JP2017193976A - DETERIORATION DETERMINATION DEVICE FOR NOx SENSOR - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a deterioration determination device for an NOx sensor capable of accurately determining deterioration of responsiveness of the NOx sensor without deteriorating exhaust gas characteristics.SOLUTION: The deterioration determination device for an NOx sensor is provided downstream of an NOx occlusion reduction type NOx catalyst 9 to determine deterioration of responsiveness of an NOx sensor 23 that reacts with NOx and ammonia. In the deterioration determination device, in order to emit and reduce NOx occluded by the NOx catalyst 9, after completion of rich spike for supplying a reducing agent to the NOx catalyst 9, a response time TM_NOxRES is calculated as a reduction speed parameter indicating reduction speed of a detection value NOx_SG obtained by the NOx sensor 23 (Step 60), and by comparing the response time TM_NOxRES with a reference time TM_REF, deterioration of responsiveness of the NOx sensor 23 is determined (Steps 61-63 in Fig. 6).SELECTED DRAWING: Figure 7

Description

  The present invention relates to a NOx sensor deterioration determination device that determines deterioration of a NOx sensor provided downstream of a NOx catalyst that purifies NOx in exhaust gas.

  As this type of conventional degradation determination device, for example, one disclosed in Patent Document 1 is known. In this conventional apparatus, a NOx occlusion reduction type is used as the NOx catalyst, and a type that reacts with NOx and ammonia (NH3) is used as the NOx sensor. This device calculates the amount of NOx occluded in the NOx catalyst during the lean burn operation in which the air-fuel mixture leaner than the stoichiometric air-fuel ratio is burned, and when the calculated amount of NOx reaches a predetermined amount Further, in order to recover the storage capacity of the NOx catalyst, a rich spike for supplying a reducing agent to the NOx catalyst is executed.

  Further, when determining an abnormality in the NOx sensor, the rich spike is executed a plurality of times as a two-stage double rich spike, and each double rich spike is necessary for releasing / reducing NOx stored in the NOx catalyst. An excessive amount of reducing agent is supplied to the NOx catalyst. Then, when such excessive reducing agent is supplied to the NOx catalyst, ammonia is generated and detected by the NOx sensor, so that the detected value of the NOx sensor in a predetermined period after the reducing agent is supplied. When the integrated value is greater than the predetermined upper limit value, it is determined that an output increase abnormality has occurred in the NOx sensor, and when the integrated value is smaller than the predetermined lower limit value, an output decrease abnormality has occurred in the NOx sensor. judge.

Japanese Patent No. 4485553

  In the conventional apparatus described above, in order to determine the abnormality of the NOx sensor, an excessive reducing agent exceeding the amount necessary for releasing / reducing NOx stored in the NOx catalyst is supplied to the NOx catalyst by the double rich spike. Is done. As a result, in the state where oxygen is not present in the NOx catalyst, the excess reducing agent is supplied to the NOx catalyst, so that CO and HC in the reducing agent are discharged from the NOx catalyst. End up. In addition, since such a double rich spike is executed a plurality of times, the exhaust gas characteristics are further deteriorated. Moreover, since abnormality determination is performed based on the integrated value of the detected value of the NOx sensor, it is not possible to satisfactorily determine deterioration of the responsiveness of the NOx sensor.

  The present invention has been made in order to solve the above-described problems, and provides a NOx sensor deterioration determination device that can accurately determine deterioration in response of a NOx sensor without deteriorating exhaust gas characteristics. The purpose is to provide.

  In order to achieve the above object, the invention according to claim 1 is provided downstream of the NOx occlusion reduction type NOx catalyst 9 for purifying NOx in the exhaust gas discharged from the internal combustion engine 3. A NOx sensor deterioration determination device for determining deterioration of the NOx sensor 23 that reacts with ammonia, and a rich spike for supplying a reducing agent to the NOx catalyst 9 in order to release and reduce NOx occluded in the NOx catalyst 9 Rich spike means to be executed (the fuel injection valve 6 in the embodiment (hereinafter, the same in this section), ECU 2, step 9 in FIG. 2) and the decrease rate of the detected value NOx_SG of the NOx sensor 23 after the end of the rich spike. Deceleration rate parameter calculation means (ECU2, step 6 in FIG. 6) for calculating the decrease rate parameter (response time TM_NOxRES) ) And, based on the decrease rate parameter calculated, it determines deterioration determining unit (ECU 2 the response of the deterioration of the NOx sensor 23, step 61 to 63 in FIG. 6, characterized in that it comprises as FIG. 7), the.

  The present invention focuses on the following operating characteristics of the NOx catalyst and NOx sensor associated with execution / stop of the rich spike for supplying the reducing agent to the NOx catalyst. For example, it is assumed that a rich spike is executed in order to release and reduce NOx stored in the NOx catalyst during the lean burn operation. In this case, in the NOx catalyst, NOx is reduced by the reaction between the reducing component in the reducing agent supplied by the rich spike and the NOx stored in the NOx catalyst. At the same time, in the reducing atmosphere, ammonia (NH 3) is generated by the reaction between the reducing component in the reducing agent and nitrogen (N 2) in the exhaust gas.

  For this reason, as shown in FIG. 8 (d), during the rich spike, the NOx concentration (broken line) of the exhaust gas flowing out from the NOx catalyst once increases and then decreases to a value of 0, whereas ammonia continuously As it is produced, it continues to flow out of the NOx catalyst. Further, since the NOx sensor arranged on the downstream side of the NOx catalyst has a characteristic of reacting not only with NOx but also with ammonia, as shown by a solid line in FIG. 8D, the detected value of the NOx sensor during the rich spike. (NOx_SG) is maintained at a value corresponding to the ammonia concentration. From this state, when the rich spike is finished and the lean burn operation is restarted (time t3 in FIG. 8), NOx in the exhaust gas is occluded in the NOx catalyst and does not flow out of the NOx catalyst. On the other hand, since the production of ammonia in the NOx catalyst is stopped, the ammonia concentration of the exhaust gas flowing into the NOx sensor rapidly decreases from the value at the end of the rich spike to the value 0.

  The present invention pays particular attention to the above-mentioned phenomenon, and if the detected value of the NOx sensor rapidly decreases in response to the ammonia concentration that rapidly decreases after the end of the rich spike, the response of the NOx sensor. Is determined to be good. Otherwise, it is determined that the response deterioration of the NOx sensor has occurred. Based on this point of view, according to the present invention, a reduction rate parameter indicating the reduction rate of the detected value of the NOx sensor after the end of the rich spike is calculated, and the response of the NOx sensor is calculated based on the calculated reduction rate parameter. Therefore, the determination can be made with high accuracy. Further, unlike the conventional apparatus, it is only necessary to use a rich spike that is executed to restore the storage capacity of the NOx catalyst, a special operation mode of the internal combustion engine for determination is unnecessary, and excessive Since it is not necessary to supply a reducing agent, the exhaust gas characteristics are not deteriorated.

  The invention according to claim 2 is the NOx sensor deterioration determination device according to claim 1, wherein the pre-response reference for acquiring the detected value NOx_SG of the NOx sensor 23 detected immediately before the end of the rich spike as the pre-response reference value NOx_RES_B. Based on the value acquisition means (ECU2, step 36 of FIG. 4) and the acquired pre-response reference value NOx_RES_B, the reference interval (10% response value NOx_RES_10, 90% response value NOx_RES_90) between the pre-response reference value NOx_RES_B and the value 0. ) To set the reference interval setting means (ECU 2, steps 37 and 38 in FIG. 4), and the decrease speed parameter calculation means is configured to detect that the detected value NOx_SG of the NOx sensor 23 passes the reference interval after the end of the rich spike. The response time TM_NOxRES required for the The deterioration determining means calculates (step 60 in FIG. 6), and determines that the deterioration of the responsiveness of the NOx sensor 23 has occurred when the calculated response time TM_NOxRES is larger than the predetermined reference time TM_REF. (Steps 61 and 63 in FIG. 6).

  According to this configuration, the detection value of the NOx sensor detected immediately before the end of the rich spike is acquired as the reference value before response, and between the reference value before response and the value 0 based on the acquired reference value before response. Set the reference interval. The pre-response reference value corresponds to the initial value when the detected value of the NOx sensor starts to decrease after the end of the rich spike, and the value 0 corresponds to the final value of the detected value that decreases. Therefore, based on the reference value before response, the reference interval can be appropriately set between the reference value before response and the value 0.

  In addition, according to this configuration, the response time required for the detected value of the NOx sensor to pass through the reference section after the end of the rich spike is calculated as a decrease speed parameter, and the calculated response time is a predetermined reference time. Is greater than the NOx sensor, it is determined that the NOx sensor response has deteriorated. The response time calculated as described above represents the decrease rate of the detected value of the NOx sensor after the end of the rich spike, and the longer the response time, the lower the response of the NOx sensor. Therefore, when the response time is longer than the reference time, it can be accurately determined that the responsiveness of the NOx sensor has deteriorated.

  According to a third aspect of the present invention, in the NOx sensor deterioration determination apparatus according to the second aspect, the reference interval setting means includes a predetermined interval (10%) excluding the vicinity of the reference value before response and the vicinity of the value 0 as the reference interval. Response value NOx_RES_10, 90% response value NOx_RES_90) is set (steps 37 and 38 in FIG. 4).

  The detected value of the NOx sensor tends to become unstable near the reference value before response because it is immediately after the end of the rich spike, and the amount of change becomes smaller as the value approaches 0 near the convergence. Does not express well. According to this configuration, by setting the reference section to a predetermined section excluding the vicinity of the reference value before response and the vicinity of value 0, the vicinity of the reference value before response and the vicinity of value 0 are excluded from the reference section. It is possible to appropriately calculate a response time that favorably reflects the responsiveness of the detected value, thereby improving the accuracy of determination of response deterioration.

  According to a fourth aspect of the present invention, in the NOx sensor deterioration determining apparatus according to the second or third aspect of the present invention, a flow rate parameter acquisition that acquires an exhaust gas flow rate parameter (intake air amount GAIR) representing a flow rate of the exhaust gas flowing into the NOx sensor 23 is obtained. Means (air flow sensor 21) and reference time setting means (ECU 2, step 42 in FIG. 4, FIG. 5) for setting the reference time TM_REF in accordance with the acquired exhaust gas flow velocity parameter.

  The larger the flow rate of the exhaust gas flowing into the NOx sensor, the shorter the calculated response time and the higher the apparent responsiveness of the NOx sensor. From such a point of view, in the configuration of claim 4, an exhaust gas flow velocity parameter representing the flow velocity of the exhaust gas flowing into the NOx sensor is acquired, and a reference time is set according to the acquired exhaust gas flow velocity parameter. Thereby, the reference time can be appropriately set according to the flow rate of the exhaust gas, and thereby the accuracy of determination of response deterioration can be further improved.

  The invention according to claim 5 is the NOx sensor deterioration determination device according to any one of claims 2 to 4, wherein the deterioration determination means determines the deterioration of the responsiveness of the NOx sensor 23 each time a rich spike is executed. Are repeatedly executed, and when the response time TM_NOxRES is larger than the reference time TM_REF, it is temporarily determined that the responsiveness of the NOx sensor 23 has deteriorated (steps 61 and 63 in FIG. 6), and the temporary determination is predetermined. When the number of times N_REF continues, it is determined that deterioration of the responsiveness of the NOx sensor 23 has occurred (steps 76 and 77 in FIG. 7).

  According to this configuration, when the response time is longer than the reference time, it is temporarily determined that the NOx sensor has deteriorated in responsiveness, and such provisional determination of the occurrence of deterioration continues for a predetermined number of times. The provisional determination is confirmed on the condition. Thereby, it is possible to avoid erroneous determination due to temporary fluctuations in the detected value of the NOx sensor, and to maintain good determination accuracy.

1 is a diagram schematically showing a NOx sensor deterioration determination device according to an embodiment of the present invention, together with an internal combustion engine to which the device is applied. FIG. It is a flowchart which shows an engine control process. It is a main flow which shows the deterioration determination process of a NOx sensor. It is a flowchart which shows the reference | standard parameter calculation process among the deterioration determination processes of FIG. 5 is a table for setting a reference time used in the processing of FIG. It is a flowchart which shows the response time calculation process among the deterioration determination processes of FIG. It is a flowchart which shows the determination confirmation process among the deterioration determination processes of FIG. It is a figure which shows typically the operation example obtained by the deterioration determination process of FIG.

  Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. An internal combustion engine (hereinafter referred to as “engine”) 3 shown in FIG. 1 is a gasoline engine having, for example, four cylinders (not shown), and is mounted on a vehicle (not shown) as a power source.

  In the intake passage 4 of the engine 3, an air flow sensor 21, a throttle valve 5, and a fuel injection valve 6 are provided in order from the upstream side. The opening degree TH of the throttle valve 5 (hereinafter referred to as “throttle valve opening degree”) TH is controlled by the ECU (electronic control unit) 2 via the TH actuator 5a, and is thereby sucked into the cylinder via the intake passage 4. The intake air amount GAIR is controlled. The air flow sensor 21 detects the intake air amount GAIR and outputs a detection signal to the ECU 2.

  The fuel injection valve 6 is provided for each cylinder so as to face an intake port (not shown) (only one is shown). The valve opening time of the fuel injection valve 6 is controlled by the ECU 2, thereby controlling the fuel injection amount. The engine 3 is further provided with an ignition plug (not shown) for igniting the air-fuel mixture in the combustion chamber for each cylinder. The operation of this spark plug is also controlled by the ECU 2.

  A three-way catalyst 8, a LAF sensor 22, a NOx catalyst 9, and a NOx sensor 23 are provided in the exhaust passage 7 of the engine 3 in order from the upstream side. The three-way catalyst 8 purifies the exhaust gas at a high purification rate by oxidizing HC and CO in the exhaust gas and reducing NOx when the exhaust gas is in a stoichiometric atmosphere.

  The NOx catalyst 9 is of a so-called NOx occlusion reduction type, and stores (adsorbs / holds) NOx in the exhaust gas in an oxidizing atmosphere where the oxygen concentration in the exhaust gas is high, while the concentration of the reducing agent in the exhaust gas is high. In a reducing atmosphere, the stored exhaust gas is purified by releasing the stored NOx and reducing it with a reducing agent.

  The LAF sensor 22 linearly detects the oxygen concentration of the exhaust gas flowing into the NOx catalyst 9 from the three-way catalyst 8 in a wide range of air-fuel ratios from the rich side to the lean side with respect to the theoretical air-fuel ratio, and the detection signal is detected by the ECU 2. Output to. The ECU 2 calculates the air-fuel ratio (hereinafter referred to as “actual air-fuel ratio”) KACT of the air-fuel mixture based on the detection value O2_SG of the LAF sensor 22. The actual air-fuel ratio KACT and a target air-fuel ratio KCMD described later are represented by an equivalence ratio.

  The NOx sensor 23 detects the NOx concentration of the exhaust gas flowing in from the NOx catalyst 9 and outputs a detection signal to the ECU 2. The NOx sensor 23 is of a so-called limit current type, for example, and has a characteristic of reacting not only with NOx in exhaust gas but also with ammonia (NH3). For this reason, the detected value NOx_SG of the NOx sensor 23 reflects both the NOx concentration and the ammonia concentration, and represents the sum of both.

  A crank angle sensor 24 is provided on the crankshaft (not shown) of the engine 3. The crank angle sensor 24 outputs a CRK signal and a TDC signal, which are pulse signals, to the ECU 2 as the crankshaft rotates. The CRK signal is output every predetermined crank angle (for example, 30 °). The ECU 2 calculates the engine speed (hereinafter referred to as “engine speed”) NE of the engine 3 based on the CRK signal. The TDC signal is a signal indicating that the piston is in the vicinity of the top dead center (TDC) at the start of the intake stroke in any cylinder. When the engine 3 has four cylinders as in the present embodiment, the crank angle Output every 180 °.

  The ECU 2 further detects from the accelerator opening sensor 25 a detection signal indicating an accelerator opening AP, which is an operation amount of an accelerator pedal (not shown) of the vehicle, from the vehicle speed sensor 26 indicating a vehicle speed (vehicle speed) VP. Each signal is input.

  The ECU 2 is composed of a microcomputer including a CPU, a RAM, a ROM, an I / O interface (all not shown), and the like. The ECU 2 executes a control process for the engine 3 and a deterioration determination process for the NOx sensor 23 in accordance with the control program stored in the ROM based on the detection signals of the various sensors 21 to 26 described above. In the present embodiment, the ECU 2 constitutes the rich spike means, the decrease speed parameter calculation means, the deterioration determination means, the pre-response reference value acquisition means, the reference interval setting means, and the reference time setting means of the present invention.

  FIG. 2 shows an engine control process executed by the ECU 2. In this control process, the operation mode of the engine 3 is: (1) stoichiometric operation in which the air-fuel mixture is controlled to the stoichiometric air-fuel ratio and burned; In order to release or reduce the NOx occluded in the NOx catalyst 9 in the burn operation or (3) the rich spike operation in which the air-fuel mixture is controlled to be richer than the stoichiometric air-fuel ratio, it is determined and executed. is there. This process is executed for each cylinder in synchronization with the generation of the TDC signal.

  In this process, first, in step 1 (illustrated as “S1”, the same applies hereinafter), it is determined whether or not the rich spike flag F_RS is “1”. The rich spike flag F_RS is set to “1” during execution of the rich spike operation (see FIGS. 8A and 8B). When the answer to step 1 is NO and the rich spike operation is not being executed, it is determined whether or not the execution condition for the lean burn operation is satisfied (step 2).

  The execution conditions are, for example, (a) the engine speed NE is within a predetermined range, (b) the vehicle speed VP is within a predetermined range, and (c) the required torque TREQ is equal to or less than a predetermined value in a low load region. When all of these conditions (a) to (c) are satisfied, it is determined that the lean burn operation execution condition is satisfied. The required torque TREQ is calculated by searching a predetermined map (not shown) according to the engine speed NE and the accelerator pedal opening AP.

  When the answer to step 2 is NO and the execution condition of the lean burn operation is not satisfied, the process proceeds to step 3, the stoichiometric operation is executed, and this process is terminated. In this stoichiometric operation, the target air-fuel ratio KCMD of the air-fuel mixture is set to a value 1.0 corresponding to the stoichiometric air-fuel ratio. Further, the throttle valve opening TH is controlled according to the engine speed NE and the required torque TREQ, and the fuel injection amount is controlled so that the detected actual air-fuel ratio KACT becomes the target air-fuel ratio KCMD.

  On the other hand, if the answer to step 2 is YES and the execution condition of the lean burn operation is satisfied, it is determined whether or not the lean burn flag F_LEAN is “1” (step 4). If the answer is NO and the current processing cycle corresponds to immediately after the execution condition of the lean burn operation is satisfied, the lean burn flag F_LEAN is set to “1” in step 5 and then the lean burn operation is executed (step 6) The process is terminated.

  In this lean burn operation, the target air-fuel ratio KCMD is set to a value smaller than 1.0, which corresponds to an air-fuel ratio that is leaner than the stoichiometric air-fuel ratio (FIG. 8A). Further, the throttle valve 5 is controlled to be fully opened, and the fuel injection amount is controlled so that the actual air-fuel ratio KACT becomes the target air-fuel ratio KCMD. By the above control, the air-fuel mixture leaner than the stoichiometric air-fuel ratio burns during the lean burn operation, and NOx in the exhaust gas is occluded in the NOx catalyst 9.

  After the lean burn operation is started in this way, the answer to step 4 is YES, and in this case, the process proceeds to step 7 where the detected value NOx_SG of the NOx sensor 23 is equal to or greater than a predetermined threshold value NOx_REF. It is determined whether or not. When this answer is NO, assuming that the amount of NOx occluded in the NOx catalyst 9 is still small, the routine proceeds to step 6 and the lean burn operation is continued.

  On the other hand, when the answer to step 7 is YES and NOx_SG ≧ NOx_REF is satisfied (time t1 in FIG. 8D), NOx stored in the NOx catalyst 9 approaches a saturated state and starts to flow out of the NOx catalyst 9. As a result, the rich spike flag F_RS is set to “1” (step 8), and the rich spike operation is executed in order to recover the storage capacity of the NOx catalyst 9 (step 9), and this process is terminated.

  After the rich spike operation is thus started, the answer to step 1 is YES, and in this case, the process proceeds to step 10 where the detection value O2_SG of the LAF sensor 22 is equal to or greater than a predetermined threshold value O2_REF. It is determined whether or not. When the answer is NO, the process proceeds to step 9 and the rich spike operation is continued.

  On the other hand, when the answer to step 10 is YES and O2_SG ≧ O2_REF is satisfied (FIG. 8 (c), time point t3), the air-fuel ratio of the exhaust gas flowing into the NOx catalyst 9 becomes sufficiently rich, and the NOx catalyst 9 As a result, the rich spike flag F_RS is reset to “0” (step 11), the rich spike operation is stopped (step 12), and this process is terminated. Thereafter, the engine 3 returns to lean burn operation.

  Next, the deterioration determination process of the NOx sensor 23 executed by the ECU 2 will be described with reference to FIGS. This deterioration determination process uses the opportunity of the rich spike operation performed as described above, and based on the decrease rate of the detected value NOx_SG of the NOx sensor 23 after the end of the rich spike operation, the response of the NOx sensor 23 is determined. Deterioration (hereinafter referred to as “response deterioration” as appropriate) is determined, and is executed at a predetermined period ΔT (for example, 10 ms).

  FIG. 3 shows a main flow of the deterioration determination process. In this process, first, in step 21, a reference parameter calculation process is executed. This process calculates various reference parameters serving as a reference for calculating the response time of the detection value NOx_SG and determining the response deterioration based on the detection value NOx_SG obtained during the rich spike operation. This subroutine is executed.

  Next, in step 22, response time calculation processing is executed. This process calculates the response time of the detected value NOx_SG of the NOx sensor 23 immediately after the end of the rich spike operation, and makes a provisional determination of the response deterioration of the NOx sensor 23 based on the calculated response time. This is executed by the subroutine of FIG.

  Next, in step 23, determination determination processing is executed, and the processing in FIG. This process is a process for determining the presence or absence of response deterioration of the NOx sensor 23 based on the result of the provisional determination in step 22, and is executed by a subroutine shown in FIG.

  Next, the reference parameter calculation process executed in step 21 of FIG. 3 will be described with reference to FIG. In this process, first, in step 31, it is determined whether or not the rich spike flag F_RS is “1”. When the answer to step 31 is YES and the rich spike operation is being performed, the detected value NOx_SG of the NOx sensor 23 obtained this time is stored in a register having a register number i = 0 among a plurality of registers (not shown). Stored as register value NOx_RG (0) (step 32). The plurality of registers includes (NRES + 1) registers whose register number i is 0 to a predetermined value NRES (for example, 10).

  Next, the register value NOx_RG (i) of the register number i is shifted to the register value NOx_RG (i + 1) of the register number (i + 1) (step 33). Next, a response determination permission flag F_NOxRES instructing permission of response determination is set to “0” (step 34), and this process is terminated. Therefore, when the operations of the above steps 32 and 33 are repeated NRES times or more, the latest NRES detection values NOx_SG detected during the rich spike operation become the register values NOx_RG (1) to NOx_RG (NRES), respectively. Is remembered as

  If the answer to step 31 is NO and the rich spike operation is not being performed, it is determined whether or not the previous value F_RS_LAST of the rich spike flag is “1” (step 35). When this answer is NO, this processing is ended as it is.

  On the other hand, when the answer to step 35 is YES and the current processing cycle corresponds to immediately after the end of the rich spike operation (time point t3 in FIG. 8), the register values NOx_RG (1) to NOx_RG (NRES stored in step 33). ) (ΣNOx_RG) divided by a predetermined value NRES (average value) is calculated as a pre-response reference value NOx_RES_B (step 36).

  Next, by multiplying the calculated pre-response reference value NOx_RES_B by, for example, 0.9 as a predetermined coefficient, the detection value NOx_SG is equivalent to a value obtained by reducing the pre-response reference value NOx_RES_B by 10%, a 10% response value NOx_RES_10 (FIG. 8). (See (d)) is calculated (step 37). Similarly, by multiplying the pre-response reference value NOx_RES_B by, for example, 0.1 as a predetermined coefficient, the 90% response value NOx_RES_90 corresponding to a value obtained by reducing the detection value NOx_SG by 90% from the pre-response reference value NOx_RES_B (FIG. 8 (d )) Is calculated (step 38). The reference interval is defined by these 10% response value NOx_RES_10 and 90% response value NOx_RES_90.

  Next, upon completion of the calculation of the reference parameter, the response determination permission flag F_NOxRES is set to “1” to instruct permission of response determination (step 39), and a 10% response value passing flag F_RES10 and a later-described The 90% response value passage flag F_RES90 is reset to “0” (steps 40 and 41).

  Next, by searching the table shown in FIG. 5 according to the intake air amount GAIR detected by the airflow sensor 21, a reference time TM_REF serving as a reference for response time is calculated (step 42), and this process is terminated. . This intake air amount GAIR is used as an exhaust gas flow rate parameter indicating the flow rate of exhaust gas flowing into the NOx sensor 23. In this table, the reference time TM_REF is set to a smaller value as the intake air amount GAIR is larger, that is, as the flow rate of exhaust gas is larger.

  Next, the response time calculation process executed in step 22 of FIG. 3 will be described with reference to FIG. In this process, first, in step 51, it is determined whether or not a response determination permission flag F_NOxRES is “1”. When this answer is NO, this processing is ended as it is.

  On the other hand, when the answer to step 51 is YES and the response determination is permitted, in the next step 52 and subsequent steps, a response time is calculated and a provisional response determination based on the result is performed. First, in step 52, it is determined whether or not the detected value NOx_SG of the NOx sensor 23 is equal to or less than the 10% response value NOx_RES_10 calculated in step 37. When the answer is NO, the process proceeds to Step 54 described later. On the other hand, if the answer to step 52 is YES and the detected value NOx_SG reaches or passes through the 10% response value NOx_RES_10, the 10% response value passage flag F_RES10 is set to “1” (step 53). Proceed to 54.

  In step 54, it is determined whether or not the detected value NOx_SG is equal to or less than the 90% response value NOx_RES_90 calculated in step 38. When the answer is NO, the process proceeds to Step 56 described later. On the other hand, when the answer to step 54 is YES and the detected value NOx_SG reaches or passes the 90% response value NOx_RES_90, the 90% response value passing flag F_RES90 is set to “1” (step 55). Proceed to 56.

  In this step 56, it is determined whether or not the 10% response value passage flag F_RES10 = 1 and the 90% response value passage flag F_RES90 = 0. When the answer is NO, the process proceeds to Step 58 described later. On the other hand, when the answer to step 56 is YES, that is, when the detected value NOx_SG is between the 10% response value NOx_RES_10 and the 90% response value NOx_RES_90, the response counter value C_RES is incremented (step 57), and the process proceeds to step 58.

  In this step 58, it is determined whether or not the 10% response value passage flag F_RES10 = 1 and the 90% response value passage flag F_RES90 = 1. When this answer is NO, a temporary determination execution flag F_NOxRES_JUD, which will be described later, is set to “0” (step 59), and this process is terminated.

  On the other hand, when the answer to step 58 is YES and the detected value NOx_SG reaches the 90% response value NOx_RES_90, the response time TM_NOxRES is calculated by multiplying the response counter value C_RES by the execution period ΔT of this process (step 60). ). As apparent from the above calculation method, the response time TM_NOxRES is the time from when the detected value NOx_SG passes through the 10% response value NOx_RES_10 to when it passes through the 90% response value NOx_RES_90, that is, through the reference interval. This corresponds to the time required (between t4 and t5 in FIG. 8).

  Next, it is determined whether or not the calculated response time TM_NOxRES is greater than the reference time TM_REF calculated in step 42 of FIG. 4 (step 61). When this answer is NO, since the response time TM_NOxRES is short, it is temporarily determined that the response deterioration of the NOx sensor 23 has not occurred, and the deterioration temporary determination flag F_NOxRES_F is set to “0” (step 62). On the other hand, when the answer to step 61 is YES, since the response time TM_NOxRES is long, it is temporarily determined that the response deterioration of the NOx sensor 23 has occurred, and the deterioration temporary determination flag F_NOxRES_F is set to “1” (step 63). .

  After the above step 62 or 63, the temporary determination execution flag F_NOxRES_JUD is set to “1” in order to indicate that the response deterioration temporary determination has been executed (step 64). Further, the response determination permission flag F_NOxRES is reset to “0” (step 65), and this process is terminated.

  Next, the determination determination process executed in step 23 of FIG. 3 will be described with reference to FIG. In this process, first, in step 71, it is determined whether or not the provisional determination execution flag F_NOxRES_JUD is “1”. When this answer is NO, this processing is ended as it is.

  On the other hand, when the answer to step 71 is YES and the provisional determination of the response deterioration of the NOx sensor 23 is executed, it is determined whether or not the deterioration provisional determination flag F_NOxRES_F representing the result is “1” (step 72). ). If the answer is NO and it is temporarily determined that no response deterioration has occurred, the deterioration confirmation counter value C_RESJUD is reset to 0 (step 73), and the deterioration confirmation flag F_NOxRES_NG is set to “0” (step 74). ), This process is terminated.

  On the other hand, if the answer to step 72 is YES and it is temporarily determined that response deterioration has occurred, the deterioration confirmation counter value C_RESJUD is incremented (step 75). Next, it is determined whether or not the deterioration confirmation counter value C_RESJUD has reached a predetermined number N_REF (step 76). When this answer is NO, this processing is ended as it is.

  On the other hand, when the answer to step 76 is YES, that is, when the temporary determination result that the response deterioration of the NOx sensor 23 has occurred is obtained N_REF continuously for the predetermined number of times, the response deterioration of the NOx sensor 23 occurs. And the deterioration confirmation flag F_NOxRES_NG is set to “1” (step 77), and then the present process is terminated.

  As described above, according to the present embodiment, after the end of the rich spike operation, the response time TM_NOxRES, which is the time required for the detected value NOx_SG of the NOx sensor 23 to pass through the reference interval, is reduced by the detected value NOx_SG. Calculated as a speed parameter. Then, the response deterioration of the NOx sensor 23 is determined by comparing the response time TM_NOxRES with the reference time TM_REF, so that the determination can be performed with high accuracy. Further, unlike the conventional apparatus, it is only necessary to use a rich spike that is executed to restore the storage capacity of the NOx catalyst, a special operation mode of the engine 3 for determination is unnecessary, and an excessive amount is required. Since it is not necessary to supply a reducing agent, the exhaust gas characteristics are not deteriorated.

  In addition, an average value of a plurality of detected values NOx_SG detected immediately before the end of the rich spike operation is calculated as a pre-response reference value NOx_RES_B, and a predetermined coefficient (= 0.9 and 0.1) is added to the pre-response reference value NOx_RES_B. 10% response value NOx_RES_10 and 90% response value NOx_RES_90 calculated by multiplying ()) define a reference interval serving as a reference for calculating the response time TM_NOxRES. As a result, the vicinity of the reference value NOx_RES_B before response and the vicinity of the value 0 are excluded from the reference interval, so that the response time TM_NOxRES that appropriately reflects the responsiveness of the detected value NOx_SG can be calculated appropriately. The determination accuracy can be improved.

  Further, a reference time TM_REF to be compared with the response time TM_NOxRES is set according to the intake air amount GAIR. Thereby, the reference time TM_REF can be appropriately set while reflecting the flow rate of the exhaust gas flowing into the NOx sensor 23, and therefore the accuracy of determination of response deterioration can be further improved.

  Further, when the response time TM_NOxRES is larger than the reference time TM_REF, it is temporarily determined that the response deterioration of the NOx sensor 23 has occurred, and the provisional determination is performed on the condition that such temporary determination continues for a predetermined number of times. Determine. As a result, it is possible to avoid erroneous determination due to temporary fluctuations in the detected value NOx_SG of the NOx sensor 23 and to maintain good determination accuracy.

  In addition, this invention can be implemented in various aspects, without being limited to the described embodiment. For example, in the embodiment, the response time TM_NOxRES required for the detected value NOx_SG to pass through a predetermined reference interval is used as the decreasing rate parameter indicating the decreasing rate of the detected value NOx_SG of the NOx sensor 23. However, the present invention is not limited to this. For example, the decrease amount of the detected value NOx_SG within a predetermined time after the end of the rich spike operation may be used. Alternatively, the decrease rate may be obtained by calculating the decrease amount of the detected value NOx_SG after the end of the rich spike operation and dividing the decrease amount by the time in between.

  Further, as the start and end of the reference section, 10% and 90% response values NOx_RES_10 and 90 obtained by multiplying the reference value before response NOx_RES_B by a predetermined coefficient (= 0.9, 0.1) are used. Of course, the reference interval may be appropriately changed by changing the coefficient.

  Further, in the embodiment, the intake air amount GAIR is used as the exhaust gas flow rate parameter for setting the reference time TM_REF. However, the present invention is not limited to this, and other appropriate representations that favorably express the flow rate of the exhaust gas flowing into the NOx sensor 23. For example, an exhaust gas flow rate detected by a flow sensor provided in the exhaust passage 7 may be used, or an intake air amount calculated based on the engine speed NE, the intake pressure, or the like may be used. Furthermore, instead of setting the reference time TM_REF using the exhaust gas flow velocity parameter, the response time TM_NOxRES may be corrected.

  Furthermore, in the embodiment, when the response time TM_NOxRES is larger than the reference time TM_REF, it is temporarily determined that the response deterioration of the NOx sensor 23 has occurred, and then the temporary determination is confirmed under a predetermined condition. However, the provisional determination may be omitted.

  In the embodiment, the rich spike operation is performed by controlling the air-fuel ratio of the air-fuel mixture combusting in the cylinder to be richer than the stoichiometric air-fuel ratio. However, the present invention is not limited to this. An addition valve may be provided to supply fuel as a reducing agent upstream of the NOx catalyst.

  Furthermore, although embodiment is an example which applied this invention to the engine 3 which is a gasoline engine for vehicles, this invention is not restricted to this, Various engines using fuels other than gasoline, for example, a diesel engine etc. It is also applicable to various engines other than those for vehicles, for example, engines for marine propulsion devices such as outboard motors with a crankshaft arranged in the vertical direction, and other industrial engines. It is. In addition, it is possible to appropriately change the detailed configuration within the scope of the gist of the present invention.

2 ECU (rich spike means, decrease speed parameter calculation means, deterioration determination means, previous detection value acquisition means, reference interval setting means, reference time setting means)
3 Engine 6 Fuel injection valve (Rich spike means)
9 NOx catalyst 21 Air flow sensor (Exhaust gas flow velocity parameter acquisition means)
23 NOx sensor NOx_SG NOx sensor detection value TM_NOxRES Response time (decrease speed parameter)
NOx_RES_B Reference value before response NOx_RES_10 10% response value (reference interval)
NOx_RES_90 90% response value (reference interval)
TM_REF Reference time
GAIR intake air volume (exhaust gas flow rate parameter)

Claims (5)

A NOx sensor deterioration determination device that is provided downstream of a NOx occlusion reduction type NOx catalyst for purifying NOx in exhaust gas discharged from an internal combustion engine and determines deterioration of a NOx sensor that reacts to NOx and ammonia. And
Rich spike means for executing a rich spike for supplying a reducing agent to the NOx catalyst in order to release and reduce NOx stored in the NOx catalyst;
A decrease rate parameter calculating means for calculating a decrease rate parameter indicating a decrease rate of the detected value of the NOx sensor after the end of the rich spike;
A deterioration determining means for determining deterioration of the responsiveness of the NOx sensor based on the calculated decrease speed parameter;
A NOx sensor deterioration determination device comprising: Pre-response reference value acquisition means for acquiring a detection value of the NOx sensor detected immediately before the end of the rich spike as a pre-response reference value;
A reference interval setting means for setting a reference interval between the reference value before response and the value 0 based on the acquired reference value before response;
The decreasing speed parameter calculating means calculates a response time required for the detected value of the NOx sensor to pass through the reference section after the end of the rich spike as the decreasing speed parameter,
2. The deterioration determination unit according to claim 1, wherein when the calculated response time is longer than a predetermined reference time, the deterioration determination unit determines that the deterioration of the responsiveness of the NOx sensor has occurred. NOx sensor deterioration determination device.   3. The NOx sensor deterioration determination apparatus according to claim 2, wherein the reference interval setting means sets the reference interval to a predetermined interval excluding the vicinity of the pre-response reference value and the vicinity of the value 0. A flow velocity parameter acquisition means for acquiring an exhaust gas flow velocity parameter representing a flow velocity of the exhaust gas flowing into the NOx sensor;
Reference time setting means for setting the reference time according to the acquired exhaust gas flow velocity parameter;
The deterioration determination device for a NOx sensor according to claim 2 or 3, further comprising: The deterioration determining means includes
Each time the rich spike is executed, the determination of the deterioration of the responsiveness of the NOx sensor is repeatedly executed,
When the response time is longer than the reference time, it is temporarily determined that the NOx sensor has deteriorated in response, and when the temporary determination continues for a predetermined number of times, the response of the NOx sensor is 5. The NOx sensor deterioration determination apparatus according to claim 2, wherein the deterioration is determined as occurring.

Images