JPH0461291B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field] The present invention relates to a method for measuring output fluctuations of an internal combustion engine.

[Prior art]

In recent years, efforts have been made to optimally adjust the ignition timing and air-fuel ratio of engines as a measure to prevent exhaust pollution and save energy. Measurements are being taken.

The conventional method for measuring this output fluctuation is
51-104106, JP-A-53-65531, JP-A-57-
106834 and the like are known. In all of these conventional methods, the time Ti required for one revolution of the crankshaft is measured in time series, and the average rotational speed for each revolution is successively compared to determine fluctuations in the internal combustion engine.

On the other hand, the number of rotations during a period in an actual vehicle is also affected by the difference in the cross-sectional condition of the road, whether it is a paved road or a rough road. Figure 1 shows the variation of the rotational speed measured at a rotation interval of 30°C of the crankshaft while the vehicle was running at a constant speed. is the waveform for a paved road, and (3) is the waveform for a rough road. Here, one revolution corresponds to 360℃A,
The air-fuel ratio A/F is chosen to be 14.5. As is clear from FIG. 1, even though the vehicle is traveling at a constant speed, when the road is rough, the rotational speed changes greatly due to the unevenness of the cross section.

As described above, the average rotational speed measured by the time Ti required for one rotation of the crankshaft is greatly influenced not only by engine output fluctuations but also by engine load fluctuations due to unevenness of the cross section. Therefore, with the conventional method of measuring rotational speed fluctuations, that is, output fluctuations, by measuring the average rotational speed per revolution from the time Ti required for one crankshaft rotation, it is difficult to accurately detect output fluctuations. There is a problem.

[Purpose of the invention]

It is an object of the present invention to maintain both the steady state and the transient state of an internal combustion engine during actual vehicle driving without being affected by the cross-sectional state, load fluctuations, accelerator pedal operation, or transient states. The purpose is to enable accurate measurement of output fluctuations.

[Structure of the invention]

In the present invention, the rotational speed of an internal combustion engine is detected, and the rotational speed N _n detected at at least two predetermined crank angle positions in the explosion stroke of the engine,
An internal combustion engine that calculates a rotational speed change amount ΔN based on the deviation of N _o , corrects the rotational speed change amount ΔN more as the rotational speed increases, and measures the corrected rotational speed change amount as an output fluctuation of the engine. A method for measuring engine output fluctuations is provided.

The present invention is based on the following findings obtained by the inventor.

That is, FIG. 2 shows the change over time in the rotational speed of a 4-cycle, 4-cylinder engine. Each cylinder repeats the explosion stroke in the order of the 1st, 3rd, 4th, and 2nd cylinders. Indicates a change in rotational speed. S (TDC) represents the top dead center signal, and S' is the 30° of the crankshaft.
It represents the rotation period signal output every rotation of °CA. N _o (1), N _n (1), N _o (2), N _n (2) each represent the average rotational speed of a predetermined cylinder (first cylinder in this case) at a predetermined crankshaft rotation angle, The subscript n in N _o (1) etc. is 60 from 30℃A after TDC.
The average rotational speed up to ℃A, and N _n (1)
etc., m represents the average rotational speed after TDC and from 90℃A to 120℃A, and furthermore, N _o (1), N _o
The numbers (1) and (2) in (2), etc. represent the average rotational speed in the first and second explosion strokes of a predetermined cylinder, respectively.

If we observe changes in engine speed from a microscopic perspective, we can see that changes in engine speed consist of relatively long-cycle changes in engine speed caused by accelerator operation and the above-mentioned load fluctuations, and short-cycle changes that occur during the explosion stroke of each cylinder. The pulsating rotational speed changes are superimposed. Similarly, pulsating changes appear in engine torque or cylinder pressure.

Using the above-mentioned pulsating rotation speed that appears in the explosion stroke of the internal combustion engine, calculate the amount of rotation speed change △N at a predetermined timing in successive explosion strokes of a predetermined cylinder using the following formula, △N=N _n −N _o . . This rotational speed change △N and average effective pressure
The relationship with Pi is shown in Figures 3 and 4. In Figures 3 and 4, the horizontal axis is the average effective pressure Pi [Kg/cm ² ]
, the vertical axis represents the rotational speed change △N [rpm],
Fig. 3 shows the case of 1000 rpm, and Fig. 4 shows the case of 1500 rpm. As shown in FIGS. 3 and 4, if the rotational speed changes, the relationship between Pi and ΔN also changes.

Figure 5 shows the relationship between the standard deviation σ (Pi) of the average effective pressure Pi and the standard deviation σ (△N) of the rotational speed variation △N.
It is shown in FIG. In Figure 5, the horizontal axis is the air-fuel ratio A/
The left vertical axis (white circle) is the standard deviation σ (Pi) [Kg/cm ² ] of the average effective pressure Pi, and the right vertical axis (black square) is the standard deviation σ ( △N)
Represents [rpm]. In Figure 5, the rotation speed is
1000rpm, torque is 4Kg・m. In Figure 6, the horizontal axis represents σ (Pi) [Kg/cm ² ], and the vertical axis represents σ
(△N) represents [rpm].

As shown in FIGS. 3 to 6, the output of the internal combustion engine can be detected with considerable accuracy by using the pulsating rotational speed that corresponds to the explosion stroke of the internal combustion engine.

Figures 7 and 8 show △N calculated from the difference between the △N obtained by the conventional method when changing the road image state and the amount of pulsation change during the explosion stroke. In the case of the conventional method, FIG. 8 shows the case of the method of the present invention, which calculates from the difference in the amount of change in pulsation during the explosion stroke. In Figures 7 and 8, (1) represents the case on a platform, and (2) represents the case on a rough road.

As shown in Figure 7, the conventional method of measuring the time Ti required for one rotation of the crankshaft over time and successively comparing the rotation speeds obtained from the values to determine the rotation speed fluctuations, that is, combustion fluctuations of the internal combustion engine, is based on the cross-sectional state. greatly influenced by. On the other hand, as shown in Fig. 8, the pulsating rotational speed change that appears in the explosion stroke of the internal combustion engine is determined, and the rotational speed change in successive explosion strokes in a predetermined cylinder is compared sequentially. It is clear that the influence of the approximate surface is very small in this method of determining fluctuations. In other words, this is because the amount of pulsation change in each explosion stroke is measured in an extremely short period of time, and the amount of rotation speed change due to accelerator operation or engine load fluctuation during that time can be almost ignored. It becomes clear that output measurement that is not affected by surface conditions is possible.

On the other hand, as can be seen from Figures 3 and 4,
The relationship between the rotational speed variation △N according to the present invention at each rotational speed, that is, the rotational speed change △N calculated from the difference in pulsating rotational speed during the explosion stroke, and the average effective pressure Pi is for each rotational speed. Showing different slopes.

To find the combustion fluctuation, the rotational speed change △N that indicates the combustion state in a given cylinder, for example in Fig. 2, △N ₁ = N _n (1)−N _o (1) △N ₂ = N _n ( 2)−N _o (2) or detect this rotational speed variation ΔN ₁ ,
It is sufficient to detect the rotational speed fluctuation △N′ △N′=△N ₁ −△N ₂ which is the difference _between △N 2 , but as mentioned above, the characteristics differ depending on the rotational speed, so The amount of change ΔN cannot be simply compared.

Therefore, by detecting the pulsating rotational speed variation △N, △N=N _n −N _o that appears in the explosion stroke of the internal combustion engine, and correcting this detected △N by the rotational speed, or By correcting each rotation speed before calculating the rotation speed change amount during the explosion stroke, and using the corrected value to find the pulsating rotation speed change amount △N that appears in the explosion stroke, the rotation speed can be calculated. Since it is unaffected, that is, the rotation speed can be adjusted to a certain constant value, it is possible to accurately detect the amount of change in the rotation speed without being affected even in a transient state, and therefore, it is possible to accurately measure output fluctuations.

〔Example〕

FIG. 9 shows an apparatus for performing a method for measuring output fluctuations in an internal combustion engine as an embodiment of the present invention.

In Fig. 9, E is 4 which is the output fluctuation measurement target.
It is a cycle, four-cylinder engine, and an angle signal plate 3 is attached to a pulley 2 located at the tip of a crankshaft 1. The angle signal plate 3 is a disk made of magnetic material, and 12 teeth are formed on its circumference.
Reference numeral 4 denotes an angle signal sensor provided so as to face the teeth of the signal plate 3, which outputs a pulse signal every time each tooth of the signal plate 3, which rotates integrally with the shaft 1, passes. Therefore, sensor 4 detects 12 rotations of shaft 1.
Outputs a pulse signal. Also, shaft 1 is the
During one round of the explosion stroke from the cylinder to the 4th cylinder, 2
Rotate.

Reference numeral 5 denotes a distributor, and 6 a cylinder discrimination sensor built into the distributor 5, which outputs a pulse signal at the compression top dead center of the first cylinder. Reference numeral 7 denotes an output fluctuation calculation unit, into which pulse signals from the angle signal sensor 4 and cylinder discrimination sensor 6 are input.

FIG. 10 shows the configuration of the output fluctuation calculation unit 7, and the calculation unit will be explained with reference to FIG. The angle signal from the angle signal sensor 4 is waveform-shaped by a waveform shaping circuit 71A, and then passed to the rotation speed counting circuit 7.
2 is input. The rotation number counting circuit 72 is composed of a 16-bit binary counter, and counts the 30°A cycle of the angle signal and sends it to the calculation circuit 74 as binary data, which calculates the reciprocal of this and uses it as the rotation number. .
A microcomputer can be used as the calculation circuit 74.

The cylinder discrimination signal from the cylinder discrimination sensor 6 is waveform-shaped by the waveform shaping circuit 71B, and is inputted to the calculation circuit 74 via the reading circuit 73. The calculation circuit 7
The engine speed or torque fluctuation calculated in step 4 is converted into an analog signal by a D/A conversion circuit 75 and sent to an air-fuel ratio control device or an ignition timing control device (not shown).

The procedure for calculating the rotation speed fluctuation performed within the calculation circuit 74 will be explained with reference to FIGS. 11 and 12.

FIG. 11 shows the cylinder discrimination signal 71b. The signal 71b remains at the "0" level from a while after the first cylinder reaches top dead center until a while after the crankshaft 1 has rotated 30 degrees.

FIG. 11 shows the rotation period signal 71a. The number of each pulse of signal 71a is crankshaft 1.
is attached sequentially every time the first cylinder rotates 30 degrees from the top dead center, and the first cylinder corresponds to 0 to 5, and then corresponds to the third, fourth, and second cylinders in order up to 23. It is attached.

In FIG. 11, the average number of revolutions per 30°A revolution of the crankshaft is shown as a bar graph.

Now, every time the rotation period signal 71a falls, an interrupt signal is issued to the computer 74, and the rotation speed fluctuation calculation program whose flowchart is shown in FIG. 12 is activated.

The above calculation program saves the contents of the register before the interrupt to the memory in step S01, then issues the read pulse signal SEL3 in step S02, and transfers the cylinder discrimination signal 71b appearing on bus B to register A1.
Load into. Register A1 at step S03
The MSB, that is, the level of the cylinder discrimination signal 71b is determined, and if it is at the "0" level indicating the start of the explosion stroke of the first cylinder, the counting memory m is cleared in step S05. If the level is "1", 1 is added to the memory m in step S04. step
In S06, the contents of memory m are 1, 3, 7, 9, 13,
If it is any one of 15, 19, or 21, the process proceeds to step S07 and onwards, otherwise the process jumps to step S16 and ends the process.

At step S07, read pulse signal SEL1,
SEL2 is issued and the rotation period data T is read from the counting circuit 72 into the register A2. In step S08, the content of register A2, that is, the reciprocal of rotation period data T is taken, multiplied by an appropriate proportionality constant K to calculate the average rotation number, and stored in register A3.

Fig. 13 is a diagram in which Figs. 3 and 4 are summarized in terms of rotational speed. The horizontal axis shows the rotational speed N [rpm], and the vertical axis shows the correction coefficient α depending on the rotational speed. The rotation speed is corrected using the coefficient α.

To explain this rotational speed correction method, step S09 calculates the rotational speed correction coefficient α corresponding to the rotational speed stored in the register A3 using the map shown in FIG. The rotation speed is multiplied by the correction coefficient α, the result is stored in register A4, and the register is stored in step S11.
Store the contents of A4 in memory.

Eight such memories are prepared for one explosion stroke, corresponding to the possible values of the counting memory m, ie, 1, 3, 7, 9, 13, 15, 19, and 21. Memory M _n (1) is a memory that stores the average rotation speed in the first explosion stroke, and memory M _n
(2) is the memory for the second explosion stroke.
In other words, memory M ₁ (2) contains the information that crankshaft 1 changes from 30 degrees during the second explosion stroke of the first cylinder.
The average number of rotations when rotating up to 60 degrees is memorized,
The memory M ₂₁ (2) stores the average rotational speed when the crankshaft 1 rotates from 90 degrees to 120 degrees during the explosion stroke of the second cylinder, which is the last cylinder in the rotational speed fluctuation measurement cycle.

In step S12, the contents of memory m are 3, 9,
If it is either 15 or 21, the process proceeds to step S13 and thereafter, otherwise the process jumps to step S16 and ends the process. In step S13, previous steps S07 to S11
The average rotational speed N _n (2) at a predetermined crankshaft rotation angle of a predetermined cylinder calculated and stored in the memory M _n (2) and the memory M _n calculated in the previous measurement cycle
The rotational speed variation ΔN' of the predetermined cylinder is calculated from the average rotational speed N _n (1) of the predetermined cylinder at the predetermined crankshaft rotation angle stored in (1). This calculation formula is shown in equation (1).

△N′=(N _n-2 (1)−N _n (1)) −(N _n-2 (2)−N _n (2)) ……(1) In formula (1), m is the first , 3, 9, 15, and 21 corresponding to the third, fourth, and second cylinders.

To explain this formula (1), (N _n-2
(1)-N _n (1)) and (N _n-2 (2)-N _n (2)) are the rotational speed changes △N, which are shown in Figures 3 and 4. As shown in FIG. 8, this value corresponds to the combustion of the internal combustion engine and is not affected by the road surface condition. Furthermore, in this embodiment, when calculating the amount of change in rotational speed ΔN, the average rotational speed is
Since the rotation speed is corrected using the correction coefficient α shown in the figure, this rotation speed change amount ΔN is not affected by the rotation speed. Therefore, the rotational speed variation △ determined by formula (1)
N′, that is, the amount of change in rotational speed corresponding to combustion from the pulsating average rotational speed that appears during the explosion stroke △N
The rotational speed fluctuation △N' obtained as the difference between successive △N of a given cylinder is accurate even during transient times when the rotational speed of the internal combustion engine changes, and this value is not affected by road surface conditions. , this rotation speed fluctuation △
Combustion fluctuations in each predetermined cylinder can be accurately detected by N'.

Return to the flowchart in Figure 12 again and follow the steps.
In S14, the rotation speed fluctuation ΔN' of each cylinder calculated in the previous step S13 is outputted to the D/A conversion circuit 75.
In step S15, the memory M _n (2) is stored in the memory M _n
Move to (1). In step S16, the contents of the register before the interrupt, which were saved in the memory at the beginning of the arithmetic program, are restored.

The above rotation speed fluctuation calculation program is based on the periodic signal 71.
It is activated every time a falls, and the memory M _n (1)
The average rotational speed as schematically shown in FIG. 11(3) is stored in the memory M _n (2). The height of the middle bar graph in FIG. 11 (3) indicates the size of the average number of revolutions, and each average number of revolutions is shown at the top of each graph.

The average rotational speed of the crankshaft 1 every 30 degrees exhibits periodic pulsations as shown by the broken line in FIG. 11 as the explosion stroke of each cylinder occurs.

As shown in the flowchart of FIG. 12, the calculation circuit 74 calculates only the average rotational speed of the crankshaft 1 from 30 degrees to 60 degrees and from 90 degrees to 120 degrees for the explosion stroke of each cylinder. This is number 11
It is shown by a solid line in FIG.

For example, the rotational speed fluctuation △N′ of the first cylinder is calculated as △N′=(N ₁ (1)−N ₃ (1))−(N ₁ (2)−N ₃ (2)), and this The output fluctuation of the first cylinder can be known from the rotational speed fluctuation ΔN'.

In this way, in the device shown in Fig. 9, the output of the internal combustion engine is determined from the difference in pulsating rotational speed at a predetermined timing during the explosion stroke, that is, from the amount of change in rotational speed △N, thereby supporting combustion that is not affected by road surface conditions. Since this measured value is corrected for the rotation speed, it is possible to always measure a value that has a constant relationship with combustion fluctuations at any rotation speed. Therefore, even if engine conditions (rotational speed) change, accurate output fluctuations can always be obtained by determining the difference in pulsating rotational speed (rotational speed variation △N) at a predetermined timing in successive explosion strokes of a predetermined cylinder. Can be detected.

As a result, the measured values are not affected by load fluctuations, accelerator pedal operation, or other transient conditions because the rotational speed is corrected, making it possible to accurately measure engine output fluctuations even when the vehicle is actually running. By using this value, the ignition timing and air-fuel ratio can be optimally adjusted.

In implementing the present invention, various modifications can be made in addition to the embodiments described above. For example, in the embodiments described above, engine output fluctuations were measured by changes in engine speed, but the same number of measurements can also be made by changes in torque or cylinder pressure.

In addition, in the present invention, as shown in FIGS. 2 and 11,
Although the average rotational speed at intervals of 30° is used, it has been experimentally confirmed that a correlation with combustion in an internal combustion engine can be obtained even if the average rotational speed at intervals shorter than 30° is used.

In addition, in the above-mentioned embodiment, when calculating the pulsating rotational speed change amount △N that appears in the explosion stroke of the internal combustion engine, the instantaneous rotational speed is corrected by the rotational speed correction coefficient α. The combustion fluctuations were detected by sequentially comparing the successive rotational speed changes of the predetermined cylinders obtained by Even if the rotation speed is corrected, the same effect as in the above-mentioned embodiment can be obtained. In the embodiment described above, the rotation speed correction coefficient α is calculated for each instantaneous rotation speed, whereas in other embodiments, the correction coefficient α for the rotation speed is determined for each instantaneous rotation speed. A correction coefficient α is determined and calculations are performed based on this.

Another embodiment will be explained with reference to the flowchart of FIG. Steps S21 and S25-S27 are similar to steps S01, S02-S04 in the previous embodiment. step
In steps S22 to S24, the instantaneous rotational speed A3 and its integrated value A4 are determined, and in steps S28 to S31, the average rotational speed A5 and its correction coefficient α1 (α2 is the value before one cycle) are determined. Steps S32 and -S34 are the same as in the previous embodiment. In step S35, the rotation speed fluctuation △N′ is expressed as
(2) That is, △N′=α1・(N _n-2 (1)−N _n (1))−α ₂・ (N _n-2 (2)−N _n (2)) ……(2) Find it by

This is basically the same as the previous embodiment, but each term for which the difference in rotational speed is calculated is multiplied by the respective correction coefficients α1 and α2. Step S36~
S38 is similar to steps S14 to S16 in the previous embodiment.

In each of the above-mentioned examples, the pulsating rotational variation was successively compared and the combustion fluctuation was determined from the difference.
However, the present invention is not limited to this, and the same effect can be obtained by performing rotational correction on the amount of rotational change, statistically processing the value to obtain a standard deviation, and calculating combustion fluctuation from this standard deviation.

As yet another embodiment of the present invention, a case where the above-mentioned statistical processing is performed will be explained using the flowchart of FIG. 15. Steps S01 to S12 are the same as in the previous embodiment. The pulsating rotational speed changes corrected in steps S41 to S47 are statistically processed to calculate the standard deviation σ.
(ΔN) In other words, the combustion fluctuation is determined. In step S41, the pulsating rotational change amount is calculated, and in step S42, the pulsating rotational change amount △N _n of each cylinder calculated in the previous step S41 is stored in the memory △N _n・_o for statistical processing calculations. Store.

In step S43, pointer n is incremented by one,
It is compared with 100 in step S44, and if it is smaller, it jumps to step S16. On the other hand, if it is more than 100, the pointer n is returned to "0" in step S45, and the 100 data of △N _n・_o is returned in step S46. The standard deviation σ(ΔN) is obtained by performing statistical processing on the equation.
Therefore, the standard deviation σ(ΔN) is obtained for every 100 pieces of data of the rotational speed variation ΔN _n . At step S47, σ(△N) is digitally/
Convert to analog.

〔Effect of the invention〕

According to the present invention, during actual vehicle driving, the internal combustion engine is not affected by road surface conditions, load fluctuations, accelerator pedal operation, or transient states, and the internal combustion engine is not affected by either the steady state or the transient state. Even output fluctuations can be measured accurately. In addition, it becomes possible to accurately measure output fluctuations during actual vehicle running, and as a result, it becomes possible to apply it to lean limit control of internal combustion engines.

[Brief explanation of the drawing]

Figure 1 is a waveform diagram showing the behavior of engine speed when road surface conditions change, Figure 2 is a 4-cycle waveform diagram,
Waveform diagrams showing changes over time in the rotational speed of a 4-cylinder engine, Figures 3 and 4 show the amount of rotational speed change at the time of explosion △
Figures 5 and 6 are characteristic diagrams showing the relationship between N and indicated mean effective pressure Pi. ) Characteristic diagrams showing the relationship between 7th and 8th
9 is a waveform diagram showing the situation measured by the conventional method and the method of the present invention when the road surface condition changes. FIG. FIG. 10 is a diagram showing the configuration of the calculation unit in the device shown in FIG. FIG. 10 is a flowchart showing the calculation program of the calculation unit; FIG. 13 is a diagram in which FIGS. 3 and 4 are summarized by rotation speed; FIG. is a flowchart showing an arithmetic program of still another embodiment. E...Engine, 1...Crankshaft, 2
... Pulley, 3 ... Angle signal plate, 4 ... Angle signal sensor, 5 ... Distributor, 6 ... Cylinder discrimination sensor, 7 ... Arithmetic unit, 71A, 71
B... Waveform shaping circuit, 72... Counting circuit, 73...
...Reading circuit, 74...Calculation circuit, 75...D/
A conversion circuit.

Claims (1)

[Claims] 1. Detecting the rotational speed of the internal engine, the rotational speed N _n detected at at least two predetermined crank angle positions during the explosion stroke during the period,
An internal combustion engine that calculates a rotation speed change amount ΔN based on the deviation of N _o , corrects the rotation speed change amount ΔN to a larger value as the rotation speed increases, and measures the corrected rotation speed change amount as an output fluctuation of the engine. Method for measuring engine output fluctuations. 2. The corrected rotational speed change amount is a value of a predetermined cylinder and successive explosion strokes, and the value ΔN' obtained by successive comparison of the values is measured as the output fluctuation. Method described. 3. The method according to claim 1, wherein the corrected rotational speed variation is accumulated for a certain period of time or in a certain number, the accumulated values are statistically processed to obtain a standard deviation, and the standard deviation is measured as an output fluctuation. Method.