JP2853334B2 - Misfire detection method due to crankshaft rotation fluctuation - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a method for detecting misfire due to crankshaft rotation fluctuation, and more particularly, to removing a detection error caused by a structural error of a crank angle sensor in detecting crankshaft rotation speed fluctuation. Further, the present invention relates to a misfire detection method capable of accurately detecting the occurrence of misfire.

2. Description of the Related Art During an operation of an internal combustion engine, if a misfire state occurs in which combustion in a cylinder is not performed normally due to a failure of a fuel injection device or the like, exhaust gas characteristics of the internal combustion engine deteriorate. Therefore, as disclosed in Japanese Patent Application Laid-Open No. 2-30954, information corresponding to the number of revolutions is calculated based on a cycle for each predetermined crank angle corresponding to each cylinder of the engine, and a change amount or a change rate of this information is calculated. The engine misfire state is detected based on the or,
According to the description of Japanese Patent Application Laid-Open No. 2-49955, the deviation between the rotational angular velocity of the internal combustion engine and the rotational angular velocity before one ignition as a reference angular velocity is calculated for each ignition interval in accordance with the combustion process in the internal combustion engine. In other words, if misfire detection is performed based on rotation fluctuations, if accidental misfires occur or misfires occur about once every several revolutions, misfires may not be detected accurately. Therefore, in Japanese Patent Application Laid-Open No. 2-49955, the reference angular velocity is updated as necessary.

As described above, the crank angle sensor is used to detect the cycle of the predetermined crank angle. The crank angle sensor typically has a plurality of vanes arranged at equal angular intervals, i.e., a rotating member having projections and rotatably mounted integrally on a crankshaft, and a vane arranged facing the rotating member and passing the vanes. And a detection unit for detecting The vanes of the rotating member are formed so as to protrude in the radial direction on the periphery of the rotating member, and are provided in a number corresponding to the number of cylinders of the engine. For example, a crank angle sensor for a six-cylinder engine has three vanes. ing. In the crank angle sensor having such a configuration, when the end of one of the adjacent vanes passes through the detection unit with the rotation of the engine, the crankshaft rushes into the crankshaft rotation angle region corresponding to the vane. When the end of the other vane passes, the separation from the area is detected, whereby the time interval, that is, the period from the point of entry into the angle area to the point of separation from the area is detected. It has become so. Further, the crankshaft rotation speed in the crankshaft rotation angle region is calculated based on the detection cycle, and then the presence or absence of a misfire is determined from the magnitude of the rotation speed fluctuation.

In such a misfire detection method, the accuracy in the cycle detection, that is, the misfire detection accuracy depends on the vane angle interval in the crank angle sensor, that is, the vane circumferential length. On the other hand,
A certain degree of error in the configuration of the crank angle sensor, in particular, an error in the manufacture and installation of the vane, is unavoidable. If the vane angle interval is larger than the design value, it takes time to rotate the crankshaft in this angle region, and therefore, it is erroneously determined that the crankshaft rotation speed has decreased in the angle region, and erroneous misfire occurs. May be detected. Such erroneous detection cannot be eliminated by the technique for updating the reference angular velocity disclosed in Japanese Patent Application Laid-Open No. 2-49955.

DISCLOSURE OF THE INVENTION An object of the present invention is to eliminate the influence of a detection error caused by an error in the configuration of a crank angle sensor in detecting a crankshaft rotation fluctuation, and to accurately detect the occurrence of a misfire. To provide a method for detecting a misfire.

To achieve the above object, according to the present invention, it is detected by a crank angle sensor. A time interval from a point of entry into a crankshaft rotation angle region corresponding to a specific stroke phase of each cylinder of the internal combustion engine to a point of departure from the angle region is sequentially detected, and occurrence of misfire is detected based on the time interval. A misfire detection method is provided.

The method includes: a step of calculating a correction coefficient for compensating a configuration error of the crank angle sensor based on the time interval; and a step of correcting internal combustion engine rotation information based on the detection time interval by a correction coefficient. Detecting a misfire based on the corrected internal combustion engine rotation information.

Preferably, the internal combustion engine rotation information is an average angular acceleration of the crankshaft. More preferably, the correction coefficient is repeatedly calculated based on the time intervals sequentially detected during the operation of the internal combustion engine to update the correction coefficient, and the average angular velocity of the crankshaft is sequentially determined based on the time interval. Then, the average angular acceleration is sequentially calculated based on the time interval, the average angular velocity, and the correction coefficient.

Preferably, the update of the correction coefficient is prohibited when the internal combustion engine is operated in a specific operation state in which the load of the internal combustion engine changes suddenly, or when a predetermined time has not elapsed since the start of the internal combustion engine. More preferably, at least when the internal combustion engine is being decelerated, when a shift operation is being performed by a transmission connected to the internal combustion engine, and when a vehicle equipped with the internal combustion engine is traveling on a rough road. One of the states is determined to be the specific operation state.

Preferably, the predetermined operating parameter of the internal combustion engine is, for each of a plurality of operating regions separated from each other by:
Each independent correction coefficient is calculated and the correction coefficient for the current operation area is updated each time the current operation area is determined, while updating of the correction coefficient for operation areas other than the current operation area is prohibited. I do. More preferably, the predetermined operating parameter is a rotational speed of the internal combustion engine or a volumetric efficiency of the internal combustion engine. The plurality of operating regions are separated from each other by the rotational speed and the volumetric efficiency of the internal combustion engine as predetermined operating parameters.

Preferably, when the calculated correction coefficient deviates from a predetermined allowable range, a maximum value or a minimum value corresponding to the allowable range is used as the correction coefficient instead of the calculated correction coefficient. More preferably, the upper limit value and the lower limit value of the allowable range correspond to the maximum allowable structural error of the crank angle sensor. or,
The calculated correction coefficient is stored in a nonvolatile memory, and the correction coefficient read from the nonvolatile memory when the internal combustion engine is started is used as an initial value of the correction coefficient.

Preferably, the misfire detecting method includes: a rotating body that rotates together with the crankshaft; a plurality of identification means provided on the rotating body at intervals in a circumferential direction to respectively identify a point of entry and a point of departure; The present invention is implemented using a crank angle sensor including a detection unit provided on the fixed side member of the engine and emitting a detection signal each time the identification unit approaches. A plurality of identification means of the crank angle sensor are provided for each cylinder.
And second identification means. The detection of the time interval is performed by determining an interval between a first detection signal generated by the first identification unit in proximity to the detection unit and a second detection signal generated by the second identification unit in proximity to the detection unit. It is performed by timing.
Further, the calculation of the correction coefficient is performed based on the time interval and the rotation cycle of the crankshaft including at least one of the generation points of the first and second detection signals used for detecting the time interval. More preferably, the misfire detection method is applied to a multi-cylinder internal combustion engine in which a plurality of cylinders can sequentially go through an explosion stroke at equal intervals during crankshaft rotation. In this case, the crank angle sensor is provided such that the second identification means for each cylinder functions as the first identification means for the next cylinder that goes through an explosion stroke after each cylinder.

Preferably, during the operation of the internal combustion engine, the instantaneous correction coefficient information sequentially obtained based on the detection time interval and the crankshaft rotation period is smoothed, and the internal combustion engine rotation information is corrected with the smoothed value. For example, the expression KLm (n) = a · KLm (n−
1) + (1-a) · KLm and equation KLm = {A · Tm (n)}
The latest correction coefficient Klm (n) is calculated according to / T (n). Here, a (0 ≦ a ≦ 1) is a weighting coefficient, KLm
(N-1) represents the correction coefficient calculated last time. A is the number of identification means provided on the rotating body, Tm (n)
Represents the latest detection time interval, and T (n) represents the latest rotation cycle of the crankshaft.

The misfire detection method of the present invention having the above-described features has the following advantages.

According to the present invention, the occurrence of a misfire is detected based on the internal combustion engine rotation information corrected by a correction coefficient for compensating for a structural error in the crank angle sensor. This correction coefficient is used to determine the error of the crank angle sensor, for example, vane manufacturing,
This reflects the mounting error, and by using this to correct the internal combustion engine rotation information, it is possible to remove the error of the crank angle sensor. As a result, a misfire detection error caused by an error in the configuration of the crank angle sensor can be removed, and therefore, the presence / absence of misfire can be accurately detected.

In a specific embodiment of the present invention, the average angular acceleration of the crankshaft is determined as the internal combustion engine rotation information. The angular acceleration of the crankshaft changes in response to a change in the torque output of the internal combustion engine, and if the average is taken, detection errors generated when detecting the individual angular accelerations can be canceled out. Therefore, the misfire detection based on the crankshaft average angular acceleration is highly responsive and reliable.

In another aspect of the present invention using a crank angle sensor including a plurality of identification units including first and second identification units for each cylinder and a detection unit that emits a detection signal each time the detection unit approaches each of the cylinders, The time interval is detected by measuring the interval between the first and second detection signals generated when the first and second identification means approach the detection unit. The detection time interval accurately represents the actual crankshaft rotation state and reflects an error in the configuration of the crank angle sensor.

In this aspect, next, a correction coefficient is determined based on the detection time interval and the crankshaft rotation cycle including at least one of the generation points of the first and second detection signals used for the detection. If the crankshaft is rotating at a constant speed, the crankshaft rotation cycle corresponds to a time interval detected by a crank angle sensor having no structural error. Therefore, for example, the ratio between the crankshaft rotation cycle and the time interval is corrected based on the crankshaft rotation cycle corresponding to the time interval in which the influence of the sensor error is removed and the time interval reflecting the actually detected sensor error. When obtained as a coefficient, the correction coefficient accurately represents the degree of the sensor error.

The crankshaft rotation cycle is obtained, for example, by summing time intervals sequentially detected during one revolution of the crankshaft. In other words, the crankshaft rotation cycle and thus the correction coefficient can be obtained during the operation of the internal combustion engine based only on the output of the crank angle sensor. Therefore, it is possible to omit the effort of storing sensor error data actually measured before the crank angle sensor is mounted on the internal combustion engine and eventually on the vehicle, for example, in a control device of the vehicle.

A crank angle sensor for use with a multi-cylinder internal combustion engine to which a specific aspect of the present invention is applied is characterized in that the second identification means for each cylinder uses the first cylinder for the next cylinder which goes through an explosion stroke after each cylinder. It can be configured to function as identification means. In this case, the time intervals respectively associated with the plurality of cylinders can be sequentially detected by a relatively small number of identification means. As a result, the configuration of the crank angle sensor can be simplified.

According to another aspect of the present invention, instantaneous correction coefficient information sequentially obtained based on the detection time interval and the rotation period during operation of the internal combustion engine is smoothed. As a result, errors in obtaining individual correction coefficient information, for example, detection errors in individual time intervals are canceled. Therefore, the misfire detection based on the internal combustion engine rotation information corrected by the smoothed value has excellent detection accuracy.

In a specific aspect of the present invention, when the internal combustion engine is operated in a specific operating state such that the load of the internal combustion engine changes suddenly,
Alternatively, when the predetermined time has not elapsed since the start of the internal combustion engine, updating of the correction coefficient is prohibited. Immediately after a specific operating state or starting of the internal combustion engine, the crankshaft is not in a constant speed rotation state, and a correction coefficient that accurately represents a sensor error, for example,
This is because it cannot be obtained from the time interval and the crankshaft rotation cycle respectively detected by the crank angle sensor. By prohibiting the update of the correction coefficient immediately after the specific operating state of the internal combustion engine or immediately after starting, such an error in the calculation of the correction coefficient can be prevented from occurring, and erroneous determination in misfire detection can be prevented.

A specific aspect of the present invention in which a correction coefficient independent of each other is calculated for each of a plurality of operating regions separated from each other by a predetermined operating parameter of the internal combustion engine is a correction coefficient common to all operating regions of the internal combustion engine. It is applied when it is inappropriate to perform misfire detection by using. In this aspect, the correction coefficient for the current operation area is updated each time the current operation area is determined, while the update of the correction coefficient for the operation area other than the current operation area is prohibited. Thus, the correction coefficient becomes more suitable for the operating range of the internal combustion engine, that is, the operating state. As a result, the error of the crank angle sensor can be more appropriately removed, and the misfire detection accuracy is improved.

Further, when the calculated correction coefficient deviates from, for example, an allowable range corresponding to the maximum allowable error in the configuration of the crank angle sensor, a specific value of the present invention using the maximum value or the minimum value of the correction coefficient instead of the calculated correction coefficient is used. According to the aspect, some cause,
For example, even when the correction coefficient cannot be properly calculated due to an error in the detection of the time interval, the adverse effect due to the error in the correction coefficient calculation can be removed, and the erroneous determination in the misfire detection can be prevented.

In the specific aspect of the present invention in which the calculated correction coefficient is stored in the nonvolatile memory, the correction coefficient read from the nonvolatile memory when the internal combustion engine is started is used as its initial value, so that the correction coefficient can be more quickly optimized. That is, in order to obtain a correction coefficient capable of appropriately correcting the error of the crank angle sensor, it is generally necessary to repeat the calculation of the correction coefficient to some extent. In this aspect, an appropriate correction coefficient reflecting the result of the repeatedly executed calculation is always stored in the memory.Therefore, when the internal combustion engine is started, this is read from the memory and used as an initial value of the correction coefficient. Immediately after the start of the internal combustion engine, misfire detection can be properly performed.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic block diagram showing a detection device for implementing a misfire detection method of the present invention, FIG. 2 is a perspective view showing a crank angle sensor of the device shown in FIG. 1, FIG. 1 is a flowchart showing a misfire detection process according to a first embodiment of the present invention, which is executed by the controller of FIG. 1; FIG. 4 is a flowchart showing a part of a misfire detection process in a misfire detection method according to a second embodiment of the present invention; 5 is a flowchart showing another part of the misfire detection process partially shown in FIG. 4, FIG. 6 is a flowchart showing the rest of the misfire detection process partially shown in FIGS. 4 and 5, and FIG. 8, a graph illustrating a change in angular acceleration in a deceleration driving state, FIG. 8 is a graph illustrating a change in angular acceleration during a shift operation, FIG. 9 is a graph illustrating a change in angular acceleration during rough road traveling, FIG. FIG. 11 is a diagram showing an example of setting an engine operation area related to the method of the second embodiment; FIG. 11 is a flowchart showing a part of a misfire detection process in a misfire detection method according to a third embodiment of the present invention; 13 is a flowchart showing a part of the misfire detection process partially shown in FIG. 13, FIG. 13 is a flowchart showing a part of the misfire detection process in the misfire detection method according to the fourth embodiment of the present invention, and FIG. FIG. 15 is a flowchart showing another part of the misfire detection process, and FIG. 15 is a flowchart showing the remaining part of the misfire detection process partially shown in FIGS.

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, a method for detecting misfire due to crankshaft rotation fluctuation according to a first embodiment of the present invention will be described with reference to FIGS.

The apparatus for carrying out the misfire detection method of the present embodiment is mounted on a multi-cylinder internal combustion engine, for example, a six-cylinder engine (not shown), in which a plurality of cylinders can sequentially go through an explosion stroke at equal intervals during crankshaft rotation. As shown in FIG. 1, a controller 10, a crank angle sensor 20, and a cylinder discrimination sensor 30
As a major element.

Referring to FIG. 2, the crank angle sensor 20 includes a rotating member 21 as a rotating body that rotates integrally with the crankshaft 1 of the engine, and a fixed member (not shown) of the engine disposed facing the rotating member 21. A rotating member having a detecting unit 22 provided
The first, second, and third vanes 21a, 21b, and 21c that protrude in the radial direction of the crankshaft are formed as identification means on the periphery of 21, and the detection unit 22 optically detects the passage of the vanes 21a, 21b, or 21c. Alternatively, when electromagnetically detected, a pulse output is generated as a detection signal. 1st to 3rd
The vanes 21a, 21b, and 21c each have a circumferential length corresponding to a crankshaft rotation angle of a certain angle, and are arranged at a predetermined angular interval in a circumferential direction, and thus are adjacent to each other. The angular spacing between corresponding ends of the vanes (first and second identification means for each cylinder) is 120 degrees. However, in practice, due to errors in the configuration of the crank angle sensor 20, particularly the manufacturing and mounting errors of the vanes 21a, 21b and 21c, the angular interval between the ends of adjacent vanes is not exactly
It is not always 120 degrees, and generally there is an angular interval error of about 1 degree or less.

The cylinder discriminating sensor 30 is mounted on a camshaft (not shown) so as to be integrally rotatable therewith. During a rotation of the crankshaft 1 and a rotation of the camshaft, the camshaft rotates at a specific rotational position corresponding to one cylinder. , A pulse output is generated every time.

The controller 10 functions as a main element of the misfire detection device and executes normal various engine controls, and a processor 11 for executing various control programs.
And a read-only memory 12 storing the control program
And a random access memory 13 for temporary storage of data and the like. The memory 13 has a nonvolatile memory area backed up by a battery (not shown). The processor 11 receives the crank angle sensor 2 via the input circuit 14.
0, a cylinder discrimination sensor 30, an ignition switch 40, various sensors and switches (partially omitted) such as an intake air amount sensor, an intake air temperature sensor, a water temperature sensor, and an output circuit.
15, various actuators including the fuel injection valve 50,
Various driving circuits (element 5
Only those corresponding to 0,60 are indicated by reference numerals 51,61). In FIG. 1, reference numeral 70 denotes a throttle position sensor used in another embodiment described later, and the sensor 70 is connected to the processor 1 via the input circuit 14.

The device of the present embodiment mounted on a six-cylinder engine in which the ignition operation is performed in the order of the cylinder number is, for example, such that the end of the third vane 21c (the front end 21c 'or the rear end, which forms the first identification means) is provided. When passing through the detection unit 22, the first cylinder corresponding to one of the first cylinder and the fourth cylinder (preferably, mainly the explosion stroke in the one cylinder) of the first cylinder group
The crankshaft protrudes into the crankshaft rotation angle region, and the end (second identification means) of the first vane 21a is
When the crankshaft passes through the first rotation angle region. Similarly, when passing through the end of the first vane 21a, the second and fifth cylinders forming the second cylinder group
The cylinder enters the second crankshaft rotation angle region corresponding to one of the cylinders and separates from the region when passing through the end of the second vane 21b.
The cylinder enters the third crankshaft rotation angle region corresponding to one of the third and sixth cylinders forming the cylinder group, and is separated from the region when the end of the third vane 21c passes. I have. The identification of the first cylinder and the fourth cylinder, the identification of the second cylinder and the fifth cylinder, and the identification of the third cylinder and the sixth cylinder are performed based on the output of the cylinder determination sensor 30. Then, the detection unit 22 generates a first detection signal when the first identification unit (for example, the end of the third vane 21c) approaches the detection unit 22, and the second identification unit (for example, the first vane 21c).
The second detection signal is generated when the corresponding end 21a approaches the detection unit. In the present embodiment applied to a six-cylinder engine in which a plurality of cylinders sequentially go through an explosion stroke at equal intervals during rotation of the crankshaft, the second identification means for each cylinder sets the next explosion stroke after the cylinder. Function as the first identification means for the cylinder.

 Hereinafter, the operation of the misfire detection device having the above configuration will be described.

While the engine is running, the processor 11
While sequentially receiving the pulse output from the cylinder determination sensor 30 and the pulse output from the cylinder discrimination sensor 30, the misfire detection process shown in FIG.

The processor 11 starts the misfire detection process every time the pulse output of the crank angle sensor 20 is input.

In each detection cycle, the processor 11 first determines the order of the crank angle sensor pulse output among the crank angle sensor pulse outputs sequentially input after the input time of the pulse output from the cylinder determination sensor 30. I do. As a result, the number of the cylinder corresponding to the input crank angle sensor pulse output is identified (step S1). Preferably, the cylinder that is currently mainly performing the explosion stroke (output stroke) at this time is identified as the identification cylinder.

The processor 11 enters the crankshaft rotation angle region corresponding to the identified cylinder group m (m is 1, 2 or 3) in response to the input of the pulse output (first detection signal) of the crank angle sensor 20. Is determined, the timer for period measurement (not shown) is restarted. The identification cylinder group m includes the cylinder identified in step S1.

When the next pulse output (second detection signal) is input from the crank angle sensor 20, the processor 11 determines the departure from the crankshaft rotation angle region corresponding to the identified cylinder group m, and counts the time of the period measurement timer. Is stopped and the timing result is read (step S2). The timing result is a time interval Tm from the point of entry into the crankshaft rotation angle region corresponding to the identified cylinder group m to the point of departure from the region.
(N), that is, a period Tm (n) determined by two predetermined crank angles corresponding to the identified cylinder group. Here, the suffix n in the cycle Tm (n) indicates that the cycle corresponds to the n-th (current) ignition operation in the identified cylinder. Further, the cycle Tm (n) is a cycle between the 120-degree crank angles of the identified cylinder group in the case of the six-cylinder engine, and more generally a cycle between the (720 / N) degrees crank angle of the N-cylinder engine.

Note that the pulse output indicating the departure from the crankshaft rotation angle region corresponding to the current identification cylinder also indicates a rush into the crankshaft rotation angle region corresponding to the next identification cylinder.
Accordingly, in response to this pulse output, the cycle is executed in the cylinder identification step S1 for the next identified cylinder, and the cycle measurement timer is restarted to start the cycle measurement for the next identified cylinder.

Then, in connection with the current identification cylinder, the processor 11
Calculates the correction coefficient KLm (n) related to the identified cylinder group m in order to eliminate the cycle measurement error due to the variation of the vane angle interval in the vane manufacturing, by the equation KLm (n) = a · KLm (n
-1) + (1-a) · Calculated according to KLm (step
S3). That is, the correction coefficient is updated (learned). Here, the symbol a is a filter constant stored in the memory 12 in advance and takes a value of 0 or more and 1 or less. Symbol KLm (n
-1) represents a correction coefficient related to the identified cylinder group m calculated in the previous detection cycle and stored in the memory 13, and KLm is represented by an equation KLm = Tm (n) ÷ (T (n) / 3 ) Is calculated according to the following formula. Here, as described above, the symbol Tm (n) represents the cycle of the currently detected 120-degree crank angle of the discriminating cylinder group m. The symbol T (n) is
Sum of the cycles between the 120-degree crank angles of the first to third cylinder groups measured one after another in the two detection cycles and the current detection cycle, that is, the 360-degree crank angle cycle (T (n))
= T1 (n) + T2 (n) + T3 (n)). If the engine speed is constant, the value T (n) / 3 obtained by dividing the 360-degree crank angle cycle by the value 3 is equal to the exact 120-degree crank angle cycle when there is no error in the vane angle interval. Therefore, the calculated value KLm indicates the ratio between the accurate 120-degree crank angle period and the 120-degree crank angle period of the identified cylinder group m.

In other words, the correction coefficient KLm (n) is equal to the detection time interval Tm
(N) and the instantaneous correction coefficient information based on the rotation period T (n) of the crankshaft including the time point at which one or both of the first and second detection signals are used to detect this. Is required. Further, the instantaneous correction coefficient information sequentially obtained based on the detection time interval and the crankshaft rotation cycle is smoothed.

Further, the processor 11 outputs the period Tn (= Tm) between the 120-degree crank angles measured in step S2 of the current detection cycle.
(N)), the average angular velocity ωn (= 120 degrees / Tn) of the crankshaft in the cycle is calculated, and the average angular velocity ωn measured in the previous detection cycle and stored in the memory 13 is calculated.
Read -1. Next, the processor 11 calculates the measurement values Tn, Tn
−1 and the calculated values ωn, ωn−1 and the correction coefficient KLm (n) calculated in step S3,
The average angular acceleration Dω of the crankshaft during the cycle between 0 ° crank angles is calculated by the following equation: Dω = KLm (n) · (ωn−ωn−1) ÷
It is calculated according to {(1/2) · (Tn + Tn−1)} (step S4). Here, the symbol D is a differential operator symbol and represents d / dt. In this manner, the crankshaft angular acceleration is obtained based on the measurement cycle corrected using the correction coefficient KLm (n).

Next, the processor 11 detects the occurrence of a misfire in the identified cylinder by using the average angular acceleration Dωn calculated in step S4 and representing the rotation fluctuation of the crankshaft and a determination value for misfire determination stored in the memory 12 in advance. Is determined (step S5). Note that the determination value is set to a negative value. When it is determined that the calculated value Dωn is smaller than the determination value,
The processor 11 sends a drive signal of, for example, H level to the lamp drive circuit 61 to turn on the warning lamp 60, thereby warning that a misfire has occurred in the identified cylinder (Step S).
6) Further, the fact that misfire has occurred in the identified cylinder determined in step S1 is stored in the memory 13 (step S7). On the other hand, if it is determined in step S5 that the average angular acceleration Dωn of the crankshaft is equal to or greater than the determination value, the processor 11 determines, for example, L
A level drive signal is sent out to turn off the warning lamp 60 to notify that no misfire has occurred in the identified cylinder (step S8). The misfire detection based on the corrected cycle is accurate without being affected by the vane angle interval error.

After the storage of the misfiring cylinder in step 7 or the turning off of the warning lamp in step S8, the processor 11 waits for the input of the next pulse output from the crank angle sensor 20. When the pulse output is input, the processor 11 executes the processing in FIG. To resume.

Hereinafter, a misfire detection method according to a second embodiment of the present invention will be described with reference to FIGS.

This embodiment has a main feature of prohibiting the calculation and updating (learning) of the correction coefficient when the engine is operating in an operating state other than the operating state in which the correction coefficient can be properly calculated. When the correction coefficient deviates from the allowable range, a suitable value is used instead of the calculated value, and a stored value of the correction coefficient is used as an initial value of the correction coefficient when the engine is started. The method of the present embodiment can be implemented by the apparatus shown in FIGS. 1 and 2 according to the first embodiment, and therefore, the description of the apparatus will be omitted.

While the engine is running, the processor 11
While sequentially receiving the pulse output from the cylinder 20 and the pulse output from the cylinder discrimination sensor 30, the misfire detection processing shown in FIGS. 4 to 6 is periodically and repeatedly executed. That is, the processor 11 starts the misfire detection processing cycle every time a pulse output indicating the entry into the crankshaft rotation angle region corresponding to the current detection cycle is input from the crank angle sensor 20. In each detection cycle, the processor 11 sequentially executes the cylinder identification and the cycle measurement between the predetermined crank angles in steps S11 and S12 corresponding to steps S1 and S2 in FIG. 3, respectively. Next, the processor 11 determines whether the engine has been started or immediately after (step S1).
3). Then, at or immediately after the start of the engine, the processor 11 prohibits the calculation and update of the correction coefficient in the current detection cycle, and thus the misfire detection.

Explaining the reason, a correction coefficient calculation step S20 described later corresponding to step S3 in FIG. 3 according to the first embodiment will be described.
In FIG. 5, the sum of the periods between the 120 ° crank angles of the first to third cylinder groups, that is, the value obtained by dividing the period between the 360 ° crank angles by 3 is used as the accurate period between the 120 ° crank angles. However, in order to use a value obtained by dividing the sum of the three 120-degree crank angle periods by three as an accurate 120-degree crank angle period, the engine speed must be substantially constant. If the engine speed changes suddenly, such as at or immediately after the start of the engine, such a requirement is not satisfied. In such a case, in this embodiment, calculation and update of the correction coefficient and detection of misfire are prohibited.

On the other hand, if it is determined in step S13 that the ignition key 40 is not at or immediately after the start of the engine, the processor 11 determines whether or not the ignition key 40 is turned off or immediately after (step S14). Immediately after the ignition key is turned off, the engine is coasted for a while, and the controller 10 and various sensors are not deactivated. During coasting of the engine, the engine speed is rapidly decreasing,
The correction coefficient cannot be obtained accurately, and therefore, an appropriate misfire detection cannot be performed. Therefore, in the present embodiment, when the ignition switch is turned off or immediately thereafter, calculation and updating of the correction coefficient and misfire detection are prohibited.

Further, the processor 11 determines whether or not the engine is in a decelerating operation (step S15). If the engine is in a decelerating operation, the processing in the current cycle is immediately terminated, and is shown as a section A in FIG. The calculation and update of the correction coefficient and the misfire detection in the deceleration operation state in which the angular acceleration (corresponding to the average angular acceleration Dω of the present embodiment) largely swings to the negative side and errors in the calculation of the correction coefficient and the misfire detection are likely to occur. Ban. On the other hand, if it is determined that the vehicle is not in the deceleration operation, the processor 11 determines whether or not the gear is being shifted by a transmission (not shown) connected to the engine (step S16).

If the shift operation is being performed, the processor 11 ends the process in the current cycle, and the angular acceleration greatly swings to the negative side, which is indicated by the section B in FIG. 8, and errors in calculating the correction coefficient and detecting misfire are likely to occur. The correction coefficient calculation, update, and misfire detection during the clutch meeting by the manual transmission (more generally, during the shifting operation of the transmission) are prohibited. FIG.
Middle, section C represents a deceleration operation state before the clutch meet. On the other hand, when determining that the vehicle is not shifting, the processor 11 further violates whether or not the vehicle equipped with the engine is running on a rough road (step S17).

If the vehicle is traveling on a rough road, the processor 11 terminates the processing in this cycle, and as shown in FIG. The calculation and update of the correction coefficient and the misfire detection are prohibited. In FIG. 9, when the triangle mark is attached, the angular acceleration is greatly reduced due to the misfire actually occurred.

On the other hand, if the vehicle is not traveling on a rough road, and accordingly, the determination results in steps S13 to S17 are all negative, the processor 11 calculates the volume efficiency ηv and the engine speed N as the predetermined operation parameters, and calculates Based on the values, the current engine operating region is divided into nine operating regions I, which are separated from each other by ηv, N as illustrated in FIG.
IX or IX (step
S18). Next, the processor 11 determines whether the determination that a misfire has occurred is currently made or immediately after the misfire determination (step S19).

Then, when it is determined in step S19 that it is not at the time of or immediately after the misfire determination, the processor 11 determines in step S20 corresponding to step S3 in FIG. 3 that the identification cylinder group m and the current engine operating area i (= I, II, .. Or IX), the correction coefficient KLmi (n) is calculated by the equation KLmi.
(N) = a · KLmi (n−1) + (1−a) · Calculated according to KLmi. In the present embodiment, as indicated by the subscript i, 9
For each of the two engine operating regions, independent correction factors KLmi are used.

Next, the processor 11 previously stores the correction coefficient KLmi (n) calculated in step S20 and the maximum allowable error in the configuration of the crank angle sensor in the nonvolatile memory area of the read only memory 12 or the random access memory 13. Compare the magnitude with the stored set value (Step S2
1). This set value is set in advance as an allowable maximum value of the correction coefficient in light of the maximum allowable values such as a vane manufacturing error, a vane mounting error, a crank angle sensor mounting error, and a crank angle sensor output error in device design and manufacturing. You. In connection with the vane angular spacing, the correction factor for compensating for the vane angular spacing error is a maximum when the vane angular spacing error is equal to the maximum allowable value, and the maximum allowable vane angular spacing error is the maximum allowable vane angular error. Once the manufacturing error and the maximum allowable vane mounting error are determined, they are automatically determined.
As a result, if the maximum allowable vane manufacturing error and the maximum allowable vane mounting error are determined, the maximum value of the correction coefficient, that is, the set value can be determined.

If the processor 11 determines that the correction coefficient KLmi (n) calculated in step S20 is larger than the set value set from the above viewpoint, the processor 11 sets the set value as the correction coefficient KLmi (n) instead of the calculated correction coefficient ( Step S22). That is, when the calculated value of the correction coefficient KLmi (n) exceeds a set value equal to the upper limit value of the correction coefficient, step S12
It is determined that the correction coefficient could not be properly calculated due to an error in the period measurement or the like, and the use of the calculated correction coefficient is forcibly prohibited.

Subsequent to step S22, or the calculation correction coefficient KLmi
When it is determined in step S21 that (n) is equal to or less than the set value and the correction coefficient has been properly calculated, the processor 11 stores the correction coefficient KLmi (n) calculated in step S20 or set in step S22 in the random access memory 13. (Step S23), whereby the correction coefficient KLmi (n) for the current engine operation area is stored.
Only update. That is, the update of the correction coefficient in an operation region other than the current engine operation region is prohibited.

Next, the processor 11 calculates the average angular acceleration D of the crankshaft in the cycle between the 120 ° crank angles of the current detection cycle.
ωn is calculated by the formula Dωn = KLmi (n) · (ωn−ωn−1)
It is calculated according to {(1/2) · (Tn + Tn−1)} (step S24).

If it is determined in step S19 that it is at or immediately after the misfire determination, the processor 11 determines that the engine rotation has fluctuated due to the occurrence of the misfire, and does not execute the correction coefficient calculation step S20 or the like. And immediately execute the average angular acceleration calculation step S24. in this case,
The processor calculates the average angular acceleration Dωn using the latest correction coefficient KLmi detected in the previous or previous detection cycle and stored in the nonvolatile memory area of the memory 13.
Is calculated. Note that the correction coefficient KLmi used in the angular acceleration calculation is related to the identified cylinder group m and the current engine operation area i.

Next, in step S25 corresponding to step S5 in FIG. 3, the processor 11 stores the average angular acceleration Dωn and the memory 12
Is determined with respect to a discrimination reference value for misfire determination stored in advance. Further, according to the determination result, the processor
11 is a warning lamp lighting step S26 and a misfire cylinder storage step S27 corresponding to steps S6 and S7 of FIG. 3, or a warning lamp extinguishing step S28 corresponding to step S8.
To end the process in the current cycle.

During operation of the engine, the correction coefficient is repeatedly calculated and updated, and the latest calculated correction coefficient is stored in the nonvolatile memory area of the memory 13. Thereafter, when the engine is once stopped and then restarted, the engine is restarted (however,
At the time of the execution of the cycle immediately after the determination result in step S13 is negative), the processor 11 uses the correction coefficient read from the memory 13 as an initial value of the correction coefficient in calculating the correction coefficient, thereby starting the engine. An appropriate correction coefficient can be obtained even in the later initial stage.

Hereinafter, a misfire detection method according to a third embodiment of the present invention will be described with reference to FIGS.

This embodiment has a main feature in the rough road determination method, and stores a point that a common correction coefficient is used for all engine operation regions, a point that the magnitude of the calculated correction coefficient is not restricted, and a calculated correction coefficient. This is different from the above-described second embodiment in that there is no point. The method of this embodiment can be performed by the apparatus shown in FIGS. 1 and 2, and therefore, the description of the apparatus will be omitted.

In each cycle of the misfire detection processing (FIGS. 11 and 12) that is repeatedly executed, the processor 11 executes the processing in step S1 of FIG.
The cylinder identification step S101 and the cycle measurement step S102 respectively corresponding to 1 and S12 are sequentially executed. Next, processor 11
Is an elapsed time TI obtained by referring to a first timer (not shown) for measuring an elapsed time TIM1 from the engine start in order to determine whether or not the engine is started or immediately thereafter.
M1 is compared with the first set value (step S103). This set value is set to a value (10 seconds to 30 seconds) slightly longer than the time normally required for the engine to reach a steady rotation state from the start of the engine. It should be noted that the first timer is operated by an ignition key 40 for starting the engine.
Is switched from the off position to the on position (when the engine is started), it is restarted under the control of the processor 11 (for the sake of simplicity, illustration of a control procedure related thereto is omitted).

If it is determined in step S103 that the elapsed time TIM1 is not less than the set value and it is not immediately after the engine is started, the processor 11
In order to determine whether the ignition key is turned off or immediately thereafter, a second timer (not shown) that measures an elapsed time TIM2 from the ignition key off and a second elapsed time TIM2 obtained by referring to a second timer are shown. A comparison is made with the set value (step S104). Although not shown, the second timer for measuring the elapsed time TIM2 is restarted when the ignition key 40 is switched from the ON position to the OFF position. The second set value is from when the ignition key is turned off.
It is set to a value slightly larger than the time normally required until the state in which the controller 10 and the like are deactivated and the calculation of the correction coefficient of the controller and the like and the update and the misfire detection operation are not performed when the engine rotation is stopped. I have.

In step S103 above, the elapsed time TI from the start of the engine
It is determined that M1 is smaller than the first set value and immediately after the engine is started, or it is determined in step S104 that the elapsed time TIM2 from the time of ignition key off is smaller than the second set value and immediately after the ignition key is turned off. Then, the processing in this cycle is immediately terminated. As a result, calculation, updating and misfire detection of the correction coefficient at the time of starting the engine and the like are prohibited.

On the other hand, if it is determined in step S104 that the elapsed time TIM2 is not less than the second set value and not immediately after the ignition key is turned off after the determination in step S103 that the engine is not immediately after the engine start, the processor 11 determines whether the airflow sensor and the intake air First, an engine load state is detected based on detection outputs from a pressure sensor, a throttle sensor, and the like. In this embodiment,
The engine volume efficiency ηv representing the engine load state is calculated. When an airflow sensor is used to detect the engine load state, the processor 1 obtains the intake air amount A / N per intake stroke based on the output of the airflow sensor and the output of the engine speed sensor. By dividing by the fully open A / N, a dimensionless equivalent volume efficiency value is obtained. Next, the processor 11 compares the volumetric efficiency transferred as described above with a third set value (for example, 18%) corresponding to the volumetric efficiency indicating the deceleration operation state of the engine, and determines whether the engine is in the deceleration operation state. Is determined (step S105). Then, when it is determined in step S105 that the calculated value of the volumetric efficiency ηv is equal to or greater than the third set value and the engine is not in the deceleration operation state, the processor 11 obtains the calculated value by referring to a third timer (not shown). The elapsed time TIM3 from the entry into the deceleration operation state is compared with a fourth set value corresponding to the time normally required from the entry into the deceleration operation state to the end of the shift operation (step S106). Then, when it is determined in step S106 that the elapsed time TIM3 is equal to or longer than the fourth set value and the shift operation is not being performed, the processor 11 determines at the time of rough road traveling determination obtained with reference to a fourth timer (not shown). TIM4 is compared with a fifth set value representing a prohibition time of correction coefficient calculation, update and misfire detection related to rough road traveling (step S107).

If it is determined in step S107 that the elapsed time TIM4 is equal to or greater than the fifth set value and the prohibition time has already elapsed, the processor
Step 11 sequentially executes steps S108 to S110 corresponding to steps S19, S20 and S24 in FIG. In brief, the processor 11 refers to a fifth timer (not shown) which is restarted when the occurrence of misfire is determined as described later, and obtains an elapsed time TIM5 from the time of misfire determination. The time TIM5 is compared with a sixth set value representing a correction coefficient calculation and update prohibition time related to the occurrence of a misfire (step S10).
8). Then, when it is determined in step S108 that the elapsed time TIM5 is equal to or greater than the sixth set value and the correction coefficient calculation and update prohibition time have already elapsed, the processor 11 relates to the cylinder group m identified in the current detection cycle. Correction coefficient KLm
i (n) is calculated to update the correction coefficient, and the average angular acceleration Dωn of the crankshaft in the cycle between the 120 ° crank angles in the current detection cycle is calculated (steps S109 and S109).
S110). On the other hand, if it is determined in step S108 that the elapsed time TIM5 has fallen below the sixth set value and the correction coefficient calculation and update prohibition time have not yet elapsed, the processor 11 proceeds to step S109.
Immediately execute step S110 without executing step S110.

If the calculated value of the volumetric efficiency ηv is smaller than the set value and the engine is operated in the deceleration operation state, step S1
When the determination is made in 05, the processor 11 restarts the third timer that measures the elapsed time TIM3 from the time of entering the deceleration operation state (step S111), and sets the flag FRR to indicate that the rough road traveling determination is not being performed. Is reset to a value "0" (step S112), and the process in the current cycle is immediately terminated. That is, in the deceleration operation state, calculation and update of the correction coefficient and misfire detection are prohibited.

If the elapsed time TIM3 from the entry into the deceleration operation state is smaller than the fourth set value and the shift operation is being performed,
When determined in S106, the processor 11 immediately ends the processing in the current cycle, thereby prohibiting the calculation, the update, and the misfire detection during the shift operation. In addition, the elapsed time TI from when it is determined that the vehicle is traveling on a rough road is determined.
If it is determined in step S107 that M4 is smaller than the fifth set value and the prohibition time for the correction coefficient calculation, update, and misfire detection related to rough road running has not elapsed yet, the processor 11 performs the process in the current cycle. Immediately, the correction coefficient calculation, update and misfire detection are prohibited.

On the other hand, if the presence or absence of misfire should be detected following the calculation and update of the correction coefficient, the processor 11 calculates the average angular acceleration Dωn in step S110, and then sets the flag F
It is determined whether or not RR is a value “1” indicating that the rough road traveling determination is being performed (step S113). FRR
≠ 1 That is, if it is not during rough road traveling determination, step S1
When determined in step 13, the processor 11 calculates the calculated angular acceleration Dωn
And a misfire determination level (determination reference value) THMF (<0) (step S114). If the angular acceleration Dωn is equal to or higher than the misfire determination level THMF, the warning lamp 60 is turned off in step S115 corresponding to step S28 in FIG. 6, while the angular acceleration Dωn is smaller than the misfire determination level THMF and the probability of misfire has occurred. Is high in step S114, the processor 11 temporarily stores that a misfire state has occurred for the misfire determination cylinder identified in step S101 (step S116).

Next, rough road travel determination is started to determine whether or not the misfire state determined in step S114 is due to the influence of the vehicle traveling on a rough road. Therefore, the processor 11 restarts the sixth timer that measures the elapsed time TIM6 from the start of the rough road traveling determination (step S117), and further sets the flag FRR to a value indicating that the rough road traveling is being determined. "1" (step S118), and the process in the current cycle ends.

In the misfire detection cycle immediately after the start of the rough road determination, the processor 11 performs the above-described series of steps.
In step S113 following S101 to S110, it is determined that the value of the flag FRR is "1". In this case, the processor 11
The elapsed time TIM6 from the start of the rough road travel determination is compared with a seventh set value representing the required time from the start to the end of the rough road travel determination (step S119). elapsed time
When it is determined in step S119 that TIM6 is smaller than the seventh set value and the rough road traveling determination time has not elapsed yet, the processor 11 determines the average angular acceleration Dωn calculated in step S110 and the rough road determination level THRR (> 0). Is compared with (step S120). If the calculated angular acceleration Dωn is greater than the rough road determination level THRR, the processor 11 determines that the vehicle is traveling on a rough road.

When it is determined in step S120 that the vehicle is traveling on the rough road, the processor 11 resets the flag FRR to a value “0” indicating that the rough road traveling is not being determined (step S121), The fourth timer for measuring the elapsed time TIM4 is restarted (step S122). Further, the provisional misfire determination in step S116 of the previous detection cycle is determined to be due to the influence of rough road traveling, and the misfire determination cylinder is determined. Clear temporary storage (step S1
twenty three). Thus, the processing in the current cycle ends.

On the other hand, during the rough road traveling determination, if the angular acceleration Dωn is equal to or lower than the rough road determination level THRR and it is not determined that the vehicle is traveling on the rough road, the misfire detection process in this cycle is immediately terminated in the present embodiment. . According to such a determination method, when a misfire state occurs during rough road traveling determination, this is ignored. Therefore, if it is not determined in step S120 that the vehicle is traveling on rough road, the presence or absence of the occurrence of a misfire state may be separately determined.

If it is determined in step S119 that the elapsed time TIM6 from the start of the rough road travel determination is equal to or greater than the seventh set value and the rough road travel determination time has already elapsed while repeatedly performing the rough road travel determination, the processor 11 determines After resetting the flag FRR to a value “0” indicating that rough road traveling is not being determined (step S124), in steps S125 and S126 corresponding to steps S26 and S27 in FIG. It is determined that the misfire determination provisionally stored in step S116 of the detection cycle is not due to the influence of rough road traveling, and the misfire determination cylinder is securely stored. Then, the processor 11 restarts the fifth timer that measures the elapsed time TIM5 from the time of the misfire determination (step S127), and ends the processing in the current cycle.

According to the procedure shown in FIG. 11 and FIG. 12, the deceleration operation state, the shift operation, the rough road traveling, and the like can be determined without special hardware, which is convenient.

Hereinafter, a misfire detection method according to a fourth embodiment of the present invention will be described with reference to FIGS.

The misfire detection method according to this embodiment differs from the procedure shown in FIGS. 11 and 12 in that rough road traveling determination is performed based on an output from a throttle position sensor (indicated by reference numeral 70 in FIG. 1). Has the main features. Note that the method of the present embodiment can be performed by the detection device shown in FIGS. 1 and 2, and thus the description of the device will be omitted.

As shown in FIGS. 13 to 15, in each cycle of the misfire detection process in the method of the present embodiment, the processor 10
S201 to S201 corresponding to steps S101 to S104 of 11 respectively
S204 is sequentially executed. Then, it is determined in step S203 that the elapsed time TIM11 obtained by referring to the first timer that measures the elapsed time TIM11 from the start of the engine is less than the first set value, or from the time when the ignition key is turned off. When it is determined in step S204 that the elapsed time TIM12 obtained by referring to the second timer that measures the elapsed time TIM12 is smaller than the second set value, the processor 11 immediately ends the processing in the current cycle, This prevents errors in calculating the correction coefficient and in detecting misfire when the engine is started, the ignition key is turned off, and the like.

On the other hand, if it is determined in steps S203 and S204 that the elapsed times TIM11 and TIM12 are equal to or greater than the set value and the engine is not started, the ignition key is not turned off, and so on, the processor 11
Reads the output TPS of the throttle position sensor 70, reads the throttle position sensor output TPS read in the previous cycle and stored in the memory 13 from the memory, and further reads the throttle position in the current cycle from the throttle position in the previous cycle. , The sensor output change rate ΔTPS representing the throttle position change rate is calculated. Next, in order to determine whether or not the engine is operating in a decelerating state, the processor 11 sets the sensor output change rate ΔTPS and the third set value THTP (THTP <0).
Is compared with (step S205). The determination of the start of deceleration operation is based on the actual load level change rate detected by another engine load state detection sensor such as an air flow sensor or an intake negative pressure sensor instead of the throttle sensor output (change rate ΔTPS). May be used.

In step S205, when it is determined that the throttle position sensor output change rate ΔTPS is smaller than the set value THTP and the engine is continuously rotating in a deceleration state, the processor 11 starts the deceleration operation of the engine or determines the rough road traveling. The third timer for measuring the elapsed time TIM13 from is restarted (step S220 in FIG. 12), and the processing in the current cycle is ended.

On the other hand, when it is determined in step S205 that the sensor output change rate ΔTPS is equal to or larger than the set value THTP and the engine is not in the deceleration operation state, the processor 11 determines whether or not the deceleration operation of the engine is started with reference to the third timer. The elapsed time TIM13 from the rough road running determination is compared with the fourth set value THT13 representing the time for calculating, updating, and preventing misfire detection of the correction coefficient (step S206). The processor 11 calculates the elapsed time TIM1
If 3 is smaller than the value THT13 and the prohibition time has not yet elapsed, the processing in this cycle is immediately terminated, while if the prohibition time has already elapsed, the rough road determination flag FRR
Is compared with the value “0” (step S207).

Then, if the flag FRR is a value “0” indicating that rough road traveling is not being determined, the processor 11 reads the throttle position sensor output TPS read from the previous cycle and the current cycle and stored in the memory 13 from the memory. Then, the sensor output change rate ΔTPS from the previous cycle to the current cycle is calculated. Next, in order to determine whether or not to start the rough load determination, the processor 11 determines whether or not the sign of the sensor output change rate ΔTPS has been inverted, that is, the value ΔTPS has changed from positive to negative or from negative to positive. It is determined whether or not the process has been performed (step S208). Preferably, in order to provide the determination with a hysteresis characteristic, when the sensor output change rate ΔTPS changes over a predetermined threshold or more and its sign changes from positive to negative or from negative to positive, the processor 11 sets the sign of the value ΔTPS. Is determined to have been inverted.

If it is determined in step S208 that the sign of the sensor output change rate ΔTPS has not been inverted, the processor 11 is restarted at the time when misfire occurrence is determined as described below, in order to determine whether correction coefficient calculation is necessary. Elapsed time TIM15 from the misfire determination measured by the fifth timer and a sixth set value THT15 representing the correction coefficient calculation inhibition time from the misfire determination
(Step S209 (corresponding to Step S108 in FIG. 11)). Then, when it is determined in step S209 that the elapsed time TIM15 is equal to or greater than the set value THT15 and the correction coefficient calculation and update prohibition time have already elapsed, the processor 11 proceeds to step S20, step S24 in FIG. 5, and step S25 in FIG. Corresponding correction coefficient calculation step S210, average angular acceleration calculation step S211 and misfire determination step S212 are sequentially executed. On the other hand, if the elapsed time TIM15 is smaller than the set value THT15, the correction coefficient is calculated, and if the update prohibition time has not yet elapsed, the process proceeds to step S209.
Is determined, the processor 11 immediately proceeds to the angular acceleration calculation step S211 without calculating and updating the correction coefficient in step S210.

When it is determined in step S212 that the calculated angular acceleration Dωn in the current cycle is lower than the determination value and misfire has occurred in the cylinder identified in step S201 of the current cycle,
The processor 11 turns on the warning lamp 60 (step S2
13), the discriminating cylinder is stored (step S214), and the fifth timer for measuring the elapsed time TIM15 from the discrimination determination is restarted (step S215), and the process in the current cycle is ended. On the other hand, if it is determined in step S212 that a misfire has not occurred in the identified cylinder, the warning lamp 60 is turned off (step S216), and the processing of the current cycle is ended.

If it is determined in step S208 that the sign of the throttle position sensor output change rate ΔTPS has been inverted, the processor 11 adds a value “1” to the count value COUNT of a rough load determination counter (not shown) to start rough load determination. (Step S217), a fourth timer (not shown) for measuring the elapsed time TIM14 from the start of the rough load determination is restarted (step S218), and the flag FRR is set to a value "1" indicating that rough load determination is being performed. (Step S219), and the process in the current cycle ends.

In a detection cycle started immediately after setting the flag FRR to the value "1", the processor 11 determines that the value of the flag FRR is not "0" in step S207, and determines the elapsed time TIM14 from the start of the rough road determination. A comparison is made with a fifth set value THT14 representing the rough road determination time (step S221). When it is determined in step S221 that the elapsed time TIM14 is less than the set value THT14 and the rough road determination time has not yet elapsed, the processor 11 determines whether the sign of the throttle position sensor output change rate ΔTPS has been inverted. (Step S222) If the sign of the value ΔTPS is not inverted, the processing in the current cycle is terminated, while the value ΔTPS is
If the sign of the TPS is inverted, the value "1" is added to the count value COUNT of the rough load determination counter (step S22).
3) The updated count value COUNT is compared with a seventh set value THRR corresponding to the number of sign inversions of the sensor output change rate ΔTPS indicating that the vehicle is traveling on rough road (step S224).

If the count value COUNT is less than the set value THRR and it cannot be determined that the vehicle is traveling on rough road, step S224
, The processor 11 ends the processing in the current cycle. On the other hand, when it is determined in step S224 that the count value COUNT is equal to or greater than the set value THRR, the processor 11 performs the accelerator pedal depressing operation and the releasing operation alternately a predetermined number of times or more, and accordingly, the vehicle is running on a rough road. Judge that there is. When the rough road traveling determination is performed in this manner, the processor 11 restarts a third timer that measures the elapsed time TIM13 from the rough road traveling determination (step S225), and determines that the flag FRR is not under the rough road traveling determination. The value is reset to “0” (step S22).
6) In addition, the count value COU of the rough load determination counter
NT is reset to a value “0” (step S227).

If it is determined in step S221 that the elapsed time TIM14 from the start of the rough load determination is equal to or greater than the set value THT14 and the rough load determination time has already elapsed, the processor 11
Determines that it is not necessary to continue determining whether or not the vehicle is traveling on a rough road. In this case, processor 11
Resets the flag FRR to a value “0” indicating that the rough load determination is not being performed (step S228), resets the count value COUNT of the rough load determination counter to a value “0” (step S229), and Is completed.

The present invention is not limited to the first to fourth embodiments, and can be variously modified.

For example, in the above embodiment, the case where the present invention is applied to a six-cylinder engine has been described. However, the present invention is applicable to various engines such as a four-cylinder engine.

In the second to fourth embodiments, the calculation and update of the correction coefficient are prohibited during the deceleration operation of the engine. However, this is not essential for implementing the method of the present invention. At the time, both calculation and updating of the correction coefficient may be permitted as in the first embodiment, or only calculation of the correction coefficient may be permitted.

Further, in the second embodiment, for each of the plurality of engine operating regions that are divided in advance, independent correction coefficients are used, the magnitude of the correction coefficient is regulated, and the calculated correction coefficient is used when starting the engine. Are stored in a memory so that they can be reused, but none of these are essential items for implementing the present invention. On the other hand, the first, third, and / or fourth embodiments can be modified so as to use the correction coefficient for each engine operation area, regulate the correction coefficient, and / or store the calculated correction coefficient in the memory. is there. Furthermore,
In the second embodiment, when the calculated correction coefficient exceeds the maximum value of the correction coefficient, the maximum value is used instead of the calculated correction coefficient. However, when the calculated correction coefficient falls below the minimum value of the correction coefficient. A minimum value may be used instead of the calculated value, or an appropriate value, for example, a corresponding one of the maximum value and the minimum value of the correction coefficient is used instead of the calculated value when the calculated value deviates from the allowable range of the correction coefficient. May be.

In the third or fourth embodiment, the continuation or start of the deceleration operation state is detected based on the volumetric efficiency ηv or the throttle position sensor output change rate ΔTPS, and a predetermined time elapses after the vehicle departs from the deceleration operation state. Until the shift operation is determined, the misfire state is once detected, and then the angular acceleration exceeds the rough road determination level, or the sign of the sensor output change rate ΔTPS is inverted (depressing and releasing the accelerator pedal). Is performed a predetermined number of times,
Although it is determined that the vehicle is traveling on a rough road, the deceleration operation state, the changing operation, and the rough road traveling can be detected by various methods. For example, a deceleration operation state can be detected based on a throttle valve opening, an intake air amount, and the like, and rough road traveling can be detected based on front and rear wheel speeds, vehicle body acceleration, and the like. In particular, in the fourth embodiment, the start of the deceleration operation is determined based on the throttle position sensor output change rate ΔTPS, and after this determination, during the set time (ie, the elapsed time counted by the third timer) TIM1
Until 3 exceeds the set value), it is determined that there is a high probability of continuing deceleration operation (or there is a high possibility of misfire misjudgment based on deceleration operation or the effect of gear shifting remains), and misfire detection and correction coefficient calculation However, in the determination of the start of the deceleration operation, instead of or in addition to the determination of the throttle position sensor output change rate ΔTPS, the rate of change Δη of the volumetric efficiency ηv obtained from the output of the air flow sensor or the like is determined.
It may be determined that v has fallen below the set value, or it may be detected that the intake pressure PB obtained from the boost sensor output or the like has fallen below the set value.

Continued on the front page (72) Inventor Yasuhisa Yoshida 5-33-8 Shiba, Minato-ku, Tokyo Inside Mitsubishi Motors Corporation (72) Inventor Mitsuhiko Yanagisawa 5-33-8 Shiba, Minato-ku, Tokyo Mitsubishi Motors (72) Inventor Hiroyuki Nakajima 5-33-8 Shiba, Minato-ku, Tokyo Mitsubishi Motors Corporation (72) Inventor Koichi Namiki 5-33-8 Shiba, Minato-ku, Tokyo Mitsubishi (72) Inventor Satoshi Kasai 5-33-8 Shiba, Minato-ku, Tokyo Mitsubishi Motors Corporation (56) References JP-A-4-66755 (JP, A) JP-A-Hei 5-18311 (JP, A) JP-A-4-203341 (JP, A) JP-B-63-57616 (JP, B2) (58) Fields investigated (Int. Cl. ⁶ , DB name) F02D 45/00 F02P 17/00

Claims (17)

(57) [Claims] A time interval from a point of entry into a crankshaft rotation angle region corresponding to a specific stroke phase of each cylinder of an internal combustion engine to a point of departure from the angle region detected by a crank angle sensor is sequentially detected. A misfire detection method for detecting the occurrence of a misfire based on the time interval, wherein a correction coefficient for compensating a structural error of the crank angle sensor is calculated based on the time interval, A method for detecting misfire due to crankshaft rotation fluctuations, comprising correcting internal combustion engine rotation information based on the correction coefficient, and detecting occurrence of misfire based on the corrected internal combustion engine rotation information. 2. The method according to claim 1, wherein the internal combustion engine rotation information is an average angular acceleration of a crankshaft. 3. The method according to claim 1, wherein the correction coefficient is repeatedly calculated based on a time interval sequentially detected during operation of the internal combustion engine to update the correction coefficient, and an average angular velocity of the crankshaft is sequentially determined based on the time interval. The misfire detection method according to claim 2, wherein the average angular acceleration is sequentially calculated based on the time interval, the average angular velocity, and the correction coefficient. 4. A specific operating state in which the correction coefficient is updated by repeatedly calculating the correction coefficient based on time intervals sequentially detected during the operation of the internal combustion engine, while the load of the internal combustion engine changes suddenly. 2. The misfire detection method according to claim 1, wherein when the internal combustion engine is operating, updating of the correction coefficient is prohibited. 5. A vehicle equipped with the internal combustion engine is running on a rough road when the internal combustion engine is performing a deceleration operation, when a shift operation is performed by a transmission connected to the internal combustion engine, and when the vehicle is mounted on a rough road. 5. The misfire detection method according to claim 4, wherein at least one of the states is determined as the specific operation state. 6. The method according to claim 1, wherein the correction coefficient is repeatedly calculated based on time intervals sequentially detected during operation of the internal combustion engine to update the correction coefficient, while a predetermined time has elapsed since the start of the internal combustion engine. 2. The method according to claim 1, wherein updating of the correction coefficient is prohibited when there is no misfire. 7. For each of a plurality of operating regions separated from each other by predetermined operating parameters of the internal combustion engine,
Each of the independent correction coefficients is calculated, and the correction coefficient for the current operation area is updated each time the current operation area is determined, while the correction coefficient for the operation area other than the current operation area is updated. 2. The method according to claim 1, wherein the misfire is prohibited. 8. The method according to claim 7, wherein the predetermined operating parameter is a rotation speed of the internal combustion engine. 9. The method according to claim 7, wherein the predetermined operating parameter is a volumetric efficiency of the internal combustion engine. 10. The system according to claim 7, wherein the plurality of operating regions are separated from each other by a rotational speed and a volumetric efficiency of the internal combustion engine as the predetermined operating parameters.
5. A misfire detection method based on crankshaft rotation fluctuation described in 4. 11. When the correction coefficient is updated by repeatedly calculating the correction coefficient based on time intervals sequentially detected during operation of the internal combustion engine, and when the calculated correction coefficient deviates from a predetermined allowable range. The method according to claim 1, wherein a maximum value or a minimum value corresponding to the allowable range is used as the correction coefficient instead of the calculated correction coefficient. 12. The crankshaft according to claim 11, wherein the upper limit and the lower limit of the allowable range correspond to a maximum allowable error in the structure of the crank angle sensor. Misfire detection method. 13. The correction coefficient stored in a nonvolatile memory, and the correction coefficient read from the nonvolatile memory when the internal combustion engine is started is used as an initial value of the correction coefficient. 2. The method for detecting misfire due to crankshaft rotation fluctuation according to 1. 14. A rotating body which rotates together with a crankshaft; and a plurality of identification means provided on the rotating body at intervals in a circumferential direction for distinguishing between the point of entry and the point of disengagement, respectively. A detection unit that is provided on a fixed member of the internal combustion engine and emits a detection signal each time the identification unit approaches, and the plurality of identification units include first and second identification units for each cylinder. Using the crank angle sensor, a first detection signal generated by the first identification unit in proximity to the detection unit and a second detection signal generated by the second identification unit in proximity to the detection unit Detecting the time interval by measuring an interval with a detection signal; and detecting the time interval and the crank including at least one of generation times of the first and second detection signals used for the detection of the time interval. Misfire detecting method according to the crankshaft rotational fluctuation of claim 1, characterized in that determining the correction factor based rotation period of the to. 15. The present invention is applied to a multi-cylinder internal combustion engine in which a plurality of cylinders can sequentially perform an explosion stroke at equal intervals during rotation of a crankshaft as the internal combustion engine, and the second identification means for each of the cylinders includes: 15. The method for detecting misfire due to crankshaft rotation fluctuation according to claim 14, wherein the crank angle sensor provided to function as the first identification means for the next cylinder going through an explosion stroke is used. 16. Smoothing of instantaneous correction coefficient information sequentially obtained based on the detection time interval and the crankshaft rotation cycle during operation of the internal combustion engine, and correcting the internal combustion engine rotation information with the smoothed value. Claim 14
5. A misfire detection method based on crankshaft rotation fluctuation described in 4. 17. The correction coefficient is updated based on time intervals sequentially detected during operation of the internal combustion engine to update the correction coefficient, and the latest detection time interval is set to Tm (n), A is the number of the identification means provided on the body, T (n) is the latest rotation cycle of the crankshaft, and K is the correction coefficient calculated last time.
Assuming that Lm (n-1) and the weighting coefficient are a (0 ≦ a ≦ 1), the latest correction coefficient KLm (n) is KLm (n) = a · KLm (n−1) + (1-a) 15. The method for detecting misfire due to crankshaft rotation fluctuation according to claim 14, wherein KLm = {A.Tm (n)} / T (n).