JP2022018447A - Diagnostic device for temperature sensor - Google Patents

Abstract

To accurately diagnose a temperature sensor.SOLUTION: There is provided a diagnostic device for a temperature sensor 44 arranged at an inlet side of a selective reduction type NOx catalyst 24 in an exhaust passage 4 of an internal combustion engine 1. The diagnosis device comprises a filter 23 arranged in the exhaust passage, and a regeneration unit 100 automatically regenerating the filter. The diagnosis device calculates a deterioration index of the NOx catalyst on the basis of an exhaust temperature which is detected by the temperature sensor at a finish of the automatic regeneration, and creates a regression equation indicating the number of times of the automatic regeneration on the basis of data of the deterioration index before the calculation of the deterioration index. At completion of this-time automatic regeneration, the diagnosis device estimates a this-time deterioration index on the basis of the regression equation which is created before the finish, and when a difference between the estimated this-time deterioration index and the calculated this-time deterioration index is out of a prescribed normal range, the diagnosis device diagnoses that the temperature sensor is in a failure.SELECTED DRAWING: Figure 1

Description

The present disclosure relates to a temperature sensor diagnostic device, and more particularly to a diagnostic device for diagnosing a temperature sensor arranged on the inlet side of a selective reduction NOx catalyst in an exhaust passage of an internal combustion engine.

Generally, for diagnostic methods in this type of diagnostic device, the voltage as the output of the temperature sensor is monitored and this voltage is compared to the upper and lower thresholds that define the normal range. Then, when the voltage continuously exceeds the upper limit threshold value or falls below the lower limit threshold value for a predetermined time or more, the temperature sensor is diagnosed as a failure.

Japanese Unexamined Patent Publication No. 2005-307745

However, if a high voltage exceeding the upper limit threshold value or a low voltage lower than the lower limit threshold value is intermittently generated from the temperature sensor for a shorter time than a predetermined time, the temperature sensor cannot be diagnosed as a failure. On the other hand, in order to avoid this, it is conceivable to set a predetermined time very short, but there is a possibility that a failure may be misdiagnosed due to temporary electrical noise unrelated to the failure. Therefore, it is practically impossible to set a predetermined time very short.

Therefore, the present disclosure was devised in view of such circumstances, and an object thereof is to provide a temperature sensor diagnostic device capable of accurately diagnosing a temperature sensor.

According to one aspect of the present disclosure.
It is a diagnostic device of a temperature sensor arranged on the inlet side of the selective reduction NOx catalyst in the exhaust passage of an internal combustion engine.
A filter placed in the exhaust passage and collecting particulate matter in the exhaust,
A playback unit configured to perform automatic playback of the filter, and
A calculation unit configured to calculate the deterioration index of the NOx catalyst based on the exhaust temperature detected by the temperature sensor when the automatic regeneration is completed.
A creation unit configured to create a regression equation that represents the relationship between the number of autoplay completions and the degradation index at the completion of autoplay, based on previous degradation index data.
At the completion of this automatic reproduction, the deterioration index of this time is estimated based on the regression equation created by the creation unit before that, and the estimated deterioration index of this time and the current deterioration index calculated by the calculation unit are calculated. A diagnostic unit configured to diagnose the temperature sensor as a failure when the difference from the deterioration index of is out of the predetermined normal range.
A temperature sensor diagnostic device is provided that comprises.

Preferably, the calculation unit calculates a larger deterioration index as the exhaust temperature detected by the temperature sensor is higher.
The diagnostic unit calculates the difference by subtracting the estimated current deterioration index from the calculated current deterioration index, and the difference falls within the normal range of the lower limit threshold value or more and the upper limit threshold value or less. Judge whether it is included or not,
The absolute value of the upper threshold is greater than the absolute value of the lower threshold.

Preferably, at the completion of the current automatic regeneration, when the temperature sensor is diagnosed as a failure by the diagnostic unit, the creation unit creates a regression equation using the estimated deterioration index of the present time.

Preferably, the regression equation is represented by Y = aX (where Y is the deterioration index, X is the number of automatic regeneration completions, and a is a predetermined coefficient).

Preferably, the diagnostic device further comprises a manual regeneration unit configured to perform manual regeneration of the filter.
The calculation unit calculates the deterioration index of the NOx catalyst based on the exhaust temperature detected by the temperature sensor when the automatic regeneration and the manual regeneration are completed.
When the automatic reproduction and the manual reproduction are completed, the creation unit creates a regression equation representing the number of automatic reproduction completions and the relationship between the number of manual reproduction completions and the deterioration index based on the data of the deterioration index before that.
At the completion of the automatic regeneration or the manual regeneration, the diagnostic unit estimates the deterioration index of the present time based on the regression equation created by the creation unit before that, and the estimated deterioration index and the calculation thereof. The temperature sensor is configured to diagnose a failure when the difference from the deterioration index calculated by the unit is out of a predetermined normal range.

Preferably, the regression equation is represented by Y = a1, X1 + a2, X2 (where Y is the deterioration index, X1 is the number of automatic reproduction completions, X2 is the number of manual reproduction completions, and a1 and a2 are predetermined coefficients).

According to the present disclosure, the temperature sensor can be diagnosed with high accuracy.

It is a schematic diagram of an internal combustion engine. It is a graph which shows the correlation of the deterioration index and the number of times of automatic reproduction completion. It is a graph for demonstrating the diagnostic method of a temperature sensor. It is a flowchart of a diagnostic routine.

Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. It should be noted that the present disclosure is not limited to the following embodiments.

FIG. 1 is a schematic view of an internal combustion engine according to the present embodiment. The internal combustion engine (also referred to as an engine) 1 is a multi-cylinder compression ignition type internal combustion engine mounted on a vehicle, specifically, an in-line 4-cylinder diesel engine. The vehicle is a large vehicle such as a truck. However, there are no particular restrictions on the type, type, application, etc. of the vehicle and internal combustion engine. For example, the vehicle may be a small vehicle such as a passenger car, and the engine may be a spark-ignition internal combustion engine (for example, a gasoline engine).

The engine may be mounted on a moving body other than a vehicle, for example, a ship, a construction machine, or an industrial machine. Further, the engine does not have to be mounted on a moving body, and may be a stationary engine.

The engine 1 includes an engine main body 2, an intake passage 3 and an exhaust passage 4 connected to the engine main body 2, and a fuel injection device 5. The engine body 2 includes structural parts such as a cylinder head, a cylinder block, and a crankcase, and moving parts such as a piston, a crankshaft, and a valve housed therein. The intake and exhaust flows are indicated by white arrows and black arrows, respectively.

The fuel injection device 5 comprises a common rail type fuel injection device, and includes a fuel injection valve provided in each cylinder, that is, an injector 7, and a common rail 8 connected to the injector 7. The injector 7 injects fuel directly into the cylinder 9. The common rail 8 stores the fuel injected from the injector 7 in a high pressure state.

The intake passage 3 is mainly defined by an intake manifold 10 connected to the engine body 2 (particularly a cylinder head) and an intake pipe 11 connected to the upstream end of the intake manifold 10. The intake pipe 11 is provided with an air cleaner 12, an air flow meter 13, a compressor 14C of a turbocharger 14, an intercooler 15, and an electronically controlled intake throttle valve 16 in this order from the upstream side. The air flow meter 13 is a sensor (intake flow rate sensor) for detecting the intake air amount (intake flow rate) per unit time of the engine 1.

The exhaust passage 4 is mainly defined by an exhaust manifold 20 connected to the engine body 2 (particularly a cylinder head) and an exhaust pipe 21 arranged on the downstream side of the exhaust manifold 20. A turbine 14T of a turbocharger 14 is provided between the exhaust pipe 21 or the exhaust manifold 20 and the exhaust pipe 21. The exhaust pipe 21 on the downstream side of the turbine 14T is provided with an exhaust pipe injector 35, an oxidation catalyst 22, a filter 23, a urea water injection valve 25, a selective reduction NOx catalyst 24, and an ammonia oxidation catalyst 26 in this order from the upstream side. ..

The oxidation catalyst 22 oxidizes and purifies the unburned components (hydrocarbon HC and carbon monoxide CO) in the exhaust gas, heats and raises the exhaust gas with the heat of reaction at this time, and NO ₂ in the exhaust gas. Oxidizes to. The filter 23 is formed by a continuously regenerating type catalytic filter, collects particulate matter (PM (Particulate Matter)) in the exhaust gas, and continuously burns and removes the collected PM. The NOx catalyst 24 reduces NOx in the exhaust by using ammonia derived from urea water injected from the urea water injection valve 25 as a reducing agent. The ammonia oxidation catalyst 26 oxidizes and purifies the excess ammonia discharged from the NOx catalyst 24.

The engine 1 includes an EGR device 30. The EGR device 30 includes an EGR passage 31 for recirculating a part of the exhaust gas (referred to as EGR gas) in the exhaust passage 4 (particularly in the exhaust manifold 20) into the intake passage 3 (particularly in the intake manifold 10), and an EGR passage. An EGR cooler 32 for cooling the EGR gas flowing through the 31 and an EGR valve 33 for adjusting the flow rate of the EGR gas are provided.

Attached to the engine 1, a control unit, a circuit element (circuitry), or an electronic control unit (called an ECU (Electronic Control Unit)) 100 forming a controller is provided.

The ECU 100 includes the above-mentioned air flow meter 13, a rotation speed sensor 40 for detecting the rotation speed of the engine (specifically, the rotation speed per minute (rpm)), and an accelerator for detecting the accelerator opening degree. The opening sensor 41 is electrically connected. Further, in the ECU 100, a differential pressure sensor 45 for detecting the differential pressure of the exhaust pressure on the inlet side and the outlet side of the filter 23 and a temperature sensor 44 for detecting the exhaust temperature on the inlet side of the NOx catalyst 24 are electrically connected. Is connected. A manual regeneration switch 46 is also electrically connected to the ECU 100.

The ECU 100 controls the above-mentioned various devices, that is, the injector 7, the intake throttle valve 16, the EGR valve 33, the exhaust pipe injector 35, and the urea water injection valve 25, based on the outputs of these sensors.

The ECU 100 periodically executes filter regeneration (or filter regeneration control) for forcibly burning and removing PM collected by the filter 23. This filter reproduction includes automatic reproduction (or automatic reproduction control) and manual reproduction (or manual reproduction control). First, automatic reproduction will be described.

The differential pressure detected by the differential pressure sensor 45 is a value that correlates with the PM collection amount of the filter 23, and the differential pressure increases as the PM collection amount increases. When the differential pressure detected by the differential pressure sensor 45 exceeds a predetermined upper limit threshold value, the ECU 100 starts automatic regeneration on condition that a predetermined automatic regeneration execution condition is satisfied. During automatic regeneration, the ECU 100 injects fuel from the exhaust pipe injector 35. Then, the injected fuel is burned by the oxidation catalyst 22, the exhaust gas is heated, and the heated exhaust gas is supplied to the filter 23 to burn off the PM accumulated in the filter 23. In addition, instead of or in addition to the fuel injection by the exhaust pipe injector 35, post injection or the like may be performed by the in-cylinder injection injector 7. After that, when the differential pressure falls below a predetermined lower limit threshold value, the ECU 100 ends the automatic reproduction.

Next, manual reproduction will be described. Even if the differential pressure exceeds the upper limit threshold value, the automatic reproduction execution condition is not satisfied, so that the automatic reproduction may not be started. In this case, the ECU 100 activates a warning device (warning light or the like) (not shown) and requests the driver to perform manual reproduction. When the driver turns on the manual regeneration switch 46 in response to this request, the ECU 100 starts manual regeneration. The control during manual playback is the same as during automatic playback.

Further, the ECU 100 is configured to calculate the deterioration index Y of the NOx catalyst 24 based on the exhaust temperature on the inlet side of the NOx catalyst 24 detected by the temperature sensor 44. The deterioration index Y is an index value indicating the degree of deterioration of the NOx catalyst 24, and the deterioration index Y increases as the degree of deterioration of the NOx catalyst 24 increases. In the case of the present embodiment, the initial value of the deterioration index Y when the NOx catalyst 24 is new (zero degree of deterioration) is zero, but it may be a value other than zero.

The ECU 100 repeatedly calculates or calculates the deterioration index Y every extremely short predetermined calculation cycle (for example, 10 ms). For convenience, this is called constant calculation. In particular, the ECU 100 is configured to calculate the deterioration index Y when the automatic reproduction is completed. The interval between each completion of autoplay is, of course, much longer than the calculation cycle.

The deterioration index Y can be calculated by any method including a known method. For example, the ECU 100 calculates the deterioration index Y based on the integrated values of the exhaust temperature and the heat generation time detected by the temperature sensor 44. The calculated deterioration index Y becomes larger as the exhaust temperature detection value is higher and the heat generation time is longer. The heat-heated time is synonymous with the exhaust gas introduction time into the NOx catalyst 24, which is calculated from the time when the NOx catalyst 24 is new. The heat generation time is counted by the ECU 100.

The following alternative methods are also possible. The ECU 100 calculates the deterioration index increase amount (Δy _n ) in one calculation cycle at the current calculation time (t _n ) based on the current exhaust temperature detection value (T _n ). Then, this deterioration index increase amount (Δ y _n ) is added to the deterioration index (Y _n-1 ) calculated at the previous calculation time (t _n-1 ) to obtain the current deterioration index (Y _n ). That is, the deterioration index Y can be suitably obtained by sequentially integrating the amount of increase in the deterioration index (Δy _n ) during one calculation cycle (Y _n = Y _n-1 + Δy _n ). The deterioration index increase amount (Δy _n ) may be calculated using a predetermined map (may be a function; the same applies hereinafter).

The deterioration index Y thus calculated can be used for various purposes. For example, it can be used to determine the replacement time of the NOx catalyst 24, such as diagnosing the NOx catalyst 24 as abnormal when the deterioration index Y exceeds a predetermined threshold value and prompting the replacement thereof. Alternatively, it can be used for investigating the cause when the NOx concentration of the exhaust gas on the downstream side of the NOx catalyst 24 becomes high. For example, if the NOx concentration becomes high when the value of the deterioration index Y is large, it can be inferred that the deterioration of the NOx catalyst 24 is the cause. On the other hand, when the NOx concentration becomes high when the value of the deterioration index Y is small, it can be inferred that the cause is an abnormality other than the NOx catalyst 24. By referring to the value of the deterioration index Y in this way, the cause can be isolated.

By the way, the temperature sensor 44 may break down due to deterioration over time. If the temperature sensor 44 fails, the calculation of the deterioration index Y and the exhaust emission are adversely affected. Therefore, in this embodiment, a diagnostic device for the temperature sensor 44 is provided.

The main cause of failure of the temperature sensor 44 is an abnormality in the electrical system such as noise generation or disconnection due to an abnormality in the sensor signal circuit. When such an abnormality occurs, an abnormal value voltage is output from the sensor and the signal circuit connecting the sensor. Conventionally, the voltage is monitored and compared with the upper and lower thresholds that define the normal range. If the voltage is continuously high for a predetermined time or longer and exceeds the upper limit threshold value, or is low voltage below the lower limit threshold value, the temperature sensor 44 is diagnosed as having a failure. The output voltage of the normal temperature sensor 44 increases as the exhaust temperature increases.

However, if a high voltage exceeding the upper limit threshold value or a low voltage lower than the lower limit threshold value occurs intermittently for a shorter time than a predetermined time, the temperature sensor 44 cannot be diagnosed as a failure. On the other hand, in order to avoid this, it is conceivable to set the predetermined time very short, but this may lead to a misdiagnosis of failure due to temporary electrical noise unrelated to the failure. .. Therefore, it is practically impossible to set a predetermined time very short.

Therefore, in the present embodiment, the temperature sensor 44 is diagnosed by a new diagnostic method different from the conventional one. Hereinafter, the diagnostic method of this embodiment will be described in detail.

As a result of diligent research, the present inventor has newly found that the deterioration index Y correlates with the number of automatic regeneration completion times X, as shown in FIG. The deterioration index Y can be expressed by, for example, a linear regression equation of Y = aX. a is a predetermined coefficient. In this case, the deterioration index Y is proportional to the number of automatic reproduction completion times X.

In this embodiment, the temperature sensor 44 is diagnosed using this correlation. As shown in FIG. 3, when the automatic regeneration is completed, the ECU 100 calculates the deterioration index Y based on the detection value of the temperature sensor 44, and the calculated value is combined with the automatic regeneration completion count X (in the figure). (Indicated by white circles). This data can be expressed as (X, Y).

The number of times the i-th automatic reproduction is completed (i = 0, 1, 2, 3 ...) Is represented by X _i . In the illustrated example, the temperature sensor 44 fails after the X _{i-th time, which is the number of times the automatic reproduction is completed, and after the X i- th} time, which is one time before that. In this case, since the temperature sensor 44 is out of order when the _Xith automatic regeneration is completed, the abnormal deterioration index _Yi is calculated based on the abnormal exhaust temperature detection value, and the deterioration is abnormally high as shown in the illustrated example. The index Y _i is calculated. Although not shown, an abnormally low deterioration index Y _i may be calculated due to a failure of the temperature sensor 44.

Therefore, the ECU 100 diagnoses the temperature sensor 44 by comparing the calculated deterioration index (referred to as the calculated deterioration index) Y _i with the deterioration index (referred to as the estimated deterioration index) Y ^ _i estimated based on the regression equation. ..

When the automatic reproduction is completed, the ECU 100 creates a regression equation based on the deterioration index data (X, Y) before that. That is, for example, when the X _i-1st (previous) automatic reproduction is completed, the ECU 100 has a regression equation as shown by the line L based on all the data (including the X _i-1st data) before that. Create Y = aX. At the time of this preparation, a known method such as the least squares method is used. At the time of creation, the ECU 100 substantially determines the coefficient a.

Further, the ECU 100 is based on the regression equation created at the completion of the X _i -th (this time) automatic reproduction, before that, specifically at the time of the latest X _i-first (previous) automatic reproduction completion. The deterioration index Y ^ _i of the _i -th time (this time) is estimated, and the deterioration index Y _i of the Xi-th time (this time ₎ is calculated based on the detected value of the temperature sensor 44. Then, when the difference between the estimated current deterioration index Y ^ _i and the calculated current deterioration index Y _i is outside the predetermined normal range, the ECU 100 diagnoses the temperature sensor 44 as a failure.

In the present embodiment, the ECU 100 calculates the difference (referred to as exponential difference) ΔY _i = Y _i −Y ^ _i obtained by subtracting the estimated deterioration index Y ^ _i this time from the calculated deterioration index Y _i this time, and this index difference. It is determined whether or not ΔY _i is within the predetermined normal range defined by −α ≦ ΔY _i ≦ + β (provided that α, β> 0). -Α is the lower threshold of the normal range, and + β is the upper threshold of the normal range. The ECU 100 diagnoses the temperature sensor 44 as normal when the exponential difference ΔY _i is within the normal range (when it is within the normal range), and when the exponential difference ΔY _i is not within the normal range (outside the normal range). (At one time), the temperature sensor 44 is diagnosed as a failure. The example shown in FIG. 3 is an example of diagnosing the temperature sensor 44 as a failure because the exponential difference ΔYi is not within the normal range.

As described above, in the present embodiment, the temperature sensor 44 is diagnosed based on the index difference ΔY representing the difference between the estimated deterioration index Y ^ and the calculated deterioration index Y. Since the reference estimated deterioration index Y ^ is calculated from the integrated value of the exhaust temperature and the heated time detected before the failure occurs, the abnormally high voltage or low voltage is short and intermittent from the temperature sensor 44 or the like after the failure occurs. Even if it occurs, it is hard to be affected by it. Therefore, even if an abnormality occurs in which the high voltage or the low voltage is short and intermittently occurs, it can be accurately diagnosed as a failure of the temperature sensor 44, and the temperature sensor 44 can be diagnosed accurately.

The absolute value (magnitude) | + β | of the upper limit threshold value can be determined, for example, as the value of a predetermined ratio (B%) of the estimated deterioration index Y ^ _i this time. Similarly, the absolute value (magnitude) | −α | of the lower limit threshold value can be set as the value of the predetermined ratio (A%) of the estimated deterioration index Y ^ _i this time.

By the way, when the output voltage of the temperature sensor 44 becomes abnormally higher than the normal value, the causes are that the temperature sensor 44 fails and its output voltage becomes high, and that the temperature sensor 44 is normal but the actual exhaust temperature is high. There are two possible cases, one is when the output voltage is high because the voltage is high. Therefore, in order to prevent misdiagnosis as a failure in the latter case, it is preferable that the absolute value | + β | of the upper limit threshold value is larger than the absolute value | −α | of the lower limit threshold value. Here, the absolute value | + β | of the upper limit threshold value is a value corresponding to the maximum permissible deviation amount of the output voltage to the high temperature side with respect to the normal value. Further, the absolute value | −α | of the lower limit threshold value is a value corresponding to the maximum permissible deviation amount of the output voltage to the low temperature side with respect to the normal value.

However, the method of setting the absolute value | + β | of the upper limit threshold value and the absolute value | −α | of the lower limit threshold value is arbitrary, and both may be set equally to each other in view of the experimental results and the like. However, the latter may be set larger than the former.

Since the estimated deterioration index Y ^ is calculated based on the integrated value of the heat-heated time, Y ^ _i-1 <Y ^ _i always holds.

Next, the diagnostic routine in this embodiment will be described with reference to FIG. The illustrated routine is repeatedly executed by the ECU 100 at predetermined calculation cycles.

First, in step S101, the ECU 100 determines whether or not the Xith automatic reproduction (this time ( _i )) is completed, that is, whether or not the current time is the timing when the Xith automatic reproduction is completed.

If it is not the timing when the automatic reproduction is completed, the ECU 100 immediately ends the routine. On the other hand, when the automatic reproduction is completed, the ECU 100 proceeds to step S102.

In step S102, the ECU 100 substitutes X = X _i for the regression equation: Y = aX created in the X _i-1st time (previous (i-1)), and estimates the deterioration index Y ^ _i this time.

Next, in step S103, the ECU 100 calculates the deterioration index Y _i this time based on the detection value of the temperature sensor 44.

In step S104, the ECU 100 calculates the index difference ΔY _i between the estimated deterioration index Y ^ _i and the calculated deterioration index Y _i .

In step S105, the ECU 100 determines whether or not the exponential difference ΔY _i is within the normal range, that is, whether or not −α ≦ ΔY _i ≦ + β is satisfied.

If it is not within the normal range, the ECU 100 proceeds to step S106 and diagnoses the temperature sensor 44 as a failure. At this time, preferably, the ECU 100 activates a warning device (not shown) to urge the user to perform necessary inspections and maintenance such as replacement of the temperature sensor 44.

Next, in step 107, the ECU 100 replaces the estimated deterioration index Y ^ _i with the calculated deterioration index Y _i . Then, in step 108, the regression equation of this time is created based on all the data (X, Y) before this time including the calculated deterioration index Y _i of this time, and the routine is completed.

On the other hand, when the exponential difference ΔY _i is within the normal range in step S105, the ECU 100 replaces the calculated deterioration index Y _i with the calculated deterioration index Y _i as it is in step 109, and creates the current regression equation in step 108. Finish the routine.

In this routine, when the failure of the temperature sensor 44 is detected for the first time at the X _i times and the transition is made to the X _{i + 1} times as it is, in step S102, the regression equation created at the X _i times is used to X _{i +.} The _first estimated deterioration index Y ^ _{i + 1} is calculated. At the X _i times, the regression equation is created using the estimated deterioration index Y ^ _i , which is close to the normal value, instead of the abnormal calculated deterioration index Y _i . Therefore, the estimated deterioration index Y ^ _{i + 1} calculated at the X _{i + 1st} time can also be a value close to the normal value. Then, in step S108, a regression equation is created substantially using this estimated deterioration index Y ^ _{i + 1} . Therefore, even at the next X _{i + 2nd} time, the estimated deterioration index Y ^ _{i + 2} close to the normal value can be calculated, and the diagnosis can be performed accurately.

Next, a modified example will be described. The same parts as those of the basic embodiment will be omitted, and the differences from the basic embodiment will be mainly described below.

It is conceivable that the deterioration index Y correlates not only with the number of automatic reproduction completions X but also with the number of manual reproduction completions. Therefore, in consideration of this case as well, in this modification, the ECU 100 creates a regression equation including the number of manual regeneration completions.

When the number of automatic reproduction completions is X1 and the number of manual reproduction completions is X2, the regression equation can be expressed by, for example, a multiple regression equation of Y = a1 · X1 + a2 · X2. Here, a1 and a2 are predetermined coefficients.

As a result, a more accurate regression equation can be created in consideration of the number of manual regeneration completions, and by extension, the diagnosis can be performed accurately by using the regression equation.

In this modification, the ECU 100 calculates the deterioration index Y based on the exhaust temperature detected by the temperature sensor 44 at the completion of the automatic regeneration and the manual regeneration, and the deterioration index data (X1, X2, Y) before that. Based on the above, a regression equation expressing the relationship between the number of automatic reproduction completions X1 and the number of manual reproduction completions X2 and the deterioration index Y is created. Then, the ECU 100 estimates the deterioration index of this time based on the regression equation created at the completion of the previous automatic reproduction or manual reproduction at the completion of the automatic reproduction or the manual reproduction of the present time, and calculates the deterioration index of the present time. When the difference between the estimated deterioration index and the calculated deterioration index is out of the normal range, the temperature sensor 44 is diagnosed as a failure.

As can be understood from the above description, the ECU 100 of the present embodiment constitutes a regeneration unit, a calculation unit, a creation unit, a diagnosis unit, and a manual regeneration unit as defined in the claims. Each of these units may, of course, be configured individually or in combination of several.

Although the embodiments of the present disclosure have been described in detail above, various other embodiments and variations of the present disclosure can be considered.

(1) In the above embodiment, the estimated deterioration index Y ^ is subtracted from the calculated deterioration index Y to calculate the difference between the two, that is, the exponential difference ΔY, but the calculation method and the expression method of the difference between the two are arbitrary. For example, the difference may be calculated by subtracting the calculated deterioration index Y from the estimated deterioration index Y ^. Alternatively, the difference between the two can be expressed as a ratio. Y / Y ^ or Y ^ / Y can be used as a value representing the difference between the two. Those skilled in the art will know how to change the threshold value, etc. due to the change of these expressions.

(2) In the above embodiment, when the automatic reproduction is completed this time, a regression equation is created based on all the data (X, Y) before this time. By doing so, the regression equation is created using the data of the maximum number of samples, so that the regression equation can be created with high accuracy.

On the other hand, it is not always necessary to use all the data. You may extract some data out of all the data and create a regression equation based on this part of the data. By doing so, the calculation load of the ECU 100 can be reduced by reducing the number of samples.

For example, some data may be extracted from all data at equal or non-equal intervals, some data may be randomly extracted, or data may be extracted every predetermined number of times (for example, 10 times). It is possible to do it.

This point is the same in the modified example in consideration of the number of times of manual reproduction completion.

(3) The format of the regression equation can be arbitrarily set according to the experimental results and the like. For example, in the case of the basic embodiment in which only the number of automatic reproduction completions X is considered, the regression equation can be in the form of Y = aX + b. b is a predetermined constant or intercept.

When a regression equation is created based on the obtained data, the straight line L representing the regression equation may not pass through the origin (X ₀ , 0). Further, depending on how the deterioration index Y is defined, the initial value of the deterioration index Y when the NOx catalyst 24 is new may not be zero. In such a case, it is convenient to set the format of the regression equation as Y = aX + b.

The same applies to the modification in consideration of the number of manual reproduction completions, and for example, the regression equation can be set in the format of Y = a1, X1 + a2, X2 + c. c is a predetermined constant or intercept.

(4) When only post-injection by the in-cylinder injection injector 7 is performed during filter regeneration, the exhaust pipe injector 35 may be omitted.

(5) The timing for starting the automatic regeneration of the filter 23 is the differential pressure on the upstream side and the downstream side thereof, or in addition to this, the vehicle mileage or the integrated fuel injection amount from the in-cylinder injection injector 7. It may be determined based on other parameters such as.

The embodiments of the present disclosure are not limited to the above-described embodiments, and all modifications, applications, and equivalents included in the ideas of the present disclosure defined by the scope of claims are included in the present disclosure. Therefore, this disclosure should not be construed in a limited way and may be applied to any other technique that falls within the scope of the ideas of this disclosure.

1 Internal combustion engine (engine)
4 Exhaust passage 23 Filter 24 Selective reduction type NOx catalyst 44 Temperature sensor 100 Electronic control unit (ECU)

Claims (6)

It is a diagnostic device of a temperature sensor arranged on the inlet side of the selective reduction NOx catalyst in the exhaust passage of an internal combustion engine.
A filter placed in the exhaust passage and collecting particulate matter in the exhaust,
A playback unit configured to perform automatic playback of the filter, and
A calculation unit configured to calculate the deterioration index of the NOx catalyst based on the exhaust temperature detected by the temperature sensor when the automatic regeneration is completed.
A creation unit configured to create a regression equation that represents the relationship between the number of autoplay completions and the degradation index at the completion of autoplay, based on previous degradation index data.
At the completion of this automatic reproduction, the deterioration index of this time is estimated based on the regression equation created by the creation unit before that, and the estimated deterioration index of this time and the current deterioration index calculated by the calculation unit are calculated. A diagnostic unit configured to diagnose the temperature sensor as a failure when the difference from the deterioration index of is out of the predetermined normal range.
A diagnostic device for a temperature sensor, characterized in that it comprises. The calculation unit calculates a larger deterioration index as the exhaust temperature detected by the temperature sensor is higher.
The diagnostic unit calculates the difference by subtracting the estimated current deterioration index from the calculated current deterioration index, and the difference falls within the normal range of the lower limit threshold value or more and the upper limit threshold value or less. Judge whether it is included or not,
The diagnostic device for a temperature sensor according to claim 1, wherein the absolute value of the upper limit threshold value is made larger than the absolute value of the lower limit threshold value. When the temperature sensor is diagnosed as a failure by the diagnostic unit at the completion of the automatic regeneration, the creation unit creates a regression equation using the estimated deterioration index of the present time according to claim 1 or 2. The diagnostic device for the temperature sensor described. The regression equation is represented by Y = aX (where Y is the deterioration index, X is the number of automatic regeneration completions, and a is a predetermined coefficient).
The diagnostic device for a temperature sensor according to any one of claims 1 to 3. Further equipped with a manual regeneration unit configured to perform manual regeneration of the filter,
The calculation unit calculates the deterioration index of the NOx catalyst based on the exhaust temperature detected by the temperature sensor when the automatic regeneration and the manual regeneration are completed.
When the automatic reproduction and the manual reproduction are completed, the creation unit creates a regression equation representing the number of automatic reproduction completions and the relationship between the number of manual reproduction completions and the deterioration index based on the data of the deterioration index before that.
At the completion of the automatic regeneration or the manual regeneration, the diagnostic unit estimates the deterioration index of the present time based on the regression equation created by the creation unit before that, and the estimated deterioration index and the calculation thereof. The temperature sensor according to any one of claims 1 to 3, which is configured to diagnose the temperature sensor as a failure when the difference from the deterioration index calculated by the unit is out of a predetermined normal range. Diagnostic device. The regression equation is represented by Y = a1, X1 + a2, X2 (where Y is the deterioration index, X1 is the number of automatic reproduction completions, X2 is the number of manual reproduction completions, and a1 and a2 are predetermined coefficients).
The diagnostic device for a temperature sensor according to claim 5.

Images