JP2012077668A - Sensor control device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To properly make a failure diagnosis of a particulate matter detection sensor (PM sensor) on the upstream and downstream sides of a filter.SOLUTION: The PM filter 16 is equipped in a discharge tube 14 of an engine 11, and the PM sensors 18 and 19 are equipped on the upstream and downstream sides of the PM filter 16, respectively. An ECU 20 executes the sensor regeneration treatment (PM forced combustion) of the PM sensors 18 and 19 at the same time, based on the PM deposition detected by the PM sensors 18 and 19. The ECU 20 also executes the failure diagnosis of the PM sensor 18 and 19, based on the order of an execution request (burning request) of the next generation treatment of the upstream side PM sensor 18. and an execution request (combustion request) of the next generation treatment of the downstream side PM sensor 19, after the execution of the regeneration treatments of the PM sensors 18 and 19.

Description

  The present invention relates to a sensor control device that monitors the amount of particulate matter (PM) in exhaust gas based on a detection signal of a particulate matter detection sensor.

  Conventionally, various PM sensors (particulate matter detection sensors) for detecting the amount of PM discharged from an engine have been proposed. For example, in the PM sensor of Patent Document 1, a pair of counter electrodes is provided on an insulating substrate, and the resistance between the electrodes changes when PM adheres between the pair of counter electrodes. The detection signal is output. Then, the PM amount is calculated based on the detection signal of the PM sensor.

JP 59-196453 A

  By the way, a PM filter for collecting PM may be provided in the exhaust pipe of the engine, and in a configuration including the PM filter, filter regeneration processing for removing the collected PM by combustion is performed as necessary. In such a case, a pressure sensor for detecting the differential pressure across the PM filter is provided, and when the differential pressure detected by the pressure sensor exceeds a predetermined value, it is determined that the filter regeneration process has been performed, and PM regeneration is performed. Processing is performed.

  However, in the configuration in which the execution timing of the filter regeneration process is determined based on the differential pressure before and after the filter measured by the pressure sensor, the filter regeneration process is caused by a difference in the correlation between the actual PM collection amount and the detected value of the pressure sensor. May not be implemented at the timing that should be implemented. Particularly, in recent years, there is a tendency that the execution timing of the filter regeneration process is advanced with the strengthening of exhaust gas regulations. In such a case, the deviation of the execution timing of the PM regeneration process becomes a further problem.

  Therefore, a configuration is conceivable in which PM sensors are respectively provided before and after the PM filter, and the execution timing of the filter regeneration processing is determined based on the outputs of these PM sensors. Since this configuration can directly monitor the amount of PM collected by the PM filter, it is considered that the execution timing of the PM regeneration process can be suitably managed.

  In addition, under recent legal regulations, it is considered that it is required to perform abnormality diagnosis of PM sensors in the configuration in which PM sensors are provided before and after the PM filter as described above.

  The main object of the present invention is to provide a sensor control device that can suitably perform abnormality diagnosis for particulate matter detection sensors (PM sensors) on the upstream side and the downstream side of a filter.

  Hereinafter, means for solving the above-described problems and the effects thereof will be described.

  According to the sensor control device of the present invention, a filter device that collects conductive particulate matter contained in the exhaust gas of the internal combustion engine is provided in an exhaust pipe of the internal combustion engine, and an upstream side and a downstream side of the filter device. Further, the present invention is applied to a system in which an upstream sensor and a downstream sensor are provided as particulate matter detection sensors for attaching the particulate matter and detecting the amount of the particulate matter attached thereto.

  In the first aspect of the invention, the first combustion means for combusting the particulate matter adhering to the upstream sensor by the heating of the heater provided in the upstream sensor, and the downstream sensor A second combustion means for burning the particulate matter adhering to the downstream sensor by heating of a heater provided in the downstream sensor, and the adhesion amount of the particulate matter detected by the upstream sensor is determined by the first sensor. First combustion requesting means for issuing a combustion request for the upstream sensor when the standard value for burning particulate matter by the combustion means is reached, and adhesion of particulate matter detected by the downstream sensor Second combustion requesting means for issuing a combustion request for the downstream sensor when the amount reaches an implementation reference value for burning particulate matter by the second combustion means; and the first combustion means Based on the combustion request issued by the first combustion requesting unit after the combustion is performed and the combustion request issued by the second combustion requesting unit after the combustion is performed by the second combustion unit, the upstream sensor and the An abnormality determining means for determining whether or not there is an abnormality in the downstream sensor.

  In short, when comparing the upstream side and the downstream side of the filter device, the upstream side is before collection by the filter device, so that there are many particulate matter contained in the exhaust gas, and the downstream side is the particulate matter contained in the exhaust gas. It can be said that there are few. Therefore, the upstream sensor issues a combustion request more frequently than the downstream sensor, that is, at a relatively short time interval. The present invention has been made with this point in mind. For example, an abnormality in which particulate matter in the exhaust does not adhere to the upstream sensor has occurred, or, on the other hand, a conductive foreign matter such as iron powder in the downstream sensor. If there is an abnormality that adheres, the order relationship between the combustion request issued for the upstream sensor and the combustion request issued for the downstream sensor will be different from that in the normal state. A sensor abnormality is determined.

  For example, when the combustion request for the downstream sensor is issued prior to the combustion request for the upstream sensor in a configuration in which the particulate matter is combusted simultaneously by the upstream sensor and the downstream sensor, the upstream sensor and the downstream sensor It can be determined that one of the side sensors is abnormal. Alternatively, when “the time interval of the combustion request issued for the upstream sensor> the time interval of the combustion request issued for the downstream sensor”, it can be determined that either the upstream sensor or the downstream sensor is abnormal.

  As described above, abnormality diagnosis can be suitably performed for the particulate matter detection sensors on the upstream side and the downstream side of the filter.

  In general, the filter device has a cell partition made of a porous material, and is configured to be ventilated by a large number of micropores in the cell partition, and the micropores gradually become clogged as the collection of particulate matter proceeds. Come. In other words, considering the relationship with the filter regeneration process, there is no clogging immediately after the filter regeneration process, and no particulate matter leaks to the downstream side of the filter. You might also say that. Focusing on this point, the difference in the amount of particulate matter between the upstream side of the filter and the downstream side of the filter becomes small at the beginning after the filter regeneration process is performed, and the upstream side and downstream side sensors have a demand for combustion. It can be said that it is inconvenient for the configuration that performs abnormality determination based on the order or the like.

  Therefore, in the invention described in claim 2, filter regeneration means for performing filter regeneration processing for removing the particulate matter collected by the filter device by combustion, and a predetermined time has elapsed after execution of the filter regeneration processing. Until it does, it is set as the structure provided with the determination prohibition means which prohibits the abnormality determination based on the said combustion request | requirement by the said abnormality determination means.

  According to the above configuration, until the predetermined time elapses after the filter regeneration process is performed, that is, in a period in which the difference in the amount of particulate matter between the filter upstream side and the filter downstream side is small, the abnormality determination unit Abnormality judgment is prohibited. Thereby, the accuracy of abnormality determination can be maintained.

  In the invention according to claim 3, the first combustion is performed either when a combustion request is issued by the first combustion requesting means or when a combustion request is issued by the second combustion requesting means. The particulate matter is combusted simultaneously with respect to the upstream sensor and the downstream sensor by the means and the second combustion means. In addition, the abnormality determination unit is configured to output a combustion request from the first combustion request unit before the combustion request from the second combustion request unit and a combustion request from the first combustion request unit. When the difference between the detection amount of the particulate matter detected by the upstream sensor and the detection amount of the particulate matter detected by the downstream sensor is smaller than a predetermined abnormality determination value, it is determined that there is an abnormality. And a filter regeneration means for performing filter regeneration processing for removing particulate matter collected by the filter device by combustion, and a period other than a predetermined time after the filter regeneration processing is performed. And a determination value changing means for reducing the abnormality determination value.

  According to the above configuration, until a predetermined time elapses after the filter regeneration process is performed, that is, in a period in which the difference in the amount of particulate matter between the filter upstream side and the filter downstream side is small, other periods The abnormality determination value becomes smaller than. As a result, it is possible to avoid erroneously determining an abnormality immediately after the execution of the filter regeneration process, and thus to maintain the accuracy of the abnormality determination.

  In the invention according to claim 4, one of the adhesion amount of the particulate matter detected by the upstream sensor and the adhesion amount of the particulate matter detected by the downstream sensor has reached a predetermined amount. In this case, combustion control means for simultaneously burning the particulate matter adhering to the upstream sensor and the downstream sensor by heating of heaters provided in the sensors, and the particulate control by the combustion control means When a substance is burned, the time required for burning particulate matter adhering to the upstream sensor is compared with the time required for burning particulate matter adhering to the downstream sensor. And an abnormality determining means for determining whether there is an abnormality in the upstream sensor and the downstream sensor based on the comparison result.

  As described above, it can be said that the upstream side of the filter device has a large amount of particulate matter contained in the exhaust gas, and the downstream side has a small amount of particulate matter contained in the exhaust gas. Therefore, when the particulate matter adhering to the upstream sensor and the downstream sensor is burned simultaneously, the upstream sensor requires a longer time to burn the particulate matter than the downstream sensor. . The present invention has been made with this point in mind. For example, an abnormality that the particulate matter in the exhaust does not adhere to the upstream sensor has occurred, or a heater failure (disconnection, etc.) occurs in any of the sensors. If this occurs, the relationship between the time required for combustion by the upstream sensor and the time required for combustion by the downstream sensor will be different from the normal time. Therefore, sensor abnormality is determined by detecting this relationship. To do. For example, when “the time required for combustion in the upstream sensor <the time required for combustion in the downstream sensor” is satisfied, it can be determined that either the upstream sensor or the downstream sensor is abnormal. As described above, abnormality diagnosis can be suitably performed for the particulate matter detection sensors on the upstream side and the downstream side of the filter.

  As described above, in the filter device, the difference in the amount of particulate matter between the upstream side of the filter and the downstream side of the filter becomes small until a predetermined time elapses after the filter regeneration process is performed. It can be said that this is inconvenient for the configuration in which the abnormality determination is performed based on the time required for combustion.

  Therefore, in the invention described in claim 5, filter regeneration means for performing filter regeneration processing for removing particulate matter collected by the filter device by combustion, and a predetermined time has elapsed after the execution of the filter regeneration processing. Until it does, it is set as the structure provided with the determination prohibition means which prohibits the abnormality determination based on the said required time by the said abnormality determination means.

  According to the above configuration, until the predetermined time elapses after the filter regeneration process is performed, that is, in a period in which the difference in the amount of particulate matter between the filter upstream side and the filter downstream side is small, the abnormality determination unit Abnormality judgment is prohibited. Thereby, the accuracy of abnormality determination can be maintained.

  In the invention according to claim 6, the abnormality determination means determines that there is an abnormality when a time difference between the time required for the combustion in the upstream sensor and the downstream sensor is smaller than a predetermined abnormality determination value. Judgment. And a filter regeneration means for performing filter regeneration processing for removing particulate matter collected by the filter device by combustion, and a period other than a predetermined time after the filter regeneration processing is performed. And a determination value changing means for reducing the abnormality determination value.

  According to the above configuration, until a predetermined time elapses after the filter regeneration process is performed, that is, in a period in which the difference in the amount of particulate matter between the filter upstream side and the filter downstream side is small, other periods The abnormality determination value becomes smaller than. As a result, it is possible to avoid erroneously determining an abnormality immediately after the execution of the filter regeneration process, and thus to maintain the accuracy of the abnormality determination.

  According to a seventh aspect of the present invention, the wetness determination means for determining that both the upstream sensor and the downstream sensor have been wetted after the internal combustion engine is started, and the wetness of both sensors by the wetness determination means. Is determined, based on the sensor output of the upstream sensor, a first water-removal determination unit that determines the timing at which the upstream-side sensor has been cleared of water, and A second water-removal determination unit for determining a timing at which the downstream-side sensor has been desorbed based on a sensor output of the downstream-side sensor, and the first water-determination determination; Presence / absence of abnormality in the upstream sensor and the downstream sensor based on the water removal timing determined by the means and the water removal timing determined by the second water removal determination means Characterized in that it comprises an abnormality determination means for determining, a.

  In short, it is considered that when the internal combustion engine is started, the particulate matter detection sensor is exposed to water in the exhaust pipe. In this case, when the particulate matter detection sensor is in a wet state, water becomes a conductive medium, so that a sensor output corresponding to the resistance of water is generated. In addition, the sensor output due to the flooding returns to the sensor output before the flooding with the disappearance of the water in the exhaust pipe. Here, when comparing the sensor output of each sensor in the flooded state, since the high-temperature exhaust flows from the internal combustion engine, originally the sensor output of the upstream sensor first returns to the sensor output before the flooding, After a while, the sensor output of the downstream sensor returns to the sensor output before being flooded. The present invention has been made with this point in mind. For example, if an abnormality occurs in which any conductive foreign matter such as iron powder adheres to any sensor, the sensor output before water exposure after water exposure is eliminated. Since it cannot return, sensor abnormality is judged based on it. As described above, abnormality diagnosis can be suitably performed for the particulate matter detection sensors on the upstream side and the downstream side of the filter.

  According to an eighth aspect of the present invention, the filter device is low-captured by the collection rate determining means for determining that the collection rate of the particulate matter by the filter device is low and the collection rate determining means. An abnormality that determines whether there is an abnormality in the upstream sensor and the downstream sensor based on the output change of the upstream sensor and the output change of the downstream sensor when it is determined that the state is a collection state Determining means.

  In short, the filter device may have a reduced particulate matter collection rate. For example, since there is no clogging of micropores immediately after the filter regeneration process, the particulate matter leaks to the downstream side of the filter. It is thought that the collection rate is reduced. The present invention has been made with this point in mind, and as described above, the filter device is in a state of low collection rate, and it is supposed that output changes occur in both the upstream sensor and the downstream sensor. If it does not, sensor abnormality is determined. As described above, abnormality diagnosis can be suitably performed for the particulate matter detection sensors on the upstream side and the downstream side of the filter.

The block diagram which shows the outline | summary of the engine control system in embodiment of invention. The disassembled perspective view which decomposes | disassembles and shows the principal part structure of the sensor element of PM sensor. The electrical block diagram regarding PM sensor. The flowchart which shows PM sensor abnormality diagnosis processing. The time chart which shows transition of PM collection amount of PM filter, and transition of sensor output of PM sensor. The flowchart which shows PM sensor abnormality diagnosis processing in 2nd Embodiment. The flowchart which shows PM sensor abnormality diagnosis processing in 2nd Embodiment. The flowchart which shows the process of abnormality diagnosis (the 1). The flowchart which shows the process of abnormality diagnosis (the 2). The flowchart which shows PM sensor abnormality diagnosis processing in 3rd Embodiment. The time chart for demonstrating abnormality diagnosis processing more concretely. The flowchart which shows PM sensor abnormality diagnosis processing in 4th Embodiment. The flowchart which shows PM sensor abnormality diagnosis processing in 5th Embodiment. The time chart for demonstrating abnormality diagnosis processing more concretely.

(First embodiment)
Hereinafter, a first embodiment of the present invention will be described with reference to the drawings. This embodiment monitors the amount of PM (the amount of conductive particulate matter) in the exhaust discharged from the engine in a vehicle engine system including an on-vehicle engine. In particular, PM sensors are provided on the upstream side and the downstream side of the PM filter as a filter device in the engine exhaust pipe, respectively, and the amount of PM trapped by the PM filter is monitored based on the amount of PM attached by the PM sensor. FIG. 1 is a configuration diagram showing a schematic configuration of the present system.

  In FIG. 1, an engine 11 is a direct-injection gasoline engine, and the engine 11 is provided with a fuel injection valve 12 and an ignition device 13 as actuators related to the operation of the engine 11. The exhaust pipe 14 of the engine 11 is provided with a three-way catalyst 15 and a PM filter 16 as an exhaust purification device. The PM filter 16 is made of, for example, a porous ceramic material and has a large number of cell partition walls extending in the exhaust pipe longitudinal direction. A large number of fine holes are formed in the cell partition wall of the PM filter 16, and the upstream and downstream sides of the filter can be ventilated by the large number of fine holes. When the exhaust gas passes through the PM filter 16, PM in the exhaust gas is collected by the cell partition wall.

  An A / F sensor 17 is provided on the upstream side of the three-way catalyst 15. Further, PM sensors 18 and 19 as particulate matter detection sensors are provided on the upstream side and the downstream side of the PM filter 16, respectively. Hereinafter, the PM sensor 18 on the upstream side of the filter is also referred to as the upstream PM sensor 18, and the PM sensor 19 on the downstream side of the filter is also referred to as the downstream PM sensor 19. In addition, in this system, a rotation sensor 21 for detecting the engine rotation speed, a pressure sensor 22 for detecting the intake pipe pressure, and the like are provided.

  The ECU 20 is mainly composed of a microcomputer comprising a known CPU, ROM, RAM, etc., and executes various control programs stored in the ROM, so that the engine can be operated according to the engine operating state at each time. 11 and various peripheral devices are controlled. That is, the ECU 20 inputs signals from the various sensors and the like, calculates the fuel injection amount and ignition timing based on the various signals, and controls the driving of the fuel injection valve 12 and the ignition device 13.

  Further, the ECU 20 calculates the PM collection amount of the PM filter 16 based on the detection signals of the PM sensors 18 and 19, and performs the regeneration process (filter regeneration process) of the PM filter 16 based on the PM collection amount. . Specifically, the amount of PM collected per hour is calculated from the difference between the sensor output of the upstream PM sensor 18 and the sensor output of the downstream PM sensor 19, and the amount of PM collected per hour is sequentially integrated. By doing so, the total amount of PM collected by the PM filter 16 is calculated. When the total PM collection amount reaches a predetermined value (start threshold value for filter regeneration processing), filter regeneration processing is performed. The filter regeneration process may be performed by a well-known method. In brief, the unburned fuel is supplied into the exhaust pipe 14 and burned in the exhaust pipe 14 to increase the temperature in the exhaust pipe. PM is burned off. Alternatively, the PM filter 16 is integrated with a heater, and PM is burned and removed by heating the heater. In addition to the above, the filter regeneration process may be performed each time the vehicle travel distance or elapsed time from the previous regeneration execution reaches a predetermined value.

  Further, the ECU 20 carries out PM forced combustion for removing PM adhering to the PM sensors 18 and 19 by combustion, or performs an abnormality diagnosis of each of the sensors 18 and 19 based on detection signals of the PM sensors 18 and 19. Or However, details thereof will be described later.

  Next, the configuration of the PM sensors 18 and 19 and the electrical configuration related to the PM sensors 18 and 19 will be described with reference to FIGS. FIG. 2 is an exploded perspective view showing an essential configuration of the sensor element 31 constituting the PM sensors 18 and 19, and FIG. 3 is an electrical configuration diagram regarding the PM sensors 18 and 19. The PM sensors 18 and 19 have the same configuration, and the sensor element 31 has the same configuration.

  As shown in FIG. 2, the sensor element 31 has two insulating substrates 32 and 33 each having a long plate shape, and one insulating substrate 32 has a PM detector 34 for detecting the amount of PM. The other insulating substrate 33 is provided with a heater portion 35 for heating the sensor element 31. The sensor element 31 is configured by laminating insulating substrates 32 and 33 in two layers. The insulating substrate 32 corresponds to the adherend for attaching PM.

  The insulating substrate 32 is provided with a pair of detection electrodes 36a and 36b which are provided apart from each other on the surface of the substrate opposite to the other insulating substrate 33. PM detection is performed by the pair of detection electrodes 36a and 36b. Part 34 is configured. The detection electrodes 36a and 36b each have a comb shape having a plurality of comb teeth. The detection electrodes 36a and 36b are opposed to each other with a predetermined interval so that the comb teeth of the detection electrodes 36a and 36b are staggered. Moreover, the heater part 35 is comprised by the heat generating body which consists of heating wires, for example.

  However, the shape of the pair of detection electrodes 36a and 36b is not limited to the above, and a pair of detection electrodes 36a and 36b are provided in a curved shape, or a pair of electrode portions each composed of one line are parallel to each other with a predetermined distance therebetween. It may be arranged oppositely.

  Although not shown, the PM sensors 18 and 19 have a holding part for holding the sensor element 31, and the sensor element 31 is fixed to the exhaust pipe in a state where one end side thereof is held by the holding part. It has come to be. In this case, at least a part including the PM detection unit 34 and the heater unit 35 is disposed in the exhaust pipe, and the insulating substrate 32 (PM attached portion) in the sensor element 31 faces the exhaust upstream side. The PM sensors 18 and 19 are configured to be attached to the exhaust pipe. Thus, when exhaust gas containing PM flows through the exhaust pipe, the PM adheres to and accumulates on the detection electrodes 36a and 36b and the periphery thereof on the insulating substrate 32. Further, the PM sensors 18 and 19 have a protective cover that covers the protruding portion of the sensor element 31.

  When the PM in the exhaust gas adheres to and accumulates on the insulating substrate 32 of the sensor element 31, the PM sensors 18 and 19 having the above-described configuration thereby cause a resistance value of the PM detection unit 34 (that is, a resistance value between the pair of detection electrodes 36a and 36b). ) Changes and the change in the resistance value corresponds to the PM deposition amount, the PM amount is detected using the change in the resistance value.

  As shown in FIG. 3, a sensor power supply 41 is connected to one end of each PM detector 34 (34A, 34B in the figure) of the PM sensor 18, 19 as an electrical configuration related to the PM sensor 18, 19, and the other end. A shunt resistor 42 is connected to the side. The sensor power supply 41 is constituted by a constant voltage circuit, for example, and the constant voltage Vcc is 5V. In this case, a voltage dividing circuit 40 is formed by the PM detection unit 34 and the shunt resistor 42, and an intermediate voltage between them is input to the ECU 20 as a PM detection voltage Vpm (sensor detection value). Note that the PM detection voltage of the upstream PM sensor 18 is Vpm1, and the PM detection voltage of the downstream PM sensor 19 is Vpm2. That is, in each PM detection part 34 of the PM sensors 18 and 19, the resistance value Rpm changes according to the PM accumulation amount, and the PM detection voltage Vpm changes depending on the resistance value Rpm and the resistance value Rs of the shunt resistor 42. Then, the PM detection voltage Vpm is input to the microcomputer 44 via an A / D converter (not shown).

Here, when Vcc = 5 V and Rs = 100 kΩ, the PM detection voltage Vpm is obtained by the following equation (1).
Vpm = 5V × 100 kΩ / (100 kΩ + Rpm) (1)
At this time, if the PM deposition amount is 0 (or substantially 0), the resistance value Rpm of the PM detection unit 34 becomes infinite, so Vpm = 0V. Further, when the resistance value Rpm of the PM detection unit 34 is reduced to, for example, 1 kΩ due to PM deposition, Vpm = 4.95V. Thus, the PM detection voltage Vpm changes according to the amount of PM accumulated in the PM detection unit 34. The microcomputer 44 calculates the PM accumulation amount according to the PM detection voltage Vpm.

  The voltage dividing circuit 40 constitutes a signal output circuit, and the voltage dividing circuit 40 can change the PM detection voltage Vpm with an output range of 0 to 5V. In this case, the output upper limit value of the PM detection voltage Vpm is 5V, more strictly, a voltage value slightly lower than 5V.

  Further, heater power sources 45 are connected to the heater portions 35 (35A and 35B in the figure) of the PM sensors 18 and 19, respectively. The heater power supply 45 is, for example, an in-vehicle battery, and each heater unit 35 is heated by power supply from the in-vehicle battery. In this case, a transistor 46 (46A, 46B in the figure) as a switching element is connected to the low side of the heater unit 35, and each transistor 46 is turned on / off by the microcomputer 44 to control the heating of the heater unit 35. Done.

  When energization of the heater unit 35 is started in a state where PM is deposited on the insulating substrate 32, the temperature of the deposited PM rises, and the deposited PM is forcibly burned accordingly. Due to such forced combustion, PM deposited on the insulating substrate 32 is removed by combustion. For example, the microcomputer 44 performs heating control by the heater unit 35 when a request for forced combustion of PM is generated when the engine is started or when the operation is finished, or when it is determined that the PM accumulation amount has reached a predetermined amount. More specifically, the microcomputer 44 detects when the PM accumulation amount detected by the upstream PM sensor 18 reaches the execution reference value for forcibly burning the accumulated PM, or detected by the downstream PM sensor 19. When the PM accumulation amount reaches the execution standard value for forcibly burning the accumulated PM, a combustion request is issued and the heater portions 35a and 35b of the PM sensors 18 and 19 are energized to perform the PM combustion processing ( The implementation standard values are the same). Note that the PM forced combustion process of the PM sensors 18 and 19 regenerates the PM accumulation amount detection function in the PM sensors 18 and 19, and is also referred to as a sensor regeneration process in that sense.

  In addition, the ECU 20 is provided with an EEPROM 47 as a backup memory for storing various learning values, abnormality diagnosis values (diagnostic data), and the like.

  Next, abnormality diagnosis processing for the PM sensors 18 and 19 will be described. In the abnormality diagnosis of the present embodiment, when the upstream side and the downstream side of the PM filter 16 are compared, the upstream side is before collecting the filter, so that the amount of PM in the exhaust gas is large, and the downstream side has a small amount of PM in the exhaust gas. Therefore, the upstream side PM sensor 18 is made by paying attention to the fact that the combustion request is issued more frequently than the downstream side PM sensor 19, that is, at a relatively short time interval.

  In particular, after the regeneration process for the upstream PM sensor 18 is performed, the next timing when the combustion request (regeneration process execution request) is issued, and the next combustion after the regeneration process for the downstream PM sensor 19 is performed. Based on the timing at which a request (reproduction processing execution request) is issued, abnormality diagnosis of the PM sensors 18 and 19 is performed. Further, the time required for burning the attached PM in the upstream PM sensor 18 and the time required for burning the attached PM in the downstream PM sensor 19 are compared, and the abnormality of the PM sensors 18 and 19 is determined based on the comparison result. Make a diagnosis.

  FIG. 4 is a flowchart showing the PM sensor abnormality diagnosis process. This process is repeatedly executed by the microcomputer 44 in the ECU 20 at a predetermined cycle.

  In FIG. 4, in step S11, it is determined whether or not the sensor regeneration process (forced combustion process) is currently being performed for both PM sensors 18 and 19. If the sensor regeneration process is not being performed, the process proceeds to step S12. If the sensor regeneration process is being performed, the process proceeds to step S16.

  First, assuming that the sensor regeneration process is not being performed, in step S12, it is determined whether or not there is a request for performing the sensor regeneration process for both PM sensors 18 and 19. Specifically, for the PM sensors 18 and 19, it is determined whether at least one of the PM detection voltages Vpm (PM adhesion amount) has reached a predetermined execution reference value serving as an execution determination reference for the sensor regeneration process. To do. In this case, if it is YES, it will progress to subsequent step S13, and if it is NO, this processing will be once ended as it is.

  In step S13, regarding the execution request of the current sensor regeneration process, the sensor regeneration process is determined based on which PM detection voltage Vpm of the upstream and downstream PM sensors 18 and 19 is requested, thereby detecting the sensor. Carry out abnormality diagnosis. In other words, whether the PM detection voltage Vpm1 of the upstream PM sensor 18 first reached the execution reference value for the sensor regeneration process, or the PM detection voltage Vpm2 of the downstream PM sensor 19 first reached the execution reference value for the sensor regeneration process. Determine whether it was done. At this time, the time interval from when the regeneration process is performed on the upstream side PM sensor 18 until the next execution request for the regeneration process is issued is the next regeneration process after the regeneration process is performed on the downstream side PM sensor 19. If it is shorter than the time interval until the execution request is issued, the PM detection voltage Vpm1 of the upstream PM sensor 18 first reaches the execution reference value of the sensor regeneration process.

  Then, if the PM detection voltage Vpm1 of the upstream PM sensor 18 has previously reached the execution reference value for the sensor regeneration process, the process proceeds to step S14, and the regeneration process is performed for both PM sensors 18, 19. At this time, heating by the heater unit 35 is started for the PM sensors 18 and 19. That is, sensor regeneration processing is performed simultaneously for both sensors.

  Further, if the PM detection voltage Vpm2 of the downstream PM sensor 19 has reached the execution reference value for the sensor regeneration process first, the process proceeds to step S15. In step S15, assuming that an abnormality has occurred in either of the PM sensors 18 and 19, information on the occurrence of the sensor abnormality is stored in the EEPROM 47 as diagnostic data.

  For example, if the upstream PM sensor 18 has an abnormality in which PM in the exhaust does not adhere, or the downstream PM sensor 19 has an abnormality in which conductive foreign matter such as iron powder adheres, the upstream PM sensor The order of the regeneration process execution request for 18 and the regeneration process execution request for the downstream PM sensor 19 is reversed from that in the normal state, and the regeneration process execution request for the upstream PM sensor 18 is later downstream. The reproduction process execution request for the PM sensor 19 is first. Thereby, it is determined that one of the PM sensors 18 and 19 is abnormal.

  On the other hand, when the process proceeds to step S16 after the start of the sensor regeneration process, it is determined in step S16 whether or not the sensor regeneration process has been completed for each of the PM sensors 18 and 19. If sensor regeneration processing is not completed, regeneration required times TA1 and TA2 are measured in step S17. The regeneration required times TA1 and TA2 are measured for each of the PM sensors 18 and 19, and are measured based on changes in the PM detection voltages Vpm1 and Vpm2 of the PM sensors 18 and 19.

  More specifically, regarding the measurement of the regeneration required times TA1 and TA2, in the sensor regeneration process, when the attached PM is burned and removed by heating the heater, the PM detection voltages Vpm1 and Vpm2 change and decrease to a predetermined value (this embodiment). Will reach 0V). In such a case, the time from when the heater energization starts until the PM detection voltages Vpm1 and Vpm2 reach predetermined values is measured as the regeneration required times TA1 and TA2.

  Moreover, if the sensor regeneration process of each PM sensor 18 and 19 is completed, it will progress to step S18. In step S18, sensor abnormality diagnosis is performed by determining whether TA1 <TA2 based on the regeneration required times TA1 and TA2 of the PM sensors 18 and 19 measured during the sensor regeneration process. If TA1 <TA2, that is, if the regeneration required time TA1 of the upstream PM sensor 18 is shorter than the regeneration required time TA2 of the downstream PM sensor 19, the process proceeds to step S15. In step S15, sensor abnormality occurrence information is stored in the EEPROM 47 as diagnostic data as described above.

  For example, if there is an abnormality that PM in the exhaust does not adhere to the upstream PM sensor 18 or a heater failure (disconnection, etc.) occurs in any of the PM sensors 18 and 19, each regeneration required time TA1, The magnitude relationship of TA2 is opposite to that in the normal state, and “reproduction required time TA1 <required reproduction time TA2” is satisfied. Therefore, it can be determined that one of the PM sensors 18 and 19 is abnormal.

  According to the embodiment described in detail above, the following excellent effects can be obtained.

  After the sensor regeneration process (PM forced combustion) is performed on both PM sensors 18 and 19 at the same time, the next regeneration process execution request (combustion request) of the upstream PM sensor 18 and the downstream PM sensor 19 An abnormality diagnosis of the PM sensors 18 and 19 is performed based on the order of the regeneration processing execution request (combustion request). More specifically, after performing the sensor regeneration process on both PM sensors 18 and 19 at the same time, whether the PM detection voltage Vpm1 of the upstream PM sensor 18 first reached the execution reference value of the sensor regeneration process, or whether the downstream PM It is determined whether the PM detection voltage Vpm2 of the sensor 19 has first reached the reference value for performing the sensor regeneration process, and abnormality diagnosis of the PM sensors 18 and 19 is performed based on the determination result.

  Also, the regeneration required times TA1 and TA2 of the PM sensors 18 and 19 are compared, and abnormality diagnosis of the PM sensors 18 and 19 is performed based on the comparison result.

  As described above, abnormality diagnosis can be suitably performed for the PM sensors 18 and 19 on the filter upstream side and the filter downstream side.

  Since the abnormality diagnosis of the PM sensors 18 and 19 can be suitably performed as described above, the reliability of the PM collection amount of the PM filter 16 calculated based on the detection signals of the PM sensors 18 and 19 is improved. Thereby, it is possible to optimize the filter regeneration process of the PM filter 16.

(Second Embodiment)
Next, the second embodiment will be described focusing on the differences from the first embodiment. In the PM filter 16, a filter regeneration process is performed in which the collected PM is removed by combustion periodically or irregularly. Further, when the filter regeneration process is performed, looking at the process in which PM is newly collected by the PM filter 16, the PM is not attached to the filter micropores immediately after the filter regeneration process. There is no clogging, and it is considered that PM leaks to the downstream side of the filter. That is, it is considered that the amount of PM on the downstream side of the filter differs depending on the degree of PM collection by the PM filter 16. Further, when the regeneration process of the PM filter 16 and the regeneration processes of the PM sensors 18 and 19 are compared, the period of execution of the regeneration process is greatly different, the period of execution of the filter regeneration process is relatively long, and the period of execution of the sensor regeneration process Is relatively short. For example, the filter regeneration process is performed every time the vehicle travel distance becomes several hundred km to several thousand km, for example, whereas the sensor regeneration process is performed every time the vehicle travel distance becomes several tens km, for example (one time). Will be performed several times during the vehicle driving).

  Referring to the time chart of FIG. 5, the timings t1 and t2 are the execution timing of the filter regeneration process of the PM filter 16, and the execution period of the filter regeneration process is Tf. In addition to the filter regeneration process, the sensor regeneration process is performed in small cycles with a short period.

  Here, the sensor outputs of the PM sensors 18 and 19 immediately after the execution of the filter regeneration process (for example, TX1 in FIG. 5) and immediately after the execution of the next filter regeneration process (for example, TX2 in FIG. 5). In the upstream PM sensor 18, if the PM concentration in the exhaust gas is constant, the sensor output changes are the same, that is, the slope of the output change is the same. On the other hand, for the downstream PM sensor 19, even if the PM concentration in the exhaust gas is constant, the change in sensor output (due to the PM collection amount (the degree of PM collection)) in the PM filter 16 ( The slope of the output change is different between TX1 and TX2. Specifically, the slope of the output change of the downstream PM sensor 19 is relatively large immediately after the execution of the filter regeneration process (TX1), and immediately before the execution of the next filter regeneration process (TX2), The slope of the output change is relatively small. That is, at the beginning after the filter regeneration process is performed, the difference in the PM amount in the exhaust between the upstream side and the downstream side of the filter becomes small.

  In the following, in consideration of the fact that the relationship between the PM adhesion amounts of the PM sensors 18 and 19 is related to the degree of PM collection of the PM filter 16 as described above, a predetermined time after the filter regeneration process is performed. A configuration for prohibiting abnormality determination until the time has elapsed and a configuration for changing the determination value for abnormality determination until the predetermined time has elapsed will be described.

<Configuration that prohibits abnormality determination>
FIG. 6 is a flowchart showing a part of the PM sensor abnormality diagnosis process shown in FIG. 5 in a modified manner. Hereinafter, the difference from FIG. In addition, about the process common to FIG. 5, the same step number is attached | subjected and description is abbreviate | omitted.

  In FIG. 6, first, in step S21, it is determined whether or not the elapsed time since the filter regeneration process of the PM filter 16 has reached a predetermined time. The predetermined time is, for example, a time of 1/10, 1/20, or the like of the execution period of the filter regeneration process, and is predetermined. In this case, if step S21 is YES, the process after subsequent step S11 is implemented, and if step S21 is NO, this process is ended as it is. That is, the abnormality diagnosis is prohibited before the predetermined time elapses after the filter regeneration process is performed.

<Configuration for changing the judgment value for abnormality judgment>
FIG. 7 is a flowchart showing a part of the PM sensor abnormality diagnosis process shown in FIG. 5 in a modified manner. Hereinafter, the difference from FIG. In addition, about the process common to FIG. 5, the same step number is attached | subjected and description is abbreviate | omitted.

  In FIG. 7, when it is determined in step S13 that the PM detection voltage Vpm1 of the upstream PM sensor 18 has previously reached the reference value for performing the sensor regeneration process, the process proceeds to step S30, where “abnormal diagnosis ( Part 1) ”is carried out.

  In “abnormality diagnosis (part 1)” in FIG. 8, in step S31, it is determined whether or not the elapsed time since the filter regeneration process of the PM filter 16 has reached a predetermined time (step in FIG. 6). Same as S21). And if step S31 is YES, it will progress to step S32, and if step S31 is NO, it will progress to step S33. In step S32, KE1 is set as the abnormality determination value. In step S33, KE2 is set as the abnormality determination value. Here, KE1 and KE2 are threshold values for carrying out an abnormality determination based on the difference between the PM adhesion amounts of the PM sensors 18 and 19, and KE1> KE2.

  Thereafter, in step S34, the difference between the current PM adhesion amount of the upstream PM sensor 18 and the PM adhesion amount of the downstream PM sensor 19 is calculated. In the subsequent step S35, the difference in the PM adhesion amount is determined as an abnormality determination. It is determined whether it is smaller than the value. The abnormality determination value is the set value (KE1 or KE2) in step S32 or S33. If the difference from the PM adhesion amount is smaller than the abnormality determination value, it is determined in step S36 that the PM sensor is abnormal. If it is determined in step S36 that the PM sensor is abnormal, the sensor regeneration process in step S14 in FIG. 7 is not performed.

  Further, in FIG. 7, when it is determined in step S16 that the regeneration process has been completed for each PM sensor 18, 19, the process proceeds to step S40 and "abnormality diagnosis (part 2)" shown in FIG. 9 is performed.

  In “abnormality diagnosis (part 2)” of FIG. 9, in step S41, it is determined whether or not the elapsed time since the filter regeneration processing of the PM filter 16 has reached a predetermined time (step of FIG. 6). Same as S21). And if step S41 is YES, it will progress to step S42, and if step S41 is NO, it will progress to step S43. In step S42, KE3 is set as the abnormality determination value. In step S43, KE4 is set as the abnormality determination value. Here, KE3 and KE4 are threshold values for performing an abnormality determination based on the difference between the regeneration required times TA1 and TA2 of the PM sensors 18 and 19, and KE3> KE4.

  Thereafter, in step S44, the difference between the required regeneration times TA1 and TA2 (measured values in step S17) of the PM sensors 18 and 19 is calculated, and in the subsequent step S45, the difference between the required regeneration times TA1 and TA2 is determined as an abnormality determination value. Or less. The abnormality determination value is the set value (KE3 or KE4) in step S42 or S43. If the difference between the reproduction required times TA1 and TA2 is smaller than the abnormality determination value, it is determined in step S46 that the PM sensor is abnormal.

  As described above, according to the second embodiment, the abnormality determination is prohibited until a predetermined time elapses after the filter regeneration process is performed (step S21 in FIG. 6). As a result, the abnormality determination is prohibited during the period in which the difference in PM amount between the filter upstream side and the filter downstream side is small after the filter regeneration process is performed, and the accuracy of the abnormality determination can be maintained.

  Further, after the sensor regeneration process is simultaneously performed for both PM sensors 18 and 19, the PM detection voltage Vpm1 of the upstream PM sensor 18 first reaches the execution reference value for the sensor regeneration process (combustion by the first combustion requesting means). In the case where the request is issued prior to the combustion request by the second combustion requesting means), if the difference between the PM adhesion amounts of the PM sensors 18, 19 is smaller than a predetermined abnormality determination value, it is determined that there is an abnormality. Further, the abnormality determination value is set to be smaller than the other periods until a predetermined time elapses after the filter regeneration process is performed (abnormality diagnosis (part 1 in FIG. 8)). As a result, it is possible to avoid erroneously determining an abnormality immediately after the execution of the filter regeneration process, and thus to maintain the accuracy of the abnormality determination.

  Further, when the time difference between the regeneration required times TA1 and TA2 of the PM sensors 18 and 19 is smaller than a predetermined abnormality determination value, it is determined that there is an abnormality, and further, a predetermined time has elapsed after the filter regeneration process is performed. Until this is done, the abnormality determination value is set to be smaller than the other periods (abnormality diagnosis (part 2) in FIG. 9). As a result, it is possible to avoid erroneously determining an abnormality immediately after the execution of the filter regeneration process, and to maintain the accuracy of the abnormality determination.

(Third embodiment)
Next, the third embodiment will be described focusing on the differences from the first embodiment. In the present embodiment, using the fact that the PM sensors 18 and 19 are flooded by the water present in the exhaust pipe after the engine is started, abnormality diagnosis of the PM sensors 18 and 19 is performed based on the change in the sensor output accompanying the cancellation of the moisture. Is to implement.

  FIG. 10 is a flowchart showing the PM sensor abnormality diagnosis process in the present embodiment. This process is repeatedly executed by the microcomputer 44 in the ECU 20 at a predetermined cycle. In the present embodiment, the sensor regeneration process (PM forced combustion process) is performed in the PM sensors 18 and 19 when the operation of the engine 11 is stopped, and the PM in each PM sensor 18 and 19 is started when the engine is started. The amount of adhesion is zero or very small.

  In FIG. 10, in step S51, it is determined by the abnormality diagnosis execution flag whether or not the sensor abnormality diagnosis due to the moisture in the exhaust pipe is not performed after the engine is started. At this time, if the abnormality diagnosis execution flag = 0, it is determined that the sensor abnormality diagnosis has not been performed. And when abnormality diagnosis is not performed, it progresses to subsequent step S52. In step S52, it is determined based on the state of the water generation flag whether or not each of the PM sensors 18 and 19 has been subjected to water determination after the engine is started. At this time, if the water exposure flag = 0, it is determined that the sensor water has not been determined yet, and the process proceeds to step S53.

  In step S53, based on the PM detection voltages Vpm1 and Vpm2 (sensor outputs) of the PM sensors 18 and 19, it is determined whether or not water is generated in the PM sensors 18 and 19. At this time, if the PM detection voltages Vpm1 and Vpm2 of the PM sensors 18 and 19 correspond to the resistance values of water, respectively, it is determined that water has been generated. In the case of water generation, 1 is set to the water generation flag in the subsequent step S54.

  If NO in step S53, it is determined in step S55 whether or not a predetermined time (moisture monitoring time) has elapsed after the engine is started. In such a case, when the predetermined time has passed without the occurrence of water exposure after the engine is started, step S55 becomes YES and 1 is set to the abnormality diagnosis execution flag. That is, the abnormality diagnosis of the present embodiment is performed on the assumption that the wetness is generated by the PM sensors 18 and 19 at the time of starting the engine. It is set as “completed” without performing abnormality diagnosis.

  After 1 is set to the water generation flag, step S52 becomes NO, and the process proceeds to step S57. In step S57, it is determined whether or not water is eliminated by both PM sensors 18 and 19. That is, when the PM sensors 18 and 19 are removed from the flooded state, the resistance between the electrodes in each PM sensor 18 and 19 increases, so that the PM detection voltage changes to the decrease side. Therefore, if a decrease in the PM detection voltage is confirmed in both PM sensors 18 and 19, it is determined that water exposure has been eliminated in both PM sensors 18 and 19. Determining whether water is eliminated means that the abnormality diagnosis has been completed without determining that there is a sensor abnormality, and then the process proceeds to step S62, where 1 is set in the abnormality diagnosis execution flag.

  When step S57 is NO, it progresses to step S58 and it is determined whether one of PM sensors 18 and 19 has eliminated water exposure. This is determined based on the decreasing change in the PM detection voltage Vpm (sensor output), as in step S57.

  If step S58 is YES, the process proceeds to step S59, and the time difference counter is incremented by one. The time difference counter measures the time difference of the water removal timing at the PM sensors 18 and 19.

  Thereafter, in step S60, it is determined whether or not the value of the time difference counter is equal to or greater than a predetermined abnormality determination value. If the counter value <the abnormality determination value, the process is temporarily terminated as it is. If counter value ≧ abnormal determination value, the process proceeds to step S61, and sensor abnormality occurrence information is stored in the EEPROM 47 as diagnostic data, assuming that a sensor abnormality has occurred. Thereafter, in step S62, 1 is set to the abnormality diagnosis execution flag.

  In step S61, it may be configured to identify which of the PM sensors 18 and 19 is abnormal. That is, in step S58, it is determined that one of the PM sensors 18 and 19 has eliminated water exposure, and if the other PM sensor does not determine that water exposure has been eliminated, it is determined that there is an abnormality. Those who are not judged to have eliminated water are judged as abnormal sensors. For example, if it is determined in step S58 that the upstream PM sensor 18 has eliminated water exposure, and then the downstream PM sensor 19 does not determine that water exposure has been eliminated, it is determined that the downstream PM sensor 19 is abnormal. .

  In addition, when both step S57 and S58 are continued with NO and predetermined time passes in that state, the structure which determines with abnormality about both PM sensors 18 and 19 at that time may be sufficient.

  Further, since high-temperature exhaust gas flows from the engine 11 into the exhaust pipe 14, it is considered that the upstream PM sensor 18 has a faster timing of water removal than the downstream PM sensor 19. Therefore, the water removal timing of the upstream PM sensor 18 and the water removal timing of the downstream PM sensor 19 are compared, and when the former is later than the latter, it can be determined that there is a sensor abnormality. It is.

  FIG. 11 is a time chart for more specifically explaining the abnormality diagnosis process in the present embodiment.

  In FIG. 11, since water is present in the exhaust pipe 14 after the engine is started, water adheres to each PM sensor 18, 19 (soaked), and the sensor output is caused by a change in interelectrode resistance due to water adhesion. Both (PM detection voltage Vpm) increase. The sensor output at this time is an output range corresponding to the resistance of water.

  Then, at timing t11, the sensor output of the sensor 18 starts to change to the decreasing side as the upstream PM sensor 18 is dehydrated. Along with this, the increment of the time difference counter is started.

  Thereafter, when the downstream PM sensor 19 is also covered with water at timing t12, the sensor output of the sensor 19 starts to change to the decreasing side. At this time, if the value of the time difference counter is less than the abnormality determination value, the sensor abnormality is not diagnosed. On the other hand, for example, if an abnormality occurs in which conductive foreign matters such as iron powder adhere to the downstream PM sensor 19, it is not possible to return to the sensor output before being wet after the water is removed. For this reason, if the value of the time difference counter becomes equal to or greater than the abnormality determination value without returning the sensor output, it is diagnosed that there is a sensor abnormality (timing t13).

  As described above, according to the third embodiment, after it is determined that the PM sensors 18 and 19 are both wet after the engine is started, the timings of water removal of the PM sensors 18 and 19 are compared, and based on the comparison result. Thus, an abnormality diagnosis of the PM sensors 18 and 19 is performed. Thereby, abnormality diagnosis can be suitably performed for the PM sensors 18 and 19 on the filter upstream side and the filter downstream side.

(Fourth embodiment)
Next, the fourth embodiment will be described focusing on the differences from the first embodiment. In this embodiment, it is determined that the PM collection rate by the PM filter 16 is low, and the output change of the upstream PM sensor 18 and the output change of the downstream PM sensor 19 in the low collection rate state. Based on the above, abnormality diagnosis of the PM sensors 18 and 19 is performed.

  FIG. 12 is a flowchart showing the PM sensor abnormality diagnosis process in the present embodiment. This process is repeatedly executed by the microcomputer 44 in the ECU 20 at a predetermined cycle.

  In FIG. 12, first, in steps S71 to S73, it is determined whether or not the PM collection rate by the PM filter 16 is currently low. Specifically, in step S71, it is determined by a filter failure determination process (not shown) that the PM filter 16 is in a failure state, and it is determined whether or not the PM failure is a failure with a low PM collection rate. In step S72, it is determined whether or not it is immediately after the execution of the filter regeneration process, that is, whether or not a predetermined time has elapsed after the execution of the filter regeneration process. In step S73, it is determined whether or not the bypass valve provided in the bypass valve connecting the upstream side and the downstream side of the PM filter 16 is in an open state.

  And if any of step S71-S73 is YES, it will progress to subsequent step S74, and if all are NO, this process will be once complete | finished as it is.

  When the process proceeds to step S74, in step S74, it is determined whether or not an output change has occurred in only one of the upstream PM sensor 18 and the downstream PM sensor 19. In this case, if both PM sensors 18 and 19 have output changes, the present process is temporarily terminated as it is. If only one output change has occurred, the process proceeds to step S75.

  In step S75, it is determined whether or not the PM sensor in which the output change has occurred is the upstream PM sensor 18. At this time, if the output change occurs only in the upstream PM sensor 18, the process proceeds to step S76, and if the output change occurs only in the downstream PM sensor 19, the process proceeds to step S79.

  In step S76, the first counter C1 is incremented by 1. In subsequent step S77, it is determined whether or not the counter value is equal to or greater than a predetermined value K1. If C1 ≧ K1, it is determined in step S78 that the downstream PM sensor 19 is abnormal. In other words, although output changes should occur in both the upstream and downstream PM sensors 18 and 19, output changes occur in the upstream PM sensor 18, but output changes in the downstream PM sensor 19 occur. If not, it is determined that the sensor is abnormal. In this case, the abnormality occurrence information of the downstream PM sensor 19 is stored in the EEPROM 47 as diagnostic data. By using the first counter C1, it is possible to accurately perform abnormality diagnosis in consideration of the transport delay of exhaust from the upstream PM sensor 18 to the downstream PM sensor 19.

  In step S79, the second counter C2 is incremented by 1. In the subsequent step S80, it is determined whether or not the counter value is equal to or greater than a predetermined value K2. If C2 ≧ K2, it is determined in step S81 that the upstream PM sensor 18 is abnormal. That is, although output changes should occur in both the upstream and downstream PM sensors 18 and 19, output changes have occurred in the downstream PM sensor 19, but output changes in the upstream PM sensor 18 have occurred. If not, it is determined that the sensor is abnormal. In this case, the abnormality occurrence information of the upstream PM sensor 18 is stored in the EEPROM 47 as diagnostic data.

  According to the fourth embodiment described in detail above, when it is determined that the PM collection rate by the PM filter 16 is low, the PM sensors 18 and 19 are based on the output changes of the PM sensors 18 and 19. It was set as the structure which performs abnormality diagnosis of. Thereby, abnormality diagnosis can be suitably performed for the PM sensors 18 and 19 on the filter upstream side and the filter downstream side.

(Fifth embodiment)
Next, the fifth embodiment will be described focusing on the differences from the first embodiment. In this embodiment, it is determined that the amount of PM emission from the engine 11 is small, and in the low PM emission state, based on the output change of the upstream PM sensor 18 and the output change of the downstream PM sensor 19. Thus, abnormality diagnosis of the PM sensors 18 and 19 is performed.

  FIG. 13 is a flowchart showing the PM sensor abnormality diagnosis process in the present embodiment. This process is repeatedly executed by the microcomputer 44 in the ECU 20 at a predetermined cycle.

  In FIG. 13, first, in step S91, it is determined whether or not the PM emission amount from the engine 11 is currently small. Specifically, it is determined that the PM emission amount is low when the engine is idling or when the fuel is being cut. In addition, when the PM emission amount is estimated based on the engine operating state such as the engine rotation speed and the load, and the PM emission amount (estimated value) is equal to or less than a predetermined value, it is determined that the PM emission amount is low. Also good. If step S91 is YES, the process proceeds to step S92, and if NO, the process proceeds to step S98.

  When the process proceeds to step S92, in steps S92 to S94, it is determined whether or not there is an output change for each of the PM sensors 18, 19. Specifically, in step S92, it is determined whether or not an output change has occurred in the upstream PM sensor 18, and an output change has occurred in the downstream PM sensor 19 in each of cases where YES or NO in step S92. It is determined whether or not there are (steps S93, S94).

  At this time, if output changes have occurred in both the upstream PM sensor 18 and the downstream PM sensor 19 (both S92 and S93 are YES), the process proceeds to step S95, and either the engine 11 or the PM sensors 18 and 19 It is determined that an abnormality has occurred. If an output change has occurred only in the upstream PM sensor 18 (YES in S92, NO in S93), the process proceeds to step S96, where it is determined that an abnormality has occurred in the upstream PM sensor 18. If the output change has occurred only in the downstream PM sensor 19 (S92 is NO, S94 is YES), the process proceeds to step S97, where it is determined that an abnormality has occurred in the downstream PM sensor 19.

  On the other hand, if step S91 is NO and the process proceeds to step S98, the output change amount per unit time of the upstream PM sensor 18 (upstream output change amount) and the output per unit time of the downstream PM sensor 19 are output. The change amount (downstream output change amount) is compared, and it is determined whether or not the upstream output change amount ≦ the downstream output change amount. If the upstream output change amount ≦ the downstream output change amount, the process proceeds to step S99, and it is determined that one of the PM sensors 18, 19 is abnormal. In addition, the abnormality occurrence information by each abnormality determination described above is sequentially stored in the EEPROM 47 as diagnostic data.

  According to the fifth embodiment described in detail above, when it is determined that the amount of PM emission from the engine 11 is small, the PM sensor 18, It was set as the structure which implements 19 abnormality diagnosis. In this case, the diagnosis accuracy can be improved by performing the abnormality diagnosis under the low PM emission state. Further, by determining which of the two PM sensors 18 and 19 has an output change, it is possible to identify the PM sensor in which an abnormality has occurred.

  Further, even if the state is not low PM emission, the upstream output change amount ≦ the downstream output change amount cannot be satisfied. Therefore, the sensor abnormality diagnosis can be suitably performed by comparing these output change amounts. .

(Other embodiments)
The present invention is not limited to the description of the above embodiment, and may be implemented as follows, for example.

  In the first embodiment, the sensor regeneration process is simultaneously performed for the upstream PM sensor 18 and the downstream PM sensor 19, but the sensor regeneration process is changed at different timings. It is good. Specifically, when the PM adhesion amount detected by the upstream PM sensor 18 reaches a predetermined amount, regeneration processing (PM forced combustion) of the upstream PM sensor 18 is performed, and the downstream PM sensor 19 When the detected PM adhesion amount reaches a predetermined amount, regeneration processing (PM forced combustion) of the downstream PM sensor 19 is performed.

  In this configuration, the time interval TB1 from when the regeneration process is performed on the upstream PM sensor 18 until the next regeneration process request (combustion request) is issued, and the regeneration process is performed on the downstream PM sensor 19. The time interval TB2 until the next regeneration processing request (combustion request) is issued is calculated, and abnormality diagnosis of the PM sensors 18 and 19 is performed based on the time intervals TB1 and TB2. In this case, if TB1> TB2, it is determined that an abnormality has occurred in one of the PM sensors 18, 19.

  In the configuration in which the time intervals TB1 and TB2 until the regeneration processing request (combustion request) is issued as described above, the abnormality diagnosis cannot be performed until the regeneration processing request for the downstream PM sensor 19 is issued. . In that respect, according to the configuration of the first embodiment (configuration of FIG. 4), abnormality diagnosis can be performed in the execution cycle of the regeneration process of the upstream PM sensor 18. Therefore, it can be said that the configuration of the first embodiment is more desirable in terms of implementation frequency.

  In the first embodiment, the time required for the regeneration at the time of performing the sensor regeneration process is measured by the time from when the heater energization is started until the PM detection voltages Vpm1 and Vpm2 each reach a predetermined value (0 V). However, this is changed as follows. A modification thereof will be described with reference to the time chart of FIG. In FIG. 14, sensor regeneration processing is started at timing t <b> 21, and accordingly, heater energization of each PM sensor 18, 19 is started. At this time, since the PM itself has a temperature characteristic in which the resistance value changes with respect to the temperature, the inter-electrode resistance decreases due to the temperature rise of the PM, and the sensor output increases accordingly. And each sensor output will be in the state stuck on the output upper limit (for example, 5V).

  After that, when PM actually starts to burn, the output of each sensor starts to drop. At this time, the timing at which each sensor output begins to drop (t22, t23 in the figure) corresponds to the time required for PM combustion, and therefore the abnormality of each PM sensor 18, 19 is caused by the sensor output fall timing. Diagnosis can be performed. Specifically, when the PM sensors 18 and 19 are normal, the sensor output drop of the downstream PM sensor 19 is the previous timing (t22), and the sensor output drop of the upstream PM sensor 18 is the later timing (t23). . On the other hand, when one of the PM sensors 18 and 19 is abnormal, the sensor output drop of the upstream PM sensor 18 is the earlier timing (t23), and the sensor output drop of the downstream PM sensor 19 is the later timing (t24). Become. Thus, the presence or absence of abnormality of the PM sensors 18 and 19 can be determined based on the sensor output lowering timing.

  In the third embodiment, the sensor regeneration process (PM forced combustion process) is performed in each PM sensor 18 and 19 when the operation of the engine 11 is stopped, and the PM adhesion amount in each PM sensor 18 and 19 is determined when the engine is started. It is assumed that the amount is zero or a minute amount, but this may be changed. For example, when the engine is started, the abnormality diagnosis execution condition is that the sensor outputs of the PM sensors 18 and 19 are smaller than a value corresponding to the resistance of water. In this case, at the time of starting the engine, it is possible to determine that each PM sensor 18, 19 has been wetted and that the water has been removed, so that an abnormality diagnosis can be performed.

  -In the above-mentioned embodiment, although application about a direct-injection type gasoline engine was illustrated, it is applicable also to other types of engines. For example, the present invention can be applied to a diesel engine (particularly, a direct injection type diesel engine), and the present invention can be used for a PM sensor provided in an exhaust pipe of the diesel engine. Alternatively, the PM amount may be detected for a gas other than the engine exhaust.

  DESCRIPTION OF SYMBOLS 11 ... Engine (internal combustion engine), 14 ... Exhaust pipe, 16 ... PM filter (filter apparatus), 18, 19 ... PM sensor (particulate matter detection sensor), 20 ... ECU, 44 ... Microcomputer (1st combustion means, 1st 2 combustion means, first combustion request means, second combustion request means, abnormality determination means, filter regeneration means, determination prohibition means, determination value change means, combustion control means, moisture determination means, first moisture elimination determination means, (2nd moisture elimination determination means, collection rate determination means).

Claims (8)

The exhaust pipe of the internal combustion engine is provided with a filter device that collects conductive particulate matter contained in the exhaust gas of the internal combustion engine, and the particulate matter is attached to the upstream side and the downstream side of the filter device. Applied to a system in which an upstream sensor and a downstream sensor are provided as particulate matter detection sensors for detecting the amount of adhesion,
First combustion means for burning particulate matter adhering to the upstream sensor by heating a heater provided in the upstream sensor;
Second combustion means for burning particulate matter adhering to the downstream sensor by heating a heater provided in the downstream sensor;
When the adhering amount of the particulate matter detected by the upstream sensor reaches an implementation reference value for burning the particulate matter by the first combustion means, a combustion request for the upstream sensor is issued. 1 combustion request means;
When the attached amount of the particulate matter detected by the downstream sensor reaches an implementation reference value for burning the particulate matter by the second combustion means, a combustion request for the downstream sensor is issued. 2 combustion request means;
Based on the combustion request issued by the first combustion request means after the combustion by the first combustion means and the combustion request issued by the second combustion request means after the combustion by the second combustion means are performed, An abnormality determining means for determining whether there is an abnormality in the upstream sensor and the downstream sensor;
A sensor control device comprising: Filter regeneration means for performing filter regeneration processing for removing particulate matter collected by the filter device by combustion;
Determination prohibiting means for prohibiting the abnormality determination based on the combustion request by the abnormality determination means until a predetermined time has elapsed after the execution of the filter regeneration process;
A sensor control device according to claim 1. Either when the combustion request is issued by the first combustion requesting means or when the combustion request is issued by the second combustion requesting means, the upstream side by the first combustion means and the second combustion means Combusting particulate matter at the same time for the sensor and the downstream sensor,
The abnormality determination means is configured to provide the upstream when the combustion request by the first combustion request means is issued before the combustion request by the second combustion request means and the combustion request is issued by the first combustion request means. When the difference between the detected amount of the particulate matter by the side sensor and the detected amount of the particulate matter by the downstream sensor is smaller than a predetermined abnormality determination value, it is determined that there is an abnormality.
Filter regeneration means for performing filter regeneration processing for removing particulate matter collected by the filter device by combustion;
Determination value changing means for reducing the abnormality determination value as compared with other periods until a predetermined time elapses after execution of the filter regeneration processing;
A sensor control device according to claim 1 or 2. The exhaust pipe of the internal combustion engine is provided with a filter device that collects conductive particulate matter contained in the exhaust gas of the internal combustion engine, and the particulate matter is attached to the upstream side and the downstream side of the filter device. Applied to a system in which an upstream sensor and a downstream sensor are provided as particulate matter detection sensors for detecting the amount of adhesion,
The upstream sensor and the downstream side when one of the particulate matter adhesion amount detected by the upstream sensor and the particulate matter adhesion amount detected by the downstream sensor reaches a predetermined amount. Combustion control means for simultaneously burning the particulate matter adhering to the sensors by heating the heaters provided in the sensors;
When the particulate matter is burned by the combustion control means, the time required for burning the particulate matter adhering to the upstream sensor and the burning of the particulate matter adhering to the downstream sensor An abnormality determination unit that compares the time required for determining whether there is an abnormality for the upstream sensor and the downstream sensor based on the comparison result;
A sensor control device comprising: Filter regeneration means for performing filter regeneration processing for removing particulate matter collected by the filter device by combustion;
Determination prohibiting means for prohibiting abnormality determination based on the required time by the abnormality determining means until a predetermined time has elapsed after the execution of the filter regeneration processing;
A sensor control device according to claim 4. The abnormality determination means determines that there is an abnormality when a time difference between the time required for the combustion in the upstream sensor and the downstream sensor is smaller than a predetermined abnormality determination value.
Filter regeneration means for performing filter regeneration processing for removing particulate matter collected by the filter device by combustion;
Determination value changing means for reducing the abnormality determination value as compared with other periods until a predetermined time elapses after execution of the filter regeneration processing;
A sensor control device according to claim 4 or 5. The exhaust pipe of the internal combustion engine is provided with a filter device that collects conductive particulate matter contained in the exhaust gas of the internal combustion engine, and the particulate matter is attached to the upstream side and the downstream side of the filter device. Applied to a system in which an upstream sensor and a downstream sensor are provided as particulate matter detection sensors for detecting the amount of adhesion,
A wetness determining means for determining that the upstream sensor and the downstream sensor are wetted after the internal combustion engine is started;
First water depletion determining means for determining the timing at which the water on the upstream sensor has been removed based on the sensor output of the upstream sensor after the water determination on the two sensors is determined by the water determination means. When,
Second moisture elimination determination means for determining the timing at which the moisture of the downstream sensor is eliminated based on the sensor output of the downstream sensor after the moisture judgment of the sensors is judged by the moisture judgment means. When,
About the upstream sensor and the downstream sensor based on the timing of water removal determined by the first water removal cancellation determination means and the timing of water exposure cancellation determined by the second water exposure cancellation determination means An abnormality determination means for determining the presence or absence of abnormality;
A sensor control device comprising: The exhaust pipe of the internal combustion engine is provided with a filter device that collects conductive particulate matter contained in the exhaust gas of the internal combustion engine, and the particulate matter is attached to the upstream side and the downstream side of the filter device. Applied to a system in which an upstream sensor and a downstream sensor are provided as particulate matter detection sensors for detecting the amount of adhesion,
A collection rate determination means for determining that the collection rate of the particulate matter by the filter device is in a low state;
The upstream sensor based on the output change of the upstream sensor and the output change of the downstream sensor when the collection device determines that the filter device is in a low collection rate state. And an abnormality determining means for determining the presence or absence of an abnormality with respect to the downstream sensor,
A sensor control device comprising:

Images