference. . Despite this ease, the experimental results have shown that the evaluation results have not deteriorated at all. This is apparent from the above description with respect to the two figures of Figures 5 (c) -5 (f).

Similarly, from the viewpoint of simplification, the method shown in FIG. 4D can be changed to replace the median term used as the multifunctional term for forming and correcting the difference with the average value. As shown in FIGS. 5 (c) -5 (f), the mean term, unlike the median term, is less well suited to compensating for errors caused by acceleration or deceleration or bank misfires. However, average value formation is also superior to the prior art. The formation of the average value is shown in FIG.
The equation shown in (f) and shown in 4 (e) is the sector time interval centered on the ignition stroke n from the time interval Tn of the ignition stroke n itself in the method of measuring only one time interval per ignition stroke. 4 shows a case where a rotation fluctuation value is formed by subtracting the average value of
(F) shows all the time interval differences as shown in FIG. 3 from the time interval difference Dn regarding the ignition process n to the ignition process n.
It shows a case where the rotation fluctuation value is formed by subtracting the average value of the time interval differences of the ignition strokes centered on.

FIG. 6 schematically shows an internal combustion engine having a toothed disk 12 detected by a pulse generator 11 at the end of the crankshaft opposite the drive side. The signal of the pulse generator 11 is supplied to a control device 13, which means 14 for measuring the time interval of the crankshaft angular sector and means 1 for determining the correction term for the rotational fluctuations.
5 are provided.

The means 14 for measuring the sector time interval comprises a predetermined crankshaft angular sector in which each at least one crankshaft angular sector is defined for each cylinder with respect to the top dead center before the ignition stroke of the respective cylinder. Is configured to measure a sector time interval passing through. The means 14 also measure the sector time interval at which each crankshaft angular sector passes for each cylinder through a predetermined crankshaft angular sector defined with respect to top dead center before the ignition stroke of each cylinder. It can also be configured to. Further, the means 14 comprises for each cylinder two crankshaft angular sectors, one crankshaft angular sector being located before the top dead center before the ignition stroke of each cylinder and the other behind the top dead center. It can also be arranged to measure the time interval through which a given predetermined crankshaft angular sector is passed.

The means 15 for determining the correction term for the rotational fluctuation can be constructed as follows. That is, the first difference is formed from the two sector time intervals for each ignition stroke n to obtain the basic term of the rotational fluctuation, and the sector time intervals of a plurality of ignition strokes centered on the target ignition stroke n are obtained. It is possible to form a correction term for the rotational fluctuation by forming a median value or an average value from the difference, and correct the basic term by subtracting this correction term, thereby obtaining the rotational fluctuation value. Furthermore, a multifunctional term is formed by obtaining the median value of the sector time intervals of a plurality of ignition strokes centered on the target ignition stroke n, and the multifunctional term is subtracted from the sector time interval of the ignition stroke n. However, it may be configured to obtain the rotation fluctuation value accordingly.

Means 14 for measuring the crankshaft angular sector
Which embodiment of the invention to use with which embodiment of the means 15 for determining the rotational fluctuation value depends on which of the methods of the invention is carried out.

[0054]

As is apparent from the above description, according to the present invention, there is provided a method and apparatus for determining the rotational fluctuation of an internal combustion engine, which can compensate the effects of nonlinear acceleration and deceleration as well as possible. be able to.

[Brief description of drawings]

FIG. 1 is a flowchart illustrating a method of determining a rotation fluctuation value of an internal combustion engine.

FIG. 2 is a table illustrating the difference formed in the method of measuring only the time interval of one crankshaft angular sector per ignition stroke.

FIG. 3 is a table similar to FIG. 2, but illustrating the differences created in the way the time intervals of two crankshaft angular sectors per ignition stroke are measured.

FIG. 4 is a table showing various expressions for calculating a rotation fluctuation value.

FIG. 5 is a graph showing the values of the difference between successive sector time intervals and sector time intervals with respect to time for various cases of driving the internal combustion engine constantly and dynamically.

FIG. 6 is a block diagram showing an outline of an apparatus for determining a rotation fluctuation value of an internal combustion engine.

[Explanation of symbols]

10 Internal combustion engine 11 pulse generator 12 toothed disc 13 Control device Means for measuring 14 sector time intervals 15 Means for determining the correction term for rotational fluctuation

─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Martin Klenk Germany 7150 Backnang Stresemannstraße 11 (72) Inventor Vinfried Mosel Germany 7140 Ludwigsberg Grundweinberge 14 (56) References 51-104130 (JP, A) JP-A-51-104133 (JP, A) JP-A-1-244313 (JP, A) International publication 90/15235 (WO, A1) (58) Fields investigated (Int.Cl) . ⁷ , DB name) F02D 45/00 F02P 5/15 F02P 17/12 G01M 15/00

Claims (18)

(57) [Claims] 1. At least one crankshaft angular sector is defined for each cylinder with respect to top dead center before the ignition stroke of the respective cylinder, and for each ignition stroke a sector is passed through a predetermined crankshaft angular sector. The time interval is measured, and for each ignition stroke n, a first difference is formed from the two sector time intervals to determine a basic term of the rotational fluctuation, and a correction term of the rotational fluctuation is determined and the correction is performed. is the basic term corrected by subtraction of the term, whereby rotation fluctuation value is determined to provide a method for determining the rotation fluctuation value of the internal combustion engine, the ignition stroke each around the ignition stroke n of interest
Find the difference between the two sector time intervals for
A method for determining a rotational fluctuation value of an internal combustion engine, characterized in that a correction term is obtained by forming a median value of a plurality of differences . 2. A method as claimed in claim 1, characterized in that for each cylinder a single crankshaft angular sector is defined which is symmetrically located with respect to top dead center before the ignition stroke of the cylinder. 3. Two crankshaft angular sectors are defined for each cylinder, one crankshaft angular sector being located before top dead center before the ignition stroke of the cylinder and the other being located after top dead center. The method of claim 1, wherein the difference is formed from two sector time intervals of one ignition stroke, respectively. 4. A sector time interval in which a similar crankshaft angle sector is defined for each cylinder with respect to the top dead center before the ignition stroke of each cylinder, and a predetermined crankshaft angle sector is passed for each cylinder. There are measured, the method for determining the rotation fluctuation value of the internal combustion engine, for each ignition stroke n, the ignition stroke each around the ignition stroke n of interest
Obtain multiple sector time intervals for the
A correction term is obtained by forming the median of the sector time intervals of the internal combustion engine, and the correction term is subtracted from the sector time interval of the target ignition stroke n, and thereby the rotational fluctuation value is obtained. A method for determining the rotational fluctuation value of an engine. 5. To determine the median, before for the ignition stroke of the number of such each before and after the ignition stroke n of interest
4. A method according to any one of claims 1 to 3, characterized in that the difference of the sector time intervals is used. 6. The method according to claim 5, characterized in that the difference of the sector time intervals for each two ignition strokes before and after the ignition stroke n of interest is used. 7. A target ignition for obtaining a median value.
For each equal number of ignition strokes before and after stroke n,
5. The method according to claim 4, wherein a Kuta time interval is used.
The method described. 8. It before and after the ignition stroke n in question
Sector time intervals for two ignition strokes are used
The method according to claim 7, wherein: 9. The maximum difference among the plurality of sector time intervals.
The difference between the first and last sector time intervals is 2 consecutive
Formed by forming an average value of the difference of two sector time intervals .
6. The method according to any one of 6 above. 10. A difference among the plurality of sector time intervals
The first sector time interval difference is equal to the sector time interval difference.
Jump over to the next sector time interval difference, and finally
The difference in the third sector time interval is from the next sector time
Interval gap and continue to the last sector time interval difference.
7. Any one of claims 1, 2 , 3 , 5 and 6 characterized by
The method described in the section. 11. The first of the plurality of sector time intervals
And the last sector time interval are two consecutive
Formed by forming the average value of the Kuta time interval
9. Any one of claims 4, 7, and 8 characterized in that
The method described . 12. The first of the plurality of sector time intervals
The sector time interval of is one sector time interval
Continues to the next sector time interval, and is the third to last sector
The data time interval skips the next sector time interval
8. The sector time interval follows, 4, 7,
8. The method according to any one of item 8 . 13. A sector time interval in which a similar crankshaft angle sector is defined for each cylinder with respect to top dead center before the ignition stroke of each cylinder, and a predetermined crankshaft angle sector is passed for each cylinder. There are measured, the method for determining the rotation fluctuation value of the internal combustion engine, for each ignition stroke n, the ignition stroke each around the ignition stroke n of interest
Obtain multiple sector time intervals for the
The internal combustion engine is characterized in that the correction term is obtained by forming an average value of the sector time intervals of, and the correction term is subtracted from the sector time interval of the target ignition stroke n, thereby obtaining the rotational fluctuation value. A method for determining the rotational fluctuation value of an engine. 14. For each cylinder there are defined two crankshaft angular sectors, one crankshaft angular sector each located before top dead center before the cylinder's ignition stroke and the other behind top dead center. And for each cylinder the sector time interval through a given crankshaft angular sector is measured, for each ignition stroke n the two sector times belonging to the ignition stroke to determine the basic term of the rotational fluctuation A first difference is formed from the interval, a correction term for the rotational fluctuation is obtained, and the basic term is corrected by subtracting the correction term,
Whereby rotation fluctuation value is determined to provide a method for determining the rotation fluctuation value of the internal combustion engine, the ignition stroke each around the ignition stroke n of interest
Find the difference between the two sector time intervals for
A method for determining a rotational fluctuation value of an internal combustion engine, characterized in that a correction term is obtained by forming an average value of a plurality of differences . 15. Measuring a sector time interval in which each at least one crankshaft angular sector passes a predetermined crankshaft angular sector defined for each cylinder with respect to top dead center before the ignition stroke of each cylinder. Means (14) for forming a first difference from two sector time intervals for the ignition stroke n to obtain a basic term of the rotational fluctuation, and a correction term for the rotational fluctuation is formed. Means for correcting the fundamental term by subtraction and thereby determining a rotational fluctuation correction term configured to determine a rotational fluctuation value (1
5) and a, in the device for determining the rotation fluctuation value of the internal combustion engine, means for determining the correction term of the rotation fluctuation, two for ignition stroke each around the ignition stroke n of interest
Obtain a plurality of sector time interval differences,
Apparatus for determining a rotational fluctuation value of an internal combustion engine, characterized in that it is arranged to determine a correction term by forming a median of the differences . 16. Means for measuring the sector time interval for each crankshaft angular sector to pass through a predetermined crankshaft angular sector defined for each cylinder with respect to top dead center before the ignition stroke of each cylinder. 1
In the device for determining the rotation fluctuation value of the internal combustion engine, which comprises 4), means (15) for determining a correction term of the rotation fluctuation is provided,
This means is centered around the target ignition stroke n.
And obtain multiple sector time intervals for the ignition stroke.
A correction term is obtained by forming a median value of the obtained plurality of sector time intervals, and the correction term is subtracted from the sector time interval of the ignition stroke n, thereby obtaining a rotation fluctuation value. An apparatus for determining a rotation fluctuation value of an internal combustion engine, which is characterized in that: 17. Means for measuring the sector time interval for each crankshaft angular sector to pass through a predetermined crankshaft angular sector defined for each cylinder with respect to top dead center before the ignition stroke of each cylinder. 1
In the device for determining the rotation fluctuation value of the internal combustion engine, which comprises 4), means (15) for determining a correction term of the rotation fluctuation is provided,
This means is centered around the target ignition stroke n.
And obtain multiple sector time intervals for the ignition stroke.
A correction term is obtained by forming an average value of the obtained plurality of sector time intervals, and the correction term is subtracted from the sector time interval of the ignition stroke n, thereby obtaining a rotation fluctuation value. An apparatus for determining a rotation fluctuation value of an internal combustion engine, which is characterized in that: 18. For each cylinder there are two crankshaft angular sectors, one crankshaft angular sector located before the top dead center before the ignition stroke of each cylinder and the other behind the top dead center. Means for measuring a sector time interval passing through a predetermined crankshaft angle sector (1
4) and for each ignition stroke n, the first difference is formed from the two sector time intervals to obtain the basic term of the rotational fluctuation, the correction term of the rotational fluctuation is determined, and the correction term is subtracted to obtain the above Means (15) for determining a rotational fluctuation correction term configured to correct the fundamental term and thereby obtain a rotational fluctuation value
With the door, in a device for determining the rotation fluctuation value of the internal combustion engine, means for determining the correction term of the rotation fluctuation, two for ignition stroke each around the ignition stroke n of interest
Obtain a plurality of sector time interval differences,
Apparatus for determining a rotational fluctuation value of an internal combustion engine, characterized in that it is arranged to determine a correction term by forming an average value of the difference .