阅读说明：本技术 曲柄角速度测定装置以及失火判定装置 (Crank angular velocity measuring device and misfire judging device ) 是由 岸信之 宇津木克洋 深谷修一 田岛克树 于 2021-05-27 设计创作，主要内容包括：提供曲柄角速度测定装置以及失火判定装置,利用与发动机的曲轴同步旋转的交流旋转电机的磁性传感器的输出来测定曲柄角速度。交流旋转电机具有以S极以及N极的各永磁体相互等间隔地呈圆周状设置而构成的转子和检测永磁体的磁化边界的第一、第二以及第三磁性传感器并与曲轴同步旋转。计时单元对从第一磁性传感器检测磁化边界到第二个磁性传感器检测磁化边界的第一经过时间、从第二磁性传感器检测磁化边界到第三个磁性传感器检测磁化边界的第二经过时间以及从第三磁性传感器检测磁化边界到第一个磁性传感器检测磁化边界的第三经过时间进行计时。移动平均计算单元基于所述第一、第二以及第三经过时间的移动平均值来计算曲柄角速度。(A crank angular velocity measuring device and a misfire judging device are provided, which measure a crank angular velocity by using an output of a magnetic sensor of an AC rotating machine rotating in synchronization with a crankshaft of an engine. The ac rotating machine includes a rotor formed by arranging permanent magnets of S-pole and N-pole circumferentially at equal intervals, and first, second, and third magnetic sensors for detecting magnetization boundaries of the permanent magnets, and rotates in synchronization with a crankshaft. The timing unit times a first elapsed time from when the first magnetic sensor detects the magnetization boundary to when the second magnetic sensor detects the magnetization boundary, a second elapsed time from when the second magnetic sensor detects the magnetization boundary to when the third magnetic sensor detects the magnetization boundary, and a third elapsed time from when the third magnetic sensor detects the magnetization boundary to when the first magnetic sensor detects the magnetization boundary. A moving average calculation unit calculates a crank angular velocity based on moving averages of the first, second, and third elapsed times.)

1. A crank angular velocity measurement device that measures a crank angular velocity of an engine (E) is characterized by comprising:

an alternating current rotating machine (M) which has a rotor formed by permanent magnets (60) of S pole and N pole arranged in a circumferential manner at equal intervals, and a first magnetic sensor (61a), a second magnetic sensor (61b) and a third magnetic sensor (61c) for detecting the magnetization boundary of the permanent magnets, and rotates synchronously with the crankshaft;

a timing unit (51a) that counts a first elapsed time from when the first magnetic sensor (61a) detects the magnetization boundary to when the second magnetic sensor (61b) detects the magnetization boundary, a second elapsed time from when the second magnetic sensor (61b) detects the magnetization boundary to when the third magnetic sensor (61c) detects the magnetization boundary, and a third elapsed time from when the third magnetic sensor (61c) detects the magnetization boundary to when the first magnetic sensor (61a) detects the magnetization boundary;

a moving average calculation unit (52) that calculates a crank angular velocity based on a moving average of the first elapsed time, the second elapsed time, and the third elapsed time.

2. The crank angular velocity measuring device according to claim 1,

the timing unit counts a first elapsed time from when the first magnetic sensor (61a) detects the nth magnetization boundary to when the second magnetic sensor (61b) detects the nth magnetization boundary, a second elapsed time from when the second magnetic sensor (61b) detects the nth magnetization boundary to when the third magnetic sensor (61c) detects the nth magnetization boundary, and a third elapsed time from when the third magnetic sensor (61c) detects the nth magnetization boundary to when the first magnetic sensor (61a) detects the (n + 1) th magnetization boundary.

3. The crank angular velocity measuring apparatus according to claim 1 or 2,

the timing unit clocks a second elapsed time from the detection of the nth magnetization boundary by the second magnetic sensor (61b) to the detection of the nth magnetization boundary by the third magnetic sensor (61c), a third elapsed time from the detection of the nth magnetization boundary by the third magnetic sensor (61c) to the detection of the (n + 1) th magnetization boundary by the first magnetic sensor (61a), and a first elapsed time from the detection of the (n + 1) th magnetization boundary by the first magnetic sensor (61a) to the detection of the (n + 1) th magnetization boundary by the second magnetic sensor (61 b).

4. The crank angular velocity measuring device according to any one of claims 1 to 3,

the timing unit times a third elapsed time from when the third magnetic sensor (61c) detects the nth magnetization boundary to when the first magnetic sensor (61a) detects the (n + 1) th magnetization boundary, a first elapsed time from when the first magnetic sensor (61a) detects the (n + 1) th magnetization boundary to when the second magnetic sensor (61b) detects the (n + 1) th magnetization boundary, and a second elapsed time from when the second magnetic sensor (61b) detects the (n + 1) th magnetization boundary to when the third magnetic sensor (61c) detects the (n + 1) th magnetization boundary.

5. A misfire judging device for judging misfire based on a crank angular velocity of an engine (E), comprising:

an alternating current rotating machine (M) which has a rotor formed by permanent magnets (60) of S pole and N pole arranged in a circumferential manner at equal intervals, and a first magnetic sensor (61a), a second magnetic sensor (61b) and a third magnetic sensor (61c) for detecting the magnetization boundary of the permanent magnets, and rotates synchronously with the crankshaft;

a timing unit (51a) that counts a first elapsed time from when the first magnetic sensor (61a) detects the magnetization boundary to when the second magnetic sensor (61b) detects the magnetization boundary, a second elapsed time from when the second magnetic sensor (61b) detects the magnetization boundary to when the third magnetic sensor (61c) detects the magnetization boundary, and a third elapsed time from when the third magnetic sensor (61c) detects the magnetization boundary to when the first magnetic sensor (61a) detects the magnetization boundary;

a moving average calculation unit (52) that calculates a crank angular velocity based on a moving average of the first elapsed time, the second elapsed time, and the third elapsed time;

and a misfire identification unit (10a) that identifies an engine misfire based on a result of calculation of the crank angular velocity.

6. The misfire judging apparatus according to claim 5,

the timing unit counts a first elapsed time from when the first magnetic sensor (61a) detects the nth magnetization boundary to when the second magnetic sensor (61b) detects the nth magnetization boundary, a second elapsed time from when the second magnetic sensor (61b) detects the nth magnetization boundary to when the third magnetic sensor (61c) detects the nth magnetization boundary, and a third elapsed time from when the third magnetic sensor (61c) detects the nth magnetization boundary to when the first magnetic sensor (61a) detects the (n + 1) th magnetization boundary.

7. The misfire judging apparatus according to claim 5 or 6,

the timing unit clocks a second elapsed time from the detection of the nth magnetization boundary by the second magnetic sensor (61b) to the detection of the nth magnetization boundary by the third magnetic sensor (61c), a third elapsed time from the detection of the nth magnetization boundary by the third magnetic sensor (61c) to the detection of the (n + 1) th magnetization boundary by the first magnetic sensor (61a), and a first elapsed time from the detection of the (n + 1) th magnetization boundary by the first magnetic sensor (61a) to the detection of the (n + 1) th magnetization boundary by the second magnetic sensor (61 b).

8. The misfire judging apparatus according to any one of claims 5 to 7,

the timing unit times a third elapsed time from when the third magnetic sensor (61c) detects the nth magnetization boundary to when the first magnetic sensor (61a) detects the (n + 1) th magnetization boundary, a first elapsed time from when the first magnetic sensor (61a) detects the (n + 1) th magnetization boundary to when the second magnetic sensor (61b) detects the (n + 1) th magnetization boundary, and a second elapsed time from when the second magnetic sensor (61b) detects the (n + 1) th magnetization boundary to when the third magnetic sensor (61c) detects the (n + 1) th magnetization boundary.

Technical Field

The present invention relates to a crank angular velocity measuring device and a misfire judging device for an internal combustion engine vehicle, and more particularly to a crank angular velocity measuring device and a misfire judging device that measure a crank angular velocity using an output of a magnetic sensor of an ac rotating machine that rotates in synchronization with a crankshaft of an engine.

Background

In a four-wheel vehicle, a technique is known for determining an engine misfire from a crank angular velocity measured from a generation time interval of crank pulses (crank inter-pulse time). Here, since the parameter for misfire identification (misfire parameter) includes the tooth space error of the crankshaft pulser rotor, it is necessary to remove the tooth space error of the crankshaft pulser rotor in order to accurately perform misfire identification.

Patent document 1 discloses a method for removing an error between teeth of a crankshaft pulser rotor, which focuses on: in the equi-spaced explosion type engine, the crank angular velocity component caused by the inertia torque varies within the pulse generation period of the TDC. Patent document 2 discloses a technique of obtaining a relative value of crank angular velocity and removing an error between teeth of a crank pulser rotor based on an integrated value of the relative value.

Documents of the prior art

Patent document

Patent document 1: japanese unexamined patent application publication No. 2008-111354

Patent document 2: WO2018-179340 publication

In recent years, it has been discussed that a misfire identification technique is also adopted in a motorcycle from the viewpoint of improvement of a repairing property and environmental protection. The method for removing the tooth-to-tooth error of the crankshaft pulser rotor of patent document 1 is not applicable to a vehicle or a single cylinder engine that does not employ a crankshaft pulser rotor system, or a multi-cylinder engine that does not employ a crankshaft pulser rotor system, on the premise that the method belongs to an equi-spaced spark-ignition engine and is equipped with the crankshaft pulser rotor system.

In patent document 2, whether the engine is of an equi-interval detonation type or an unequally-interval detonation type, the fixing error component can be removed like the tooth-to-tooth error of the crankshaft pulser rotor. On the other hand, the crank angular velocity can also be measured by the following method: instead of using the crankshaft pulser rotor, the rotational speed of a magnet provided in the ACG starter motor that rotates in synchronization with the crankshaft is detected, for example, with a magnetic sensor.

Here, since the function and sensitivity of the magnetic sensor change and deteriorate due to environmental changes such as temperature, humidity, and atmospheric pressure and aging changes, in the measurement using the magnetic sensor, it is necessary to consider a fluctuation error component that depends on environmental changes and aging changes, in addition to a fixed error component that depends on mounting accuracy. However, in patent document 2, although the fixed error component can be excluded, the variable error component cannot be excluded.

Disclosure of Invention

An object of the present invention is to solve the above-described problems and provide a crank angular velocity measuring device and a misfire judging device that can accurately measure a crank angular velocity by excluding error components that vary due to environmental changes and aging changes, included in measurement errors of the crank angular velocity, without providing a crank pulser rotor system, and that can accurately judge misfire based on the measurement results of the crank angular velocity.

In order to achieve the above object, the crank angular velocity measuring device of the present invention includes the following structures (1) and (2).

(1) The disclosed device is provided with: an alternating-current rotating electrical machine having a rotor formed by permanent magnets of S-pole and N-pole arranged circumferentially at equal intervals, and a first magnetic sensor, a second magnetic sensor, and a third magnetic sensor for detecting a magnetization boundary of the permanent magnets, and rotating in synchronization with a crankshaft; a timing unit that counts a first elapsed time from when the first magnetic sensor detects the magnetization boundary to when the second magnetic sensor detects the magnetization boundary, a second elapsed time from when the second magnetic sensor detects the magnetization boundary to when the third magnetic sensor detects the magnetization boundary, and a third elapsed time from when the third magnetic sensor detects the magnetization boundary to when the first magnetic sensor detects the magnetization boundary; and a unit that calculates a crank angular velocity based on a moving average of the first elapsed time, the second elapsed time, and the third elapsed time.

(2) The unit for calculating crank angular velocity repeats the calculation: an average value of a first elapsed time from the first magnetic sensor detecting the nth magnetization boundary to the second magnetic sensor detecting the nth magnetization boundary, a second elapsed time from the second magnetic sensor detecting the nth magnetization boundary to the third magnetic sensor detecting the nth magnetization boundary, and a third elapsed time from the third magnetic sensor detecting the nth magnetization boundary to the first magnetic sensor detecting the (n + 1) th magnetization boundary; an average value of a second elapsed time from the second magnetic sensor detecting the nth magnetization boundary to the third magnetic sensor detecting the nth magnetization boundary, a third elapsed time from the third magnetic sensor detecting the nth magnetization boundary to the first magnetic sensor detecting the n +1 th magnetization boundary, and a first elapsed time from the first magnetic sensor detecting the n +1 th magnetization boundary to the second magnetic sensor detecting the n +1 th magnetization boundary; an average value of a third elapsed time from the third magnetic sensor detecting the nth magnetization boundary to the first magnetic sensor detecting the n +1 th magnetization boundary, a first elapsed time from the first magnetic sensor detecting the n +1 th magnetization boundary to the second magnetic sensor detecting the n +1 th magnetization boundary, and a second elapsed time from the second magnetic sensor detecting the n +1 th magnetization boundary to the third magnetic sensor detecting the n +1 th magnetization boundary.

In order to achieve the above object, the misfire identification device of the present invention is characterized by having the following configurations (3) and (4).

(3) The disclosed device is provided with: an alternating-current rotating electrical machine having a rotor formed by permanent magnets of S-pole and N-pole arranged circumferentially at equal intervals, and a first magnetic sensor, a second magnetic sensor, and a third magnetic sensor for detecting a magnetization boundary of the permanent magnets, and rotating in synchronization with a crankshaft; a timing unit that counts a first elapsed time from when the first magnetic sensor detects the magnetization boundary to when the second magnetic sensor detects the magnetization boundary, a second elapsed time from when the second magnetic sensor detects the magnetization boundary to when the third magnetic sensor detects the magnetization boundary, and a third elapsed time from when the third magnetic sensor detects the magnetization boundary to when the first magnetic sensor detects the magnetization boundary; a unit that calculates a crank angular velocity based on a moving average of the first elapsed time, the second elapsed time, and the third elapsed time; and a misfire determination unit that determines an engine misfire based on a result of calculation of the crank angular velocity.

(4) The unit for calculating crank angular velocity repeats the calculation: an average value of a first elapsed time from the first magnetic sensor detecting the nth magnetization boundary to the second magnetic sensor detecting the nth magnetization boundary, a second elapsed time from the second magnetic sensor detecting the nth magnetization boundary to the third magnetic sensor detecting the nth magnetization boundary, and a third elapsed time from the third magnetic sensor detecting the nth magnetization boundary to the first magnetic sensor detecting the (n + 1) th magnetization boundary; an average value of a second elapsed time from the second magnetic sensor detecting the nth magnetization boundary to the third magnetic sensor detecting the nth magnetization boundary, a third elapsed time from the third magnetic sensor detecting the nth magnetization boundary to the first magnetic sensor detecting the n +1 th magnetization boundary, and a first elapsed time from the first magnetic sensor detecting the n +1 th magnetization boundary to the second magnetic sensor detecting the n +1 th magnetization boundary; an average value of a third elapsed time from the third magnetic sensor detecting the nth magnetization boundary to the first magnetic sensor detecting the n +1 th magnetization boundary, a first elapsed time from the first magnetic sensor detecting the n +1 th magnetization boundary to the second magnetic sensor detecting the n +1 th magnetization boundary, and a second elapsed time from the second magnetic sensor detecting the n +1 th magnetization boundary to the third magnetic sensor detecting the n +1 th magnetization boundary.

Effects of the invention

(1) According to the crank angular velocity measuring apparatus of the present invention, when the crank angular velocity is measured using the output of the magnetic sensor that detects the UVW phase of the ac rotating machine that rotates in synchronization with the crankshaft, even if an error occurs in the measured value of the crank angular velocity due to a change in the interval and function between the magnetic sensors due to environmental or temporal influences, the error can be cancelled out by the calculation process of the present invention and excluded from the crank angular velocity, and thus accurate crank angular velocity measurement can be performed.

(2) According to the misfire determination device of the present invention, when misfire determination is performed based on the crank angular velocity measured using the output of the magnetic sensor that detects the UVW phase of the ac rotating machine that rotates in synchronization with the crankshaft, even if an error occurs in the measured value of the crank angular velocity due to the change in the mutual spacing and function of the magnetic sensors due to environmental or temporal influences, the error can be cancelled out by the calculation process of the present invention and excluded from the crank angular velocity, so that accurate crank angular velocity measurement can be performed, and as a result, misfire determination with high accuracy can be performed.

Drawings

Fig. 1 is a functional block diagram showing the structure of a misfire determination apparatus to which the present invention is applied.

Fig. 2 is a functional block diagram showing a structure of a crank angular velocity measuring unit to which the present invention is applied.

Fig. 3 is a time chart showing the relationship of the crank angle to the outputs of the respective magnetic sensors.

Fig. 4 is a diagram for explaining a principle in which errors in the mutual spacing of the magnetic sensors are cancelled.

Fig. 5 is a diagram showing a calculation method of the moving average value of the elapsed time Δ t.

Fig. 6 is a diagram for explaining the function of the 720-degree filtering processing unit.

Fig. 7 (a) and (b) are diagrams for explaining the function of the relative angular velocity calculating section.

Fig. 8 (a) and (b) are diagrams for explaining the function of the integrated angular velocity calculating unit.

FIG. 9 is an exemplary graph showing the results of on-computer calculations of inertial torque

Fig. 10 is a graph showing an example of the relative value of the amount of change in crank angular velocity due to inertia torque with reference to the compression upper limit point.

Fig. 11 is an example graph showing the suction torque component.

Fig. 12 is a diagram for explaining the function of the crank angular velocity measurement error removing unit.

Fig. 13 (a) to (c) are diagrams showing an example of a misfire determination method (no misfire) in which relative angular velocities are obtained from crank angular velocities and an integrated value of the relative angular velocities is used as a misfire parameter.

Fig. 14 (a) to (c) are diagrams showing an example of a misfire determination method (presence of misfire) in which relative angular velocities are obtained from crank angular velocities and an integrated value of the relative angular velocities is used as a misfire parameter.

Description of the reference numerals

5: a crank angular velocity measuring part;

6: a 720-degree filtering processing unit;

7: a relative angular velocity calculating section;

8: an accumulated angular velocity calculation unit;

9: an inertia torque component removing unit;

10: a suction torque component removing section;

11: a crank angular velocity measurement error component removing unit;

12: a misfire judging section;

100: a misfire judging device.

Detailed Description

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Here, first, an outline of a misfire determination method using the crank angular velocity measured by the crank angular velocity measurement method of the present invention will be described, and then, an embodiment thereof will be described in detail.

Fig. 13 and 14 are explanatory diagrams of a misfire identification method in which the relative angular velocity is obtained from the measured value of the crank angular velocity [ each graph (a) ] and the integrated value (integrated angular velocity) [ each graph (c) ] is obtained as the misfire parameter, and each graph (a) is a graph showing the crank angular velocity on the vertical axis in terms of the crank angle with the upper limit point of compression as the origin. The relative angular velocity of the crank angular velocity is calculated by subtracting the reference angular velocity detected in the vicinity of the compression upper limit point of each cylinder of the engine from the measurement result of the crank angular velocity at each crank angle. Fig. 13 shows a case where combustion is normally performed, and fig. 14 shows a case where a misfire occurs.

The relative angular velocity is increased when combustion is normally performed in the combustion stroke after the compression upper limit point of the engine and is decreased when there is a misfire [ fig. (b) ]. Therefore, the integrated angular velocity obtained by integrating the relative angular velocities in the predetermined crank angle range is a positive value when the combustion is normally performed, and is a negative value when the misfire occurs, and therefore can be used as a parameter for determining the misfiring cylinder.

However, the integrated angular velocity includes, in addition to the combustion torque, an angular velocity component that fluctuates due to a measurement error of the crank angular velocity, noise caused by a dynamic change of the sensor gap, or the like, load torque, friction, inertia torque, suction torque, or the like. Therefore, in order to correctly perform misfire determination, it is necessary to remove all of these fluctuation components from the integrated angular velocity.

Among these fluctuation components, noise, load torque, friction, and inertia torque are removed by statistical processing, engine control, or computer calculation. The suction torque can be removed by the technique disclosed in patent document 2. In addition, in the crank acceleration measurement error, although the technique disclosed in patent document 2 can be used to remove the fixed error, there is no effective method for removing the error that changes due to the influence of the environment and the elapsed time.

Here, the error that changes due to the influence of the environment or the elapsed time, which is included in the measurement error of the crank angular velocity, refers to an error caused by, for example, a fluctuation in the temperature characteristic of the magnetic sensor, a positional deviation of the magnetic sensor due to deformation of the magnetic sensor holding member, a change in the responsiveness of the magnetic sensor due to deterioration with time, or the like.

Next, embodiments of the present invention will be described in detail with reference to the drawings. Fig. 1 is a functional block diagram showing the structure of a misfire determination apparatus to which the present invention is applied. Here, a motorcycle equipped with a four-stroke single cylinder engine E and an ACG starter motor (three-phase ac rotating electrical machine) M will be described as an example.

In the misfire identification apparatus 100, the crank angular velocity measuring section 5 measures the angular velocity of the crankshaft of the engine E. Fig. 2 is a functional block diagram showing a configuration related to crank angular velocity measurement to which the present invention is applied, and in the present embodiment, crank angular velocity is measured based on the outer rotor angular velocity of the ACG starter motor M that rotates in synchronization with the crankshaft of the engine E.

The ACG starter motor M includes: an outer rotor (not shown) formed by arranging the permanent magnets 60 of S-pole and N-pole circumferentially at equal intervals, and first, second, and third magnetic sensors 61(61a, 61b, 61c) for detecting magnetization boundaries of the permanent magnets 601. In the present embodiment, the magnetization boundary switching from the S pole to the N pole is detected, but the magnetization boundary switching from the N pole to the S pole may be detected. The first, second, and third magnetic sensors 61a, 61b, and 61c also have a function of detecting the U-phase, V-phase, and W-phase of the ACG starter motor M.

The crank angular velocity measuring unit 5 includes first, second, and third timing units 51a, 51b, and 51c and a moving average calculating unit 52, and calculates a crank angular velocity for misfire determination at each stage of engine control. The moving average calculation unit 52 removes noise caused by dynamic changes in the gap between the magnetic sensors 61 or between the permanent magnets 60, noise caused by dynamic changes in the sensor gap, and errors that vary with time. Here, the dynamic change of the sensor gap means a change in the distance between the magnetic sensor 61 and the permanent magnet 60 during operation (the change in the gap between the magnetic sensor 61 is a change with time due to thermal deformation of the sensor holding member).

Fig. 3 is a timing chart showing the relationship between the crank angle of the engine E and the output signal of each magnetic sensor 61.

The first timer unit 51a counts a first elapsed time Δ t1 from when the first magnetic sensor 61a detects the nth magnetization boundary Bn to when the second magnetic sensor 61b detects the nth magnetization boundary. The second timer unit 51b counts a second elapsed time Δ t2 from when the second magnetic sensor 61b detects the magnetization boundary Bn to when the third magnetic sensor 61c detects the magnetization boundary Bn. The third timer 51c counts a third elapsed time Δ t3 from the detection of the magnetization boundary Bn by the third magnetic sensor 61c to the detection of the next (n +1 th) magnetization boundary Bn +1 by the first magnetic sensor 61 a.

The moving average calculation unit 52 calculates a moving average of three consecutive elapsed times Δ t1, Δ t2, and Δ t 3. That is, after the first elapsed time Δ t1, the second elapsed time Δ t2, and the third elapsed time Δ t3 are obtained, the moving average value (Δ t1+ Δ t2+ Δ t3)/3 is calculated.

Then, after a second first elapsed time Δ t1_2 from the detection of the magnetization boundary Bn +1 by the first magnetic sensor 61a to the detection of the magnetization boundary Bn +1 by the second magnetic sensor 61b is timed, the moving average calculation section 52 calculates a moving average (Δ t2+ Δ t3+ Δ t1_2)/3 of the second elapsed time Δ t2, the third elapsed time Δ t3, and the second first elapsed time Δ t1_ 2.

Then, after a second elapsed time Δ t2_2 from when the second magnetic sensor 61b detects the magnetization boundary Bn +1 to when the third magnetic sensor 61c detects the magnetization boundary Bn +1 is counted, the moving average calculation unit 52 calculates a moving average (Δ t3+ Δ t1_2+ Δ t2_2)/3 of the third elapsed time Δ t3, the second first elapsed time Δ t1_2, and the second elapsed time Δ t2_ 2.

Here, as shown in fig. 4, when the interval between the magnetic sensors 61a/61B and 61B/61c is a, and the interval between the magnetization boundary Bn and the magnetization boundary Bn +1 (magnetization interval) is B (2A + α) longer than 2A by α, the measurement distance of the first elapsed time Δ t1 is a, the measurement distance of the second elapsed time Δ t2 is a, and the measurement distance of the third elapsed time Δ t3 is B-2A.

In the present embodiment, since the distance L to be measured is the sum of the respective measurement distances of the first, second, and third elapsed times Δ t1, Δ t2, and Δ t3, a + (B-2A) ═ B does not relate to a. In addition, since the error of the magnetization interval B is fixed without considering environmental changes or deterioration with age, it can be corrected by the method of patent document 2.

Therefore, even if the intervals of the magnetic sensors 61a/61b and 61b/61c include errors Δ d1 and Δ d2, which are a combination of manufacturing errors such as machining accuracy and assembly accuracy and errors in fluctuations due to environmental changes and temporal influences, in addition to the design interval a, the crank angular velocity ω can be measured regardless of the crank angle without being affected by the error component included in the output of each magnetic sensor 61, as shown in fig. 5.

Returning to fig. 1, the 720-degree filtering unit 6 eliminates the linear change amount in one stroke cycle for the calculation result (moving average) of the crank angular velocity ω measured as described above, and extracts the fluctuation component of a relatively long cycle. This makes it possible to remove an angular velocity fluctuation component caused by a load torque applied from a tire or an auxiliary machine of a vehicle driven by an engine or friction of a sliding member of the engine. Fig. 6 shows an example in which 720 degrees filtering is employed in a stroke in which combustion torque is generated but deceleration is caused by load torque.

As shown in fig. 7, the relative angular velocity calculation unit 7 uses the crank angular velocity ω measured in the vicinity of the compression upper limit point (TDC: crank angle 0 degrees) as a reference angular velocity ω 1ref [ see fig. (a) ], and determines the difference between the angular velocity ω 1_ i measured at each crank angle i in the crank angle range from the TDC to 180 degrees and the reference angular velocity ω 1ref as the relative angular velocity ω 1_ i (═ ω 1_ i- ω 1ref) [ see fig. (b) ].

As shown in fig. 8, the integrated angular velocity calculation unit 8 integrates the relative angular velocity ω ω 1_ i in a crank angle range of 180 degrees, and calculates an integrated angular velocity Σ ω 1_ i.

The inertia torque component removing unit 9 removes the inertia torque component from the integrated angular velocity Σ ω 1 — i to calculate the post-removal integrated angular velocity Σ ω 1' _ i. The inertia torque component can be calculated on a computer using, for example, a method disclosed in patent document 2. The inertia torque Tq of the single cylinder engine is shown in fig. 9, for example, and the relative value d ω' of the change d ω in the crank angular velocity caused by the inertia torque with reference to the value of the compression upper limit point is shown in fig. 10.

In the present embodiment, the relative value d ω' obtained from the unit engine speed is calculated by the moving average calculation unit 52 in the same manner as the moving average, and the value integrated in the same section as the integrated angular velocity calculation unit 8 is calculated and set in the ECU. The inertia torque component at an arbitrary engine speed can be calculated by multiplying this value by the engine speed. Further, using this value, the post-removal integrated angular velocity Σ ω 1' _ i can be calculated.

The suction torque component removal unit 10 calculates the post-removal integrated angular velocity Σ ω 1' _ i by removing the crank angular velocity component due to the torque generated by the pump operation of the engine E. The error removal can be performed by setting an ECU that obtains the suction torque component by the method disclosed in patent document 2 (a method of measuring the integrated angular velocity in the fuel supply cut state in a standard vehicle without an error and removing the inertia torque).

The suction torque contribution depends on the engine speed and the intake manifold pressure. The suction torque component of the single cylinder engine is found in the form of a line graph based on the engine speed and the intake manifold pressure as shown in fig. 11, for example. In the present embodiment, the line map data is set in the ECU and used for correction of the misfire parameter.

The crank angular velocity measurement error component removing unit 11 removes an accumulated angular velocity error caused by a backlash error of the pulser rotor or a fixed error such as a magnetization position shift in the present embodiment. The error removal is performed by using the method for removing the backlash error disclosed in patent document 2 (a coefficient Kpul1 for calculating the measurement error of the crank angular velocity due to the backlash error per unit rotational speed is calculated by subtracting the inertia torque component and the suction torque component set in advance in the ECU from the integrated angular velocity calculated in the fuel supply shutoff state and dividing the result by the reference angular velocity ω ref1, the coefficient is stored in the ECU, and then the error component at an arbitrary engine rotational speed is calculated by multiplying the reference angular velocity ω ref1 and the component is removed from the integrated angular velocity).

Fig. 12 shows an example in which the inertia torque component (2) and the suction torque component (3) are removed from the integrated value (1) of the relative angular velocity to extract the crank angular velocity measurement error component.

In the present embodiment, the misfire parameter is a value obtained by removing the crank angular velocity measurement error component from the removed integrated angular velocity Σ ω 1' _ i. The misfire judging section 12 judges whether or not misfire occurs based on the misfire parameter.

According to the present embodiment, when misfire determination is performed based on the crank angular velocity measured using the output of the magnetic sensor that detects the UVW phase of the ac rotating machine that rotates in synchronization with the crankshaft, even if an error occurs in the measured value of the crank angular velocity due to environmental or temporal effects on the mutual spacing and functional changes of the magnetic sensors, the error can be cancelled out by the calculation procedure of the present invention and excluded from the crank angular velocity, so that accurate crank angular velocity measurement can be performed, and as a result, misfire determination with high accuracy can be performed.

In the above embodiment, the application to a single cylinder engine has been described as an example, but the present invention is not limited to this, and can be similarly applied to a multi-cylinder engine. In this case, there is no problem in either the equi-spaced explosion type engine or the unequal-spaced explosion type engine.