JP2019090329A - Sensor failure determination device of engine - Google Patents

Abstract

To provide a sensor failure determination device of an engine which can specify a failed sensor by precisely discriminating a failure of a specified sensor and failures of a plurality of other sensors when determining a sensor failure.SOLUTION: Simulated filling efficiency Ec and simulated intake manifold pressure Pin are calculated on the basis of boost pressure Pboost which is detected by a boost pressure sensor (S1, 2), and when each simulated value deviates from actual values of an AFS and an intake manifold pressure sensor (YES in S3), a deviation ΔEc between the conversion filling efficiency Ec(in) of the actual values acquired from the intake manifold pressure Pin and actual filling efficiency Ec based on an intake amount V is calculated (S4, 5). When the deviation ΔEc is smaller than correlative index determination value ΔEc0 (YES in S6), a failure of a boost pressure sensor is determined (S7), and when the deviation is equal to or larger than the correlative index determination value ΔEc0 (No in S6), either or both of the failures of the AFS and the intake manifold pressure sensor are determined (S8 to 12).SELECTED DRAWING: Figure 2

Description

  The present invention relates to a sensor failure determination device for an engine.

  An engine mounted on a vehicle as a power source for traveling is equipped with various sensors, and based on the detection information thereof, fuel injection of the engine, ignition timing and the like are controlled to operate the engine. For example, since the failure of the sensors causes an inappropriate control of the engine, as a countermeasure, for example, in the technology described in Patent Document 1, the actual value detected by the sensor is compared with the estimated value estimated. The sensor is determined to be broken when there is a divergence between them.

For example, the case where the method of Patent Document 1 is applied to failure determination of an air flow sensor and an intake manifold pressure sensor installed in a naturally aspirated engine will be described. The air flow sensor detects the amount of intake air, the in-manifold pressure sensor detects the in-manifold pressure downstream of the throttle, and the actual measured values thereof are compared with the respective simulated values to determine the presence or absence of a failure.
Each simulated value is basically calculated based on the front / rear pressure ratio of the throttle valve, and the simulated cylinder air introduced into the cylinder by correcting the passing air amount of the throttle valve obtained from the front / back pressure ratio with volumetric efficiency Determine the quantity. Then, the simulated filling efficiency is calculated from the simulated cylinder air amount as a correlation value of the intake amount, and the simulated cylinder air amount is corrected by the intake temperature to calculate the simulated intake manifold pressure.

  In order to determine the failure of the air flow sensor and the intake manifold pressure sensor, it is necessary that the simulated values to be compared with the actual measured values have sufficient reliability. In a naturally aspirated engine, the throttle upstream pressure corresponds to the atmospheric pressure, and the throttle downstream pressure corresponds to the intake manifold pressure. Therefore, the front-to-back pressure ratio is calculated from the atmospheric pressure detected by the atmospheric pressure sensor and the simulated intake manifold pressure (previous value). It is calculated.

  The air flow sensor and the intake manifold pressure sensor are affected by the heat from the engine due to the installation to the engine, but the atmospheric pressure sensor whose installation location is not limited is a place with better thermal environmental conditions, for example, ECU (engine control Unit) is housed. Therefore, the possibility that the atmospheric pressure sensor is broken is extremely low, and it can be assumed that the atmospheric pressure which is the detected value is correct. By applying such atmospheric pressure to the calculation process, the reliability of the simulated filling efficiency and the simulated intake manifold pressure Sex is secured.

JP, 2014-31782, A

However, when the sensor failure determination method described in Patent Document 1 is applied to a supercharged engine, the following problems occur.
The throttle downstream pressure in the supercharged engine corresponds to the in-manifold pressure as in the naturally aspirated engine, but the throttle upstream pressure corresponds to the supercharging pressure by the turbocharger instead of the atmospheric pressure. Therefore, the front / rear pressure ratio of the throttle valve is calculated based on the supercharging pressure and the intake manifold pressure. In order to detect the supercharging pressure, a supercharging pressure sensor is installed on the intake passage in a region between the compressor and the throttle valve, and inevitably receives the influence of heat from the engine. Therefore, the failure frequency of the supercharging pressure sensor is much higher than that of the atmospheric pressure sensor whose thermal environmental condition is good, and it can not be assumed that the detected supercharging pressure is correct like atmospheric pressure.

  Such a situation means a decrease in the reliability of the simulated filling efficiency and the simulated intake manifold pressure calculated based on the supercharging pressure. As a result, even if a deviation occurs between each simulated value and the measured value by the air flow sensor and the intake manifold pressure sensor, the cause of the deviation is the simulated value based on the erroneous boost pressure due to the failure of the boost pressure sensor. It can not be determined whether it is an erroneous operation or an air flow sensor or an in-manifold pressure sensor failure.

The above problems occur not only in a supercharged engine but also in the case where an atmospheric pressure sensor is installed in the intake passage of a naturally aspirated engine. For this reason, the determination method of the sensor failure described in patent document 1 can not be applied as it is to these engines as it was, but a drastic measure was conventionally required from now on.
The present invention has been made to solve such problems, and the object of the present invention is to provide simulated values corresponding to state quantities detected by a plurality of other sensors based on the output value of a specific sensor. The engine sensor failure judging device capable of accurately judging the failure of a specific sensor and the failure of a plurality of other sensors and specifying the failed sensor when calculating the failure of each sensor to determine the failure of the sensor. It is to provide.

  In order to achieve the above object, a sensor failure judging device of an engine according to the present invention is characterized in that, in an engine in which a specific sensor for detecting different state quantities related to intake air and a plurality of other sensors are respectively installed in an intake passage. Simulated value calculation means for calculating simulated values corresponding to the state quantities detected by the plurality of sensors based on output values of the sensors, and detected by the plurality of sensors based on actually measured values of the plurality of sensors Intake correlation index calculation means for calculating an intake correlation index correlating to each other among state quantities, and deviation between intake correlation indices of the plurality of sensors calculated by the intake correlation index calculation means It is calculated as a correlation index deviation, and any one of the plurality of sensors is broken when the correlation index deviation is equal to or greater than a predetermined correlation index determination value. Characterized by comprising a determining failure determination means and (Claim 1).

  According to the sensor failure determination device of the engine configured as described above, while the simulated values corresponding to the state quantities detected by the plurality of sensors are calculated based on the output value of the particular sensor, the actual measurement of the plurality of sensors is performed. Based on the value, an intake correlation index correlated with each other is calculated for each sensor. Then, a deviation between intake correlation indices of a plurality of sensors is calculated as a correlation index deviation, and when the correlation index deviation is equal to or more than the correlation index determination value, it is determined that any one of the plurality of sensors is broken. As a result, it is possible to distinguish between the case where one of the plurality of sensors is broken and the case where the particular sensor is broken.

  As another aspect, when the failure index determination means determines that the correlation index deviation is greater than or equal to the correlation index determination value, each of the actual measurement values detected by the plurality of sensors and each of the sensors calculated by the simulation value calculation means It is preferable that the deviation from the simulated value is calculated as the actual measurement / simulated deviation, and the failure determination is made to the sensor for which the actual measurement / simulated deviation larger than the preset actual measurement / simulative judgment value is calculated (claim 2) .

According to this aspect, when the correlation index deviation is equal to or higher than the correlation index determination value, the deviation between each actual measurement value of the plurality of sensors and the simulation value is calculated as an actual measurement / simulation deviation, and an actual measurement or actual measurement judgment range • A fault determination is made on the sensor for which the simulated deviation has been calculated.
In another aspect, it is preferable that the specific sensor is a throttle upstream pressure sensor which is installed on the upstream side of the throttle valve of the intake passage of the engine and detects the upstream pressure of the throttle valve (Claim 3) .

According to this aspect, as the upstream pressure of the throttle valve, the supercharging pressure is detected by the throttle upstream pressure sensor in the case of a supercharged engine, and the atmospheric pressure is detected in the case of a naturally aspirated engine. Based on the supply pressure and the atmospheric pressure, simulated values corresponding to the state quantities detected by the plurality of sensors are respectively calculated.
In another aspect, the plurality of sensors are installed in an intake passage of the engine to detect an intake air amount of the engine, and are installed downstream of a throttle valve of the intake passage to detect the intake amount of the engine. An intake manifold pressure sensor for detecting intake manifold pressure, and the simulated value calculation means is based on an output value of the throttle upstream pressure sensor, and a simulated value corresponding to an intake amount detected by the air flow sensor, and the intake manifold pressure sensor It is preferable to calculate simulated values corresponding to the detected intake manifold pressure (claim 4).

According to this aspect, based on the output value of the throttle upstream pressure sensor, that is, the supercharging pressure or the atmospheric pressure, the simulation value corresponding to the intake amount detected by the air flow sensor and the intake manifold pressure detected by the intake manifold pressure sensor The simulated values are calculated respectively.
As another aspect, it is preferable that the intake correlation index calculation means calculate the filling efficiency or the intake manifold pressure for each sensor as the intake correlation index correlating to each other between the intake amount and the intake manifold pressure. Item 5).

According to this aspect, the filling efficiency or the intake manifold pressure is calculated for each sensor as an intake correlation index that correlates to each other between the intake amount and the intake manifold pressure.
As another aspect, it is preferable that the failure determination means make a failure determination of the specific sensor when the correlation index deviation is less than a correlation index determination value (claim 6).
According to this aspect, when the correlation index deviation is less than the correlation index determination value, the failure determination of the specific sensor is made.

  According to the sensor failure determination device for an engine of the present invention, when the simulation value corresponding to the state quantities detected by a plurality of other sensors is calculated based on the output value of a specific sensor to determine the failure of the sensor In the above, it is possible to accurately determine the failure of a specific sensor from the failure of a plurality of other sensors and to specify the failed sensor.

BRIEF DESCRIPTION OF THE DRAWINGS It is a whole block diagram which shows the engine to which the sensor failure determination apparatus of this invention was applied. It is a flowchart which shows the sensor failure determination routine performed by ECU. It is explanatory drawing which shows the process of the failure determination by ECU.

Hereinafter, an embodiment of a sensor failure determination device for an engine according to the present invention will be described.
FIG. 1 is a whole block diagram showing an engine to which a sensor failure judging device of the present invention is applied. The engine is mounted on a vehicle (not shown) as a power source for traveling.
A piston 4 is disposed in the cylinder 3 of each cylinder formed in the cylinder block 2 of the engine 1, and each piston 4 slides in the cylinder 3 according to the rotation of the crankshaft 5. The rotation of the crankshaft 5 is transmitted to intake and exhaust camshafts 7, 8 provided on the cylinder head 6, and the camshafts 7, 8 are rotationally driven in synchronization with the crankshaft 5. The intake valve 9 is driven according to the rotation of the intake camshaft 7 to open and close the intake port 11 at a predetermined timing, and the exhaust valve 10 is driven according to the rotation of the exhaust camshaft 8 to open and close the exhaust port 12 at a predetermined timing. Do. VVT mechanisms 13 and 14 are connected to the intake and exhaust camshafts 7 and 8, respectively, and the phase of the intake and exhaust camshafts 7 and 8 with respect to the crankshaft 5 by these VVT mechanisms 13 and 14 and consequently the intake valve 9 and the exhaust valve The opening and closing times of 10 are arbitrarily changed.

  A common surge tank 17 is connected to the intake port 11 of each cylinder via an intake manifold 16, and a downstream end of an intake pipe 18 is connected to the surge tank 17. An air cleaner 19, a compressor of a turbocharger 20, an intercooler 21 and a throttle valve 22 are provided in the intake pipe 18 from the upstream side. Reference numeral 30 denotes a waste gate 30 for controlling the supercharging pressure of the turbocharger 20.

In the present embodiment, the intake manifold of the present invention is constituted by the intake manifold 16, the surge tank 17, the intake pipe 18 and the air cleaner 19.
Further, the upstream end of an exhaust pipe 24 is connected to the exhaust port 12 of each cylinder via an exhaust manifold 23, and the exhaust pipe 24 is provided with a turbine of the turbocharger 20, a catalyst device 29 and a silencer not shown. .

  An in-cylinder injector 25 is disposed in each cylinder so as to face the cylinder, and a port injector 26 is provided in the intake manifold 16 corresponding to each cylinder. The in-cylinder injector 25 is supplied with fuel from a high pressure pump through a low pressure pump (not shown), and the port injector 26 is supplied with fuel from a low pressure pump. An ignition plug 27 is disposed in each cylinder of the engine 1 so as to face in the cylinder, and each ignition plug 27 is ignited by the drive of an igniter 28.

  During operation of the engine 1, the intake air introduced from the air cleaner 19 into the intake pipe 18 is pressurized by the compressor of the turbocharger 20, cooled by the intercooler 21, and then flow-adjusted by the throttle valve 22. Further, the intake air is distributed to each cylinder by the intake manifold 16 through the surge tank 17 and mixed with the fuel injected from the port injector 25 and introduced into the cylinder of the engine 1 as the intake valve 9 is opened. In the cylinder, fuel is injected from the in-cylinder injector 26 into the air-fuel mixture, ignited by the spark plug 27 and burned, and the generated combustion pressure rotates the crankshaft 5 via the piston 4.

Exhaust gas after combustion in the cylinder of each cylinder is discharged to the exhaust port 12 with the opening of the exhaust valve 10, collected by the exhaust manifold 23, and guided to the exhaust pipe 24 to drive the turbine of the turbocharger 20 as a catalyst. It is discharged to the outside through the device 29 and the silencer.
On the other hand, an ECU (engine control unit) provided with an input / output device (not shown), a storage device (ROM, RAM, BURAM, etc.) containing a large number of control programs, a central processing unit (CPU), a timer counter, etc. 31 is installed to perform comprehensive control of the engine 1.

  A throttle position sensor 32 for detecting the opening degree θth of the throttle valve 22 and an air flow sensor 33 for detecting an intake amount V of the engine 1 (corresponding to a plurality of sensors of the present invention) A crank angle sensor 34 outputting a crank angle signal synchronized with the rotation of the engine 1, a water temperature sensor 35 detecting a cooling water temperature Tw of the engine 1, an accelerator sensor 36 detecting an accelerator opening .theta. A supercharging pressure sensor 37 (corresponding to a specific sensor of the present invention, equivalent to a throttle upstream pressure sensor) detecting the supercharging pressure Pboost as pressure, and an intake manifold pressure sensor 38 (inventive pressure detecting Pin as the downstream pressure of the throttle valve 22) Various sensors such as the atmospheric pressure sensor 39 for detecting the atmospheric pressure Pa, etc. Detection information from the capacitors class is input. The supercharging pressure sensor 37 and the intake manifold pressure sensor 38 also have a function of detecting the intake air temperature Tair.

Further, on the output side of the ECU 31, various devices such as the intake and exhaust VVT mechanisms 13 and 14, the port injector 25, the in-cylinder injector 26, the igniter 28, the waste gate 30, and the throttle actuator 40 for opening and closing the throttle valve 22 are described. The kind is connected.
The ECU 31 calculates target values such as fuel injection amount, fuel injection timing, throttle opening, opening / closing timing, closing timing, etc. based on detection information from various sensors and a preset control map etc. Drive controls various devices based on the target value.

  By the way, in the present embodiment, the failure determination processing is executed for the AFS 33 and the intake manifold pressure sensor 38, and for that purpose, actual measurement values of the intake air amount V and the intake manifold pressure Pin are compared with respective simulated values. As will be described later, each simulated value is calculated based on the front-rear pressure ratio of the throttle valve 22 and the engine 1 of the present embodiment is a supercharged type, so the supercharging pressure Pboost detected by the supercharging pressure sensor 37 is throttled. The upstream pressure ratio is calculated, and the front-rear pressure ratio is calculated using the simulated intake manifold pressure Pin (previous value) as the throttle downstream pressure.

  Then, as described in [Problems to be Solved by the Invention], the supercharging pressure sensor 37 installed in the intake passage has a high failure frequency due to the influence of heat from the engine 1, and the detected supercharging pressure Pboost is It can not be assumed that it is correct. Inevitably, even if the reliability of the simulated filling efficiency Ec and the simulated intake manifold pressure Pin calculated based on the boost pressure Pboost lowers and a deviation occurs between the actual measurement value and the simulated value, the boost pressure sensor 37 There is a problem that it can not be determined whether the simulated value is erroneously calculated based on the erroneous supercharging pressure Pboost due to the failure of the above or the failure of the AFS 33 or the intake manifold pressure sensor 38.

  In view of such problems, the inventor focused on the consistency between the intake air amount V detected by the AFS 33 and the intake manifold pressure Pin detected by the intake manifold pressure sensor 38. That is, the engine 1 of the present embodiment corresponds to the charging efficiency Ec (corresponding to the correlated intake correlation index according to the present invention so that the intake amount V of the actual measurement value detected by the AFS 33 can be easily used for various controls. It is converted to the packing efficiency) and handled. The intake manifold pressure Pin (actually measured value), which is a state quantity related to the intake of the engine 1 as well as the intake quantity V, can be converted into the filling efficiency Ec based on the gas equation. Therefore, if the in-manifold pressure Pin detected by the in-manifold pressure sensor 38 is converted into the filling efficiency Ec (hereinafter, referred to as a converted filling efficiency Ec (in)), they can be equivalently compared with each other.

  And, even if there is a divergence between the measured value and the simulated value, if the two filling efficiencies Ec and Ec (in) do not deviate, the AFS 33 and the intake pressure sensor 38 function normally. It can be determined that the simulated value is miscalculated based on the incorrect boost pressure Pboost. Conversely, if there is a divergence between the actual measurement value and the simulation value, and if there is a divergence between the two charging efficiencies Ec and Ec (in) as well, at least one of the sensors 33 and 38 breaks down. It can be determined that the sensor on the side where the measured value and the simulated value deviate from each other is faulty.

Based on the above findings, in the present embodiment, a sensor that has failed is identified based on comparison between the actual charging efficiency Ec obtained from the intake amount V and the conversion charging efficiency Ec (in) obtained from the intake manifold pressure Pin, The details will be described below.
As shown in FIG. 1, the ECU 31 includes a simulation value calculation unit 41 that calculates the simulation charging efficiency Ec and the simulation of the intake manifold pressure Pin based on the boost pressure Pboost detected by the boost pressure sensor 37. Specifically, the pressure ratio between the front and rear of the throttle valve 22 is calculated from the supercharging pressure Pboost detected by the supercharging pressure sensor 37 and the simulated intake manifold pressure Pin (previous value), and the passing air of the throttle valve 22 determined from the front and rear pressure ratio. The amount is corrected by volumetric efficiency to determine the amount of simulated cylinder air introduced into the cylinder. Then, the simulated filling efficiency Ec is calculated from the simulated cylinder air amount as the correlation value of the intake amount V, and the simulated cylinder air amount is corrected by the intake temperature Tair to calculate the simulated intake manifold pressure Pin.

  Further, the ECU 31 includes an intake correlation index calculation means 42 for calculating the charging efficiency Ec as an intake correlation index correlating to each other between the intake amount V and the intake manifold pressure Pin. Specifically, the in-manifold pressure Pin detected by the in-manifold pressure sensor 38 is converted to the converted charging efficiency Ec (in), and the equivalent comparison with the actual charging efficiency Ec already converted from the intake amount V and used for engine control is To be possible.

  When the actual filling efficiency Ec is not converted, the intake correlation index calculating means 42 also executes a process of calculating the actual filling efficiency Ec from the intake amount V. Further, since it is possible to convert the filling efficiency Ec to the intake manifold pressure Pin based on the gas equation, the filling efficiency Ec based on the intake amount V detected by the AFS 33 is converted to the intake manifold pressure Pin and detected by the intake manifold pressure sensor 38 It may be possible to make an equivalent comparison with the intake manifold pressure Pin.

  When the AFS 33 and the intake manifold pressure sensor 38 both are normal and detect the correct measured value, the actual charging efficiency Ec and the conversion charging efficiency Ec (in) are matched with each other by establishing a linear relationship inherent to the engine, Deviation ΔEc (corresponding to the correlation index deviation of the present invention) is always maintained around zero. On the other hand, when an error is included in the measured value due to any sensor failure, the actual filling efficiency Ec and the converted filling efficiency Ec (in) gradually diverge according to the increase of the error, and the deviation ΔEc increases Do. Therefore, the deviation ΔEc can be used as an index to determine whether the AFS 33 and the intake manifold pressure sensor 38 are normal.

Further, the ECU 31 compares the actual charging efficiency Ec with the converted charging efficiency Ec (in) when there is a deviation between each simulated value (simulated filling efficiency Ec, simulated in-manifold pressure Pin) and each measured value. And a failure judging means 43 for specifying a sensor which is in failure.
Specifically, the deviation between the simulated filling efficiency Ec calculated by the simulated value calculating means 41 and the intake amount V (the actual filling efficiency Ec) detected by the AFS 33 is determined, and the simulated inflow manifold calculated by the simulated value calculating means 41 The deviation between the pressure Pin and the intake manifold pressure Pin detected by the intake manifold pressure sensor 38 is determined.

  For these determination processes, actual and simulated judgment values ΔEc (AFS) and ΔPin regarding the filling efficiency Ec and the intake manifold pressure Pin are set in advance, respectively, and the deviation between the simulated filling efficiency Ec and the actual filling efficiency Ec (this invention If the deviation between actual measurement and simulation deviation) is equal to or higher than the actual measurement and simulation judgment value ΔEc (AFS), the failure judgment is made to AFS 33, and the deviation between the simulated intake manifold pressure Pin and the intake manifold pressure Pin is If it deviates to the actual measurement / simulation judgment value ΔPin or more, a failure judgment is made to the intake manifold pressure sensor 38.

Since the difference between the actual measurement value and the simulation value may be due to the failure of the boost pressure sensor 37, the failure determination at this time has a tentative meaning.
When the failure determination means 43 makes a failure determination for at least one of the sensors 33 and 38 based on the comparison between the actual measurement value and the simulation value, the intake correlation index calculation means 42 determines from the intake amount V of the actual measurement value. A deviation ΔEc between the converted actual filling efficiency Ec and the converted filling efficiency Ec (in) converted from the measured intake manifold pressure Pin is calculated. When the deviation ΔEc is less than the correlation index determination value ΔEc0 set in advance, the failure determination of the supercharging pressure sensor 37 is made, and when the deviation ΔEc is the correlation index determination value ΔEc0 or more, the actual measurement of the AFS 33 and the intake manifold pressure sensor 38 A final failure judgment is made on the sensors 33 and 38 on the side (one or both) where the value and the simulated value deviate (fault judgment means).

FIG. 2 is a flowchart showing a sensor failure determination routine executed by the ECU 31. During operation of the engine 1, the ECU 31 executes the routine at predetermined control intervals.
First, the simulated filling efficiency Ec is calculated in step S1, and the simulated intake manifold pressure Pin is calculated in step S2 (simulated value calculating means). In the following step S3, it is determined whether the simulated filling efficiency Ec and the actual filling efficiency Ec of the measured value are diverging or not, and whether the simulated intake manifold pressure Pin and the measured intake manifold pressure Pin are deviated. If there is no divergence with respect to either the charging efficiency Ec or the intake manifold pressure Pin, both the AFS 33 and the intake manifold pressure sensor 38 are considered to be normal (fault determination means), and a determination of No is made in step S3. And once exit the routine.

  If the measured value of either one or both of the charging efficiency Ec and the intake manifold pressure Vin deviates from the simulated value, the determination in step S3 is YES, and the process proceeds to step S4. In step S4, the intake manifold pressure Pin of the actual measurement value is converted into converted filling efficiency Ec (in) (intake air correlation index calculating means), and in the following step S5 deviation ΔEc between actual filling efficiency Ec and converted filling efficiency Ec (in) is calculated Do.

  In step S6, it is determined whether or not the deviation .DELTA.Ec is less than the correlation index determination value .DELTA.Ec0. If the determination is yes, the supercharge pressure sensor 37 is determined to have a failure in step S7 (failure determination means). When the determination in step S6 is No, the process proceeds to step S8, and it is determined whether the deviation determination in step S3 relates to the filling efficiency Ec. When the determination is Yes, the process proceeds to step S9, and it is determined whether or not the deviation determination in step S3 relates to the intake manifold pressure Pin. When the determination is Yes, failure determination is made for each of the AFS 33 and the in-manifold pressure sensor 38 in step S10, and then the routine is ended (failure determination means).

When the determination in step S8 is No, the failure determination is made to the intake manifold pressure sensor 38 in step S11, and when the determination in step S9 is No, the failure determination is made to the AFS 33 in step S12 (failure determination means).
The process of the above-mentioned failure determination by the ECU 31 will be further described based on FIG.
If Yes is determined in step S3 due to the difference between the actual measurement value and the simulation value, the case is divided into three cases shown in FIG. 3 and in any case, the deviation ΔEc is compared with the correlation index determination value ΔEc0 in step S6. Be done.

  In case 1 (AFS 33 failure, intake manifold pressure sensor 38 normal), when deviation ΔEc <ΔEc0, since AFS 33 and intake manifold pressure sensor 38 can be regarded as functioning normally, simulated value based on erroneous boost pressure Pboost Can be determined as being erroneously calculated, and failure determination of the supercharging pressure sensor 37 is made. Further, when the deviation ΔEcEΔEc0, it can be considered that at least one of the sensors 33 and 38 is broken. Since the failure of the AFS 33 has been tentatively determined based on the comparison between the actual measurement value and the simulated value, the failure determination of the AFS 33 is made according to the determination result.

Judgments are made for Case 2 and Case 3 based on the same viewpoint. In case 2 (AFS 33 is normal, intake manifold pressure sensor 38 is defective), determination of failure of supercharging pressure sensor 37 is made when deviation ΔEc <ΔEc0, and when deviation ΔEc 偏差 ΔEc0, determination of failure of intake manifold pressure sensor 38 is made .
In case 3 (both AFS 33 and intake manifold pressure sensor 38 fail), determination of failure of supercharging pressure sensor 37 is made when deviation ΔEc <ΔEc0, and when deviation ΔEc ≧ ΔEc0, determination of failure of AFS 33 and intake manifold pressure sensor 38 is made. Ru.

  As described above, in the present embodiment, when deviation occurs between the measured values of the AFS 33 and the intake manifold pressure sensor 38 and the simulated values, the measured value of the intake manifold pressure Pin is converted to the conversion charging efficiency Ec (in). After conversion, it is compared with the actual filling efficiency Ec based on the intake air amount V. Then, when the conversion charging efficiency Ec (in) and the actual charging efficiency Ec do not deviate (ΔEc <ΔEc0), it is considered that the AFS 33 and the intake manifold pressure sensor 38 function normally, and the supercharging pressure sensor 37 If there is a discrepancy (ΔEc Δ ΔEc0), at least one of AFS 33 and intake manifold pressure sensor 38 is regarded as a failure, and a provisional determination is made based on comparison of measured values and simulated values. The results are finally finalized.

Therefore, the failure of the supercharging pressure sensor 37 and the failure of the AFS 33 and the intake manifold pressure sensor 38 can be accurately determined, and the failure of the AFS 33 and the failure of the intake manifold pressure sensor 38 can be accurately determined. 33 and 38 can be identified with certainty.
Although the description of the embodiment is finished above, the aspect of the present invention is not limited to this embodiment. For example, in the said embodiment, although embodied in the engine 1 provided with the turbocharger 20, it does not restrict to this. For example, even in a naturally aspirated engine in which an atmospheric pressure sensor is installed in the intake passage, the problems described in the [Problems to be Solved by the Invention] occur. It may apply.

1 Engine 16 intake manifold (intake passage)
17 Surge tank (intake passage)
18 Intake pipe (intake passage)
19 Air cleaner (intake passage)
37 Supercharging pressure sensor (specific sensor, throttle upstream pressure sensor)
33 Air flow sensor (multiple sensors)
38 intake manifold pressure sensor (multiple sensors)
31 ECU
41 simulated value calculation means 42 intake correlation index calculation means 43 failure determination means

Claims (6)

In an engine in which a specific sensor for detecting different state quantities relating to intake and a plurality of other sensors are respectively installed in an intake passage,
Simulation value calculation means for respectively calculating simulation values corresponding to the state quantities detected by the plurality of sensors based on the output values of the specific sensor;
Intake correlation index calculation means for calculating, for each sensor, an intake correlation index correlating to each other among state quantities detected by the plurality of sensors based on actual measurement values of the plurality of sensors;
The deviation between the intake correlation indices of the plurality of sensors calculated by the intake correlation index calculation means is calculated as a correlation index deviation, and the plurality of sensors are calculated when the correlation index deviation is greater than a predetermined correlation index determination value. An engine sensor failure judging device comprising: a failure judging means for judging that any one of the sensors is broken. The failure determination means is a deviation between each actual measurement value detected by the plurality of sensors and a simulation value of each sensor calculated by the simulation value calculation unit when the correlation index deviation is equal to or more than the correlation index determination value. The engine according to claim 1, wherein each of the engine and the simulation deviation is calculated as the measurement and simulation deviation, and the failure determination is made to the sensor for which the measurement and simulation deviation larger than the preset measurement and simulation judgment value is calculated. Sensor failure judgment device. The said specific sensor is a throttle upstream pressure sensor which is installed in the upstream of the throttle valve of the intake passage of the said engine, and detects the upstream pressure of this throttle valve, It is characterized by the above-mentioned. Engine sensor failure judgment device. The plurality of sensors are installed in an intake passage of the engine to detect an intake amount of the engine, and are installed downstream of a throttle valve of the intake passage to detect an intake manifold pressure of the engine. It is an intake pressure sensor,
The simulation value calculation means calculates a simulation value corresponding to the intake amount detected by the air flow sensor based on the output value of the throttle upstream pressure sensor, and a simulation value corresponding to the intake manifold pressure detected by the intake manifold pressure sensor. The sensor failure judging device for an engine according to claim 3, wherein each of the sensors is calculated. 5. The intake correlation index calculation means according to claim 4, wherein the filling efficiency or the intake manifold pressure is calculated for each sensor as the intake correlation index correlating to each other between the intake amount and the intake manifold pressure. Engine sensor failure judgment device. The sensor failure of an engine according to any one of claims 1 to 5, wherein the failure determination means makes a failure determination of the specific sensor when the correlation index deviation is less than a correlation index determination value. Judgment device.

Images