JP2010180743A - Exhaust emission control device of internal combustion engine - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an exhaust emission control device of an internal combustion engine capable of quickly reducing the oxygen concentration, by setting a target value for reducing oxygen low by learning a variation possessed by a device constitution, in the exhaust emission control device for restraining an excessive temperature rise by reducing the oxygen concentration, after being determined as having the possibility of the excessive temperature rise in a DPF. <P>SOLUTION: The oxygen concentration in exhaust gas flowing in the DPF in two stages, is reduced on and after the time t0 of detecting the possibility of causing the excessive temperature rise in the DPF. By learning the variation possessed by the device constitution in advance, the oxygen concentration is quickly reduced by setting a target value T1' lower than a target value T1 in a conventional technology in a first stage of quickly reducing the oxygen concentration. Afterwards, the value is converged to the final target value T2 by feedback control. A setting time up to reaching the target value T2 can be shortened in a learning existing case more than a learning nonexistent case. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an exhaust emission control device for an internal combustion engine.

  Today, with increasing awareness of environmental protection, excellent exhaust purification performance is required for internal combustion engines. Particularly in diesel engines, it is important to remove so-called exhaust particulates (particulate matter, PM) such as black smoke discharged from the engine. For this purpose, a diesel particulate filter (DPF) is often provided in the middle of the exhaust pipe.

  Although most of the PM in the exhaust gas is removed by the DPF collecting the PM, the PM continues to accumulate in the DPF, but the DPF clogs and burns the accumulated PM. It is necessary to regenerate the DPF. In order to burn PM accumulated in the DPF, a technique such as post injection in which fuel is injected after main injection in the cylinder is used.

  However, if the temperature rises too much during DPF regeneration, problems such as melting or cracking of the DPF will occur. For example, when the engine is decelerated under the condition where the DPF is being regenerated and the temperature of the DPF is high and the amount of PM accumulated in the DPF is large, the intake air amount decreases rapidly, so that the heat inside the DPF is downstream by the exhaust gas. Will not be moved to. Accordingly, heat is trapped inside the DPF, and the possibility of excessive temperature rise of the DPF is increased. FIG. 9 shows an example of occurrence of overheating in such a case.

  For example, in Patent Document 1 below, when there is a possibility of excessive DPF temperature rise (exhaust temperature is high and the vehicle is decelerating), the intake throttle valve is throttled and the EGR valve is fully opened to reduce oxygen in the exhaust. A technique for preventing PM combustion on the DPF is disclosed.

JP 2002-188493 A

  In order to prevent rapid combustion of PM and suppress excessive temperature rise of DPF, it is necessary to control the oxygen concentration to a predetermined value quickly and with high accuracy. This oxygen concentration is very low, for example, about 2 to 3%. If the oxygen concentration is too low, engine misfire or white smoke occurs, and conversely, if the oxygen concentration is too high, the DPF overheats. End up.

  Further, since a diesel engine is usually operated at a high oxygen concentration (for example, 10 to 20% or more), it is necessary to greatly reduce the oxygen concentration in order to suppress excessive temperature rise. If it is slowly reduced, PM burns rapidly in the process of reduction, and overheating cannot be suppressed. Therefore, it is necessary to adjust the oxygen concentration to a predetermined low concentration value with high accuracy and high response, but there is a problem that this is difficult in the prior art.

  Accordingly, in view of the above problems, the problem to be solved by the present invention is an exhaust purification device that reduces the oxygen concentration and suppresses the excessive temperature increase when it is determined that there is a possibility of excessive temperature increase of the DPF. It is an object of the present invention to provide an exhaust purification device for an internal combustion engine that can set the target value for rapid reduction of the oxygen concentration low by learning the variation of the engine, and can quickly reduce the oxygen concentration.

Means for Solving the Problems and Effects of the Invention

  In order to achieve the above object, an exhaust gas purification apparatus for an internal combustion engine according to the present invention includes a filter disposed in an exhaust passage of the internal combustion engine for collecting particulate matter, and the particulate matter collected by the filter. A regenerating means for regenerating the filter by burning, a detecting means for detecting that the filter is in an overheatable state in which the temperature of the filter may be excessively increased during regeneration by the regenerating means, and the detecting means A first operation that manipulates an input amount including a fuel injection amount and an intake air amount so that the oxygen concentration in the exhaust gas flowing into the filter is reduced to a target value for rapid reduction when it is detected that the temperature can be raised. The accuracy of convergence of the oxygen concentration to the target value for rapid reduction is improved by obtaining the deviation between the command value and the measured value of the oxygen concentration when the input means is operated with the reducing means. So that the input amount Compared with a learning means for correcting and a rapid reduction target value set in order to avoid that the oxygen concentration is excessively reduced when the input amount is not corrected by the learning means, the learning means And setting means for setting a target value for rapid reduction used in the first reduction means to be low when the input amount is corrected according to the above.

  Thus, in the exhaust gas purification apparatus for an internal combustion engine according to the present invention, if there is a risk of overheating during the regeneration of the filter that collects particulate matter, the system aims to reduce the oxygen concentration and avoid overheating In this case, the machine difference or the like of the apparatus is learned in advance, and the input amount is corrected based on the learned machine difference. Therefore, the oxygen concentration can be reduced to the target value for rapid reduction with high accuracy. Therefore, since the possibility that misfire or white smoke is generated due to excessive reduction of the oxygen concentration is reduced, the target value for rapid reduction can be set low. Therefore, since the target value for rapid reduction can be set low and the oxygen concentration can be rapidly and accurately reduced to the target value, the exhaust gas purification device that effectively reduces the excessive temperature rise of the filter by rapidly reducing the oxygen concentration Can be realized.

  Further, the learning means may include first learning means for operating the input amount and acquiring the deviation in a state where it is not detected by the detection means that the overheatable state is possible.

  Thus, by performing accurate learning in a state where there is no danger of overheating, it is possible to acquire information on the deviation between the command value and the actual value when the oxygen concentration is reduced. Therefore, it is possible to correct the input amount including the fuel injection amount and the intake air amount when the oxygen concentration is reduced by the information on the deviation. Therefore, by correcting the input amount, the oxygen concentration can be accurately reduced to the rapid reduction target value, so that the rapid reduction target value can be set low. Since the oxygen concentration can be rapidly and accurately reduced to the target value for rapid reduction set to a low value, it is possible to quickly reduce the oxygen concentration and effectively suppress the occurrence of overheating of the filter.

  The first learning means may acquire the deviation for each operating condition of the internal combustion engine.

  Thereby, the input value can be corrected with high accuracy for each operation condition by having the learned value acquired for each operation condition. Therefore, even if the operating conditions vary, the oxygen concentration can be accurately reduced to the rapid reduction target value by appropriately correcting the input amount accordingly, so that the rapid reduction target value can be set low. Since the oxygen concentration can be rapidly and accurately reduced to the target value for rapid reduction set to a low value, it is possible to quickly reduce the oxygen concentration and effectively suppress the occurrence of overheating of the filter.

  The first learning means may acquire the deviation during idle operation of the internal combustion engine.

  Thus, by performing learning during idle operation where the oxygen concentration is likely to vary, a high effect can be obtained with simple learning. And by correcting the input amount appropriately, the target value for rapid reduction can be set low, and the oxygen concentration can be rapidly and accurately reduced to the target value for rapid reduction that is set low, so the oxygen concentration is quickly reduced. Thus, it is possible to effectively suppress the excessive temperature rise of the filter.

  The internal combustion engine may be mounted on an automobile, and the learning unit may update the deviation value for each predetermined travel distance of the automobile or for each predetermined time.

  As a result, when the characteristics of the constituent articles change over time due to traveling for a predetermined distance or the elapse of a predetermined time, the learning value is updated appropriately, so that the input value when reducing the oxygen concentration can be corrected with a highly accurate learning value. . Therefore, the target value for rapid reduction can be set low and the occurrence of excessive temperature rise can be effectively suppressed. It is also possible to suppress learning at an unnecessary frequency even though it has not changed over time.

  The oxygen concentration in the exhaust gas flowing into the filter by the first reduction means is controlled to be close to the target value for rapid reduction, and the oxygen concentration in the exhaust gas flowing into the filter is lower than the target value for rapid reduction. It may be possible to provide a second reduction means for executing control to bring it close to the excessive temperature rise avoidance target value.

  As a result, first, when the oxygen concentration in the exhaust gas flowing into the filter rapidly to the target value for rapid reduction that is set lower by learning is reduced, and then the oxygen concentration is further reduced to the target value for avoiding excessive temperature rise, By this two-stage oxygen concentration reduction process, it is possible to achieve a reduction in oxygen concentration that has both rapidity and high accuracy, and thus suppression of excessive temperature rise of the filter.

  The second reduction means may perform control to feed back a measured value of the oxygen concentration in the exhaust gas flowing into the filter.

  Thereby, since feedback control is used in the second stage oxygen concentration reduction, it is possible to converge to the target value for avoiding excessive temperature rise with high accuracy.

  And a downstream measuring means for measuring the oxygen concentration in the exhaust passage downstream of the filter; and a calculating means for calculating the oxygen concentration in the exhaust gas flowing into the filter from the measurement value by the downstream measuring means, The measured value of the oxygen concentration may be replaced with the oxygen concentration value calculated by the calculating means.

  Thus, even when the oxygen concentration measuring means is arranged downstream of the filter, the temperature of the filter can be appropriately fed back by correcting the measured value. Therefore, highly accurate oxygen concentration control can be performed, and the excessive temperature rise of the filter can be effectively suppressed.

  Further, the measured value of the oxygen concentration may be a value corrected by a correction unit that corrects the oxygen concentration value by using a difference between the measured value of the oxygen concentration when the fuel injection is stopped and the oxygen concentration in the atmosphere.

  As a result, the measurement value of the oxygen concentration measurement means is corrected with the oxygen concentration in the atmosphere, so high-precision oxygen concentration control can be performed using the high-precision oxygen concentration measurement value, and overheating of the filter is effective. Can be suppressed.

The block diagram in the Example of the exhaust gas purification apparatus of the internal combustion engine in this invention. 3 is a flowchart of learning processing according to the first embodiment. 5 is a flowchart of a cooling process execution determination process according to the first embodiment. 3 is a flowchart of oxygen concentration reduction processing in the first embodiment. 9 is a flowchart of learning processing according to the second embodiment. 10 is a flowchart of oxygen concentration reduction processing in the third embodiment. The figure which shows the example of the time transition of oxygen concentration and DPF temperature. The figure which shows the example of a feedback control system. The figure which shows the example of excessive temperature rise generation | occurrence | production.

  Embodiments of the present invention will be described below with reference to the drawings. First, FIG. 1 is a schematic diagram of a device configuration in Embodiment 1 of an exhaust gas purification device 1 for an internal combustion engine according to the present invention.

  FIG. 1 shows an example of an exhaust purification device 1 configured for a four-cylinder diesel engine 2 (hereinafter simply referred to as an engine). The engine 2 and the exhaust purification device 1 include an intake pipe 3, an exhaust pipe 4, and an EGR pipe 5.

  Air is supplied to the engine 2 through the intake pipe 3. An air flow meter 31 and an intake throttle 32 are disposed in the intake pipe 3. The air flow meter 31 measures the intake air amount. The intake air amount here may be a mass flow rate per unit time, for example. Further, the amount of intake air supplied to the engine 2 increases or decreases by adjusting the opening of the intake throttle 32.

  The engine 2 is equipped with an injector 21 and an engine speed sensor 22. Fuel is supplied into the cylinder by injection from the injector 21. The engine speed sensor 22 measures the speed of the engine 2 (per unit time). The engine speed sensor 22 may be, for example, a crank angle sensor that measures the rotation angle of a crank connected from the engine 2, and the detected value is sent to the ECU 7 to calculate the engine speed.

  Exhaust gas is discharged to an exhaust pipe 4 connected to the engine 2. The exhaust pipe 4 is equipped with oxygen concentration sensors 41 and 42 for measuring the oxygen concentration. However, two oxygen concentration sensors 41 and 42 may not be provided. In the first and second embodiments, only the oxygen concentration sensor 41 on the upstream side of the DPF 6 is used, and in the third embodiment described later, the oxygen concentration sensor 42 on the downstream side of the DPF 6 is used. Only need to be equipped. Exhaust temperature sensors 61 and 62 are disposed on the inlet side and the outlet side of the DPF 6, respectively, and the exhaust temperature at each position is measured. Further, a differential pressure sensor 63 for measuring a front-rear differential pressure (pressure loss) that is a difference in exhaust pressure between the inlet side and the outlet side of the DPF 6 is also provided.

  The EGR pipe 5 is equipped to perform exhaust gas recirculation (EGR) from the exhaust pipe 4 to the intake pipe 3. The EGR pipe 5 is equipped with an EGR valve 51. The exhaust gas recirculation amount is adjusted by opening and closing the EGR valve 51.

  A DPF 6 is disposed in the middle of the exhaust pipe 4. For example, the DPF 6 may have a structure in which the inlet side and the outlet side are alternately clogged in a so-called honeycomb structure. The exhaust discharged during the operation of the engine 2 contains PM (particulate matter), and this PM is collected inside or on the surface of the DPF wall when the exhaust gas passes through the DPF wall having the above structure of the DPF 6. Is done. The DPF 6 may be a DPF with an oxidation catalyst on which an oxidation catalyst is supported.

  The engine 2 and the exhaust emission control device 1 may be mounted on an automobile. An accelerator opening sensor 81 and a vehicle speed sensor 82 are provided. The accelerator opening sensor 81 detects information on the depression amount of the accelerator pedal by the driver. The vehicle speed sensor 82 detects the number of rotations of the wheel by a known configuration such as a rotary encoder, and converts it to the speed of the vehicle. It is assumed that the vehicle travel distance of the vehicle can be calculated by integrating the vehicle speed obtained by the vehicle speed sensor 82 by the ECU 7.

  The measured values of the air flow meter 31, the engine speed sensor 22, the accelerator opening sensor 23, the oxygen concentration sensors 41 and 42, the exhaust temperature sensors 61 and 62, and the differential pressure sensor 63 described above are measured by the electronic control unit 7 (ECU: Electronic). Control Unit). The ECU 7 adjusts and controls the timing and amount of fuel injection to the engine 2 by the injector 21 and the opening between the intake throttle 32 and the EGR valve 51. The ECU 7 may have a structure similar to that of a normal computer, and may include a CPU that performs various calculations and a memory 71 that stores various types of information.

  Every time the amount of PM deposited on the DPF 6 becomes sufficiently large, the deposited PM is removed by burning, and the DPF 6 is regenerated. The method for estimating the PM accumulation amount is, for example, that a functional relationship (map) between the differential pressure across the DPF 6 and the PM accumulation amount is obtained in advance and stored in the memory 71, and the measured value of the differential pressure sensor 63 is the same map. From this, the amount of accumulated PM can be estimated.

  As a regeneration method of the DPF 6, for example, post injection is performed in which fuel is injected from the injector 21 at a timing after the main injection. The unburned fuel that is injected into the cylinder by the post injection and discharged to the exhaust pipe 4 as it is unburned reaches the DPF 6 and is heated by the action of the catalyst carried on the DPF 6, and the PM deposited on the DPF 6 is increased. Burn.

  In Example 1, under the above-described apparatus configuration, when the possibility of occurrence of overheating occurs during regeneration of the DPF 6, processing for quickly reducing the oxygen concentration and suppressing the occurrence of overheating is performed. Execute. At that time, oxygen concentration reduction is performed in two stages. First, in the first stage, a target value (first target value) for rapid reduction is set for the purpose of rapidly reducing the oxygen concentration, and this target value is set. Control to reach quickly. Next, in the second stage, control is performed by feedback control for the purpose of accurately reducing the oxygen concentration to the target value (second target value) for suppressing excessive temperature rise. The first target value is set higher than the second target value. In this embodiment, the input amount for reducing the output with the oxygen concentration as an output is a fuel injection amount, an intake air amount, and an EGR amount (at least one of them).

  Prior to the above process, a process for learning the machine difference of the injector 21 and the air flow meter 31 and a change in characteristics over time is performed so that the first target value can be set as low as possible, that is, close to the second target value. 3 and 4 show the processing procedure for reducing the oxygen concentration, and FIG. 2 shows the processing procedure for learning the machine difference and the change in characteristics over time. 2, 3, and 4 are programmed and stored in the memory 71, and may be automatically processed by the ECU 7 executing them. Note that the learning process of FIG. 2 may be executed in a state where it is determined that there is no possibility of overheating (a state where a cooling process execution flag described later is not 1).

  In the learning process shown in FIG. 2, it is first determined in step S10 whether or not the vehicle has traveled a predetermined distance from the previous learning. If the vehicle has traveled a predetermined distance (S10: YES), the process proceeds to S20. If the vehicle has not traveled a predetermined distance (S10: NO), S10 is repeated until the vehicle travels a predetermined distance. The predetermined distance may be set to 1000 km, for example. Alternatively, in S10, it may be determined whether a predetermined time has elapsed since the previous learning, instead of determining whether the predetermined travel distance has been reached since the previous learning. In that case, the ECU 7 may have a timer function.

  Next, in S20, it is determined whether or not the PM accumulation amount in the DPF 6 is equal to or less than a predetermined value. If the PM deposition amount is less than or equal to the predetermined value (S20: YES), the process proceeds to S30, and if it is greater than the predetermined value (S20: NO), the process returns to S10 again and the same procedure is repeated. As described above, the accumulated amount of PM may be estimated by the ECU 7 from the measured value of the differential pressure sensor 63. For this purpose, a functional relationship (map) between the differential pressure value in the DPF 6 and the PM deposition amount may be obtained in advance and stored in the memory 71.

  As described above, in the process of FIG. 2, the functions of the injector 21 and the air flow meter 31 described below are only performed when the vehicle has traveled a predetermined distance or more after the last learning or when the PM accumulation amount is small. Perform learning of differences and characteristic changes over time. The reason for limiting to the case of a predetermined travel distance or a predetermined time or longer is to appropriately learn the characteristic change of the injector 21 and the air flow meter 31 due to the predetermined distance travel or the predetermined time, and avoid learning at an unnecessary frequency. There is a meaning. The reason that learning is performed only when the amount of accumulated PM is small is to avoid a situation in which PM burns during learning and eventually overheats.

  In S30, an engine speed and a load value indicating the operating condition of the engine 2 are acquired. The engine speed may be detected by the engine speed sensor 22. The load numerical value may be the accelerator opening detected by the accelerator opening sensor 81, or may be an output torque value calculated by the ECU 7 from the engine speed and the accelerator opening. A known method may be used as a method for calculating the output torque.

  In S40, an oxygen concentration target value is set. This target value may be set arbitrarily, but may be the first target value T1, for example.

  In S50, command values for the fuel injection amount, the intake air amount, and the EGR amount are set. Here, the command value means that the measured value of the oxygen concentration sensor 41 is set to the target value set in S40 when there is no machine difference or time-dependent characteristic change of the injector 21, air flow meter 31, EGR valve 51, etc. This is the value of the fuel injection amount, intake air amount, and EGR amount that can be reached. The ECU 7 may calculate these command values according to a map or the like stored in the memory 71 in advance. Here, the EGR amount indicates an exhaust amount recirculated through the EGR pipe 5. The intake air amount may be the opening of the intake throttle 32. The EGR amount may be the opening of the EGR valve 51.

  In S60, the command value set in S50 is executed. That is, the fuel injection amount command value to the injector 21 and the opening to the intake throttle 32 and the EGR valve 51 are adjusted so as to realize the fuel injection amount, intake air amount, and EGR amount command values set in S50. Next, in S70, the measured value of the oxygen concentration sensor 41 is read when the oxygen concentration is reduced by the effect of S60.

  In S80, the difference between the oxygen concentration deviation, that is, the difference between the oxygen concentration target value set in S40 and the oxygen concentration measurement value acquired in S70 is calculated. This deviation is considered to reflect machine differences and changes with time of the injector 21, the air flow meter 31, the intake throttle valve 32, the EGR valve 51, and the like. The numerical value of this deviation is stored in the memory 71 in association with the operating condition obtained in S30.

  The above is the processing procedure of FIG. By repeatedly executing the process of FIG. 2, the memory 71 stores a deviation value for each different operating condition. For example, after dividing a plane indicating the operating conditions, that is, a plane having the engine speed and the load numerical value as coordinate axes into a plurality of areas, the memory 71 may store the deviation in each area. In addition, it is considered that the possibility of excessive temperature rise occurs in the low rotation and low load areas. Therefore, if the operation condition area for storing the deviation is limited to the low rotation and low load areas, it is simpler. It can be a system.

  Next, the processing procedure of FIG. 3 will be described. The main point of the process of FIG. 3 is to determine whether or not the DPF 6 is in a state where there is a possibility of overheating, and if there is a possibility of overheating, a cooling process execution flag that is set when the DPF 6 is cooled is determined. Setting the value to 1. 3 and 4 may be executed during a period in which the DPF 6 is regenerated by executing post injection from, for example, the injector 21 according to a command from the ECU 7 (regeneration means).

  Specifically, in the process of FIG. 3, when the PM accumulation amount is larger than the predetermined threshold A1 (S110: YES), when the internal temperature of the DPF 6 is larger than the predetermined threshold A2 (S120: YES), and the engine 2 is If the vehicle is in the deceleration state (S130: YES), 1 is stored in the value of the cooling process execution flag in S140. In other cases, that is, when the internal temperature of the DPF 6 is equal to or lower than the predetermined threshold A2 (S120: NO), or when the internal temperature of the DPF 6 is equal to or lower than the predetermined threshold A2 (S120: NO), or when the engine 2 is not in the deceleration state. (S130: NO) does not store 1 in the value of the cooling process execution flag.

  The reason for determining whether or not there is a possibility of excessive temperature rise is that if the engine is decelerating while burning PM at a high temperature with a large amount of PM accumulated, the amount of exhaust will decrease. This is because, although a large amount of PM is burning at a high temperature, the heat in the DPF 6 is not carried away by the exhaust gas downstream, and thus an excessive temperature rise may occur.

  The PM accumulation amount may be estimated from the DPF differential pressure as described above. The DPF temperature may be a measured value of the exhaust temperature sensor 61 or 62 or a DPF internal temperature estimated from the measured value. Whether or not the engine is in a deceleration state may be determined by whether or not the measured value of the engine speed sensor 22 or the vehicle speed sensor 82 is decreasing at a speed faster than a predetermined threshold. The above is the processing of FIG.

  Next, FIG. 4 will be described. 4 is based on the learning shown in FIG. 2 and in the situation where the cooling process is determined to be necessary according to the determination condition shown in FIG. Is to reduce the oxygen concentration in the exhaust gas flowing into the exhaust gas to a level where overheating can be avoided.

  In the processing procedure of FIG. 4, it is first determined in S200 whether or not the cooling processing execution flag is 1. If the cooling process execution flag is 1 (S200: YES), the process proceeds to S210. If it is not 1 (S200: NO), the process of FIG. 4 ends.

  In S210, an engine speed and a load numerical value are acquired. As described above, the engine speed may be measured by the engine speed sensor 22. The load numerical value may be an accelerator opening detected by the accelerator opening sensor 81, or an output torque value calculated by the ECU 7 from the engine speed and the accelerator opening.

  In S220, a first target value is calculated. The first target value is a target value in the first stage oxygen concentration reduction as described above. Next, in S230, command values for the fuel injection amount, the intake air amount, and the EGR amount are set. Here, the command value is a value determined so that the measured value of the oxygen concentration sensor 41 accurately reaches the first target value set in S220. However, this command value is a command value corrected based on the deviation information obtained by learning in FIG. 2 under the operating condition obtained in S210.

  Specifically, the correction method here is such that when the measured value is higher than the command value in the learning of FIG. 2 (the actual value has not been reduced to the command value), the oxygen concentration is further decreased. The correction may be made such that the command value for the fuel injection amount is corrected in the increasing direction, the command value for the intake air amount is corrected in the decreasing direction, and the command value for the EGR amount is corrected in the increasing direction. On the other hand, if the measured value is lower than the command value in the learning of FIG. 2 (the actual value has decreased too much to the command value or less), the fuel injection amount command is set so as to increase the final oxygen concentration. The correction may be made to correct the value in the decreasing direction, the command value for the intake air amount in the increasing direction, and the command value for the EGR amount in the decreasing direction. Then, the greater the deviation, the larger the correction width may be.

  Specific examples of the first target value and the like are shown in FIG. In the figure, the first target value in the prior art is indicated by T1, and the first target value in the present invention is indicated by T1 '. Further, T2 in FIG. 7 is a second stage target value (second target value). In other words, in order to suppress the occurrence of excessive temperature rise, the oxygen concentration is ultimately desired to be reduced to T2. Further, when the oxygen concentration is further decreased, an undesired oxygen concentration value that causes an emission deterioration such as engine misfire or white smoke is indicated by T0. The second target value T2 is set higher than the limit concentration T0.

  As described above, when it is detected that there is a possibility of overheating, it is desired to quickly reduce the oxygen concentration. Therefore, we want to set the first target value as low as possible. However, even if it intends to reduce to the first target value, there is a possibility that it will be further reduced due to machine differences and characteristic changes of the injector 21 and the air flow meter 31. In such a case, if the oxygen concentration is reduced to T0 or less, problems such as misfire and white smoke occur as described above, and such a situation must be avoided.

  In the case of the prior art, learning is not executed and, as a result, there is no information about machine differences and characteristic changes of the injector 21 and the air flow meter 31, so the first target is to avoid problems such as misfire and white smoke. The value T1 must be set higher. On the other hand, in the present invention, the learning shown in FIG. 3 is executed to obtain the deviation information of the oxygen concentration reflecting the machine difference and characteristic change of the injector 21 and the air flow meter 31. By correcting the command value of the intake air amount, it can be more accurately reduced to the first target value.

  Therefore, in the present invention, the first target value T1 'can be set lower than the conventional T1, that is, closer to the second target value T2. The difference between the first target value T1 ′ and the second target value T2 is the minimum necessary margin (margin width) taking into account differences other than machine differences of the injector 21 and the air flow meter 31 and changes in characteristics over time. That's fine.

  Returning to FIG. 4, in S240, the fuel injection amount from the injector 21, the intake throttle 32, and the opening degree of the EGR valve 51 are adjusted so as to realize the command value set in S230. FIG. 7 shows an example of transition of oxygen concentration when the oxygen concentration reduction processing is executed.

  Time t0 is the time when the oxygen concentration reduction process is started. A solid line is a response in the case of the present invention, that is, a response when the learning shown in FIG. 2 is executed, and a dotted line is a response in the case of the conventional case, that is, a case where the learning is not executed. As shown in the figure, since the first target value is set lower, the oxygen concentration is reduced more rapidly in the present invention than in the prior art.

  Next, when it is detected in S280 that the oxygen concentration has been reduced to the first target value, feedback control is started in S290. That is, the process proceeds to the second stage described above. FIG. 8 shows an example of a feedback control system used after S290. The target value in FIG. 8 is the second target value, and the control object 91 shows characteristics from the fuel injection amount, the intake air amount, and the EGR amount to the oxygen concentration in the exhaust gas flowing into the DPF 6.

  In FIG. 8, the fuel injection amount, the intake air amount, and the EGR amount may be, for example, the command value of the fuel injection amount, the opening degree of the intake throttle 32, and the opening degree of the EGR valve 51, respectively. In FIG. 8, a PI (proportional integration) controller 90 is used as the controller. As is well known, the integration function of the PI controller 90 causes the oxygen concentration as the output of the control object 91 to converge to the (second) target value with high accuracy. Here, the PI controller has been described as an example. However, the present invention is not limited to this, and other known control methods such as modern control may be used.

  As shown in FIG. 7, by switching to feedback control, the oxygen concentration decreases toward the second target value. However, at that time, if the feedback gain in the second stage (for example, the gain of the PI controller 90 when the control system of FIG. 8 is used) is increased too much, the response causes an undershoot, and the oxygen concentration becomes the second target value. Although there is a possibility that the value is reduced to a value lower than T2, it is necessary to avoid the reduction from the limit concentration T0 as described above. Therefore, the feedback gain is not set so large in the feedback control system used in the second stage. As a result, the convergence to the second target value in the second stage is slower than the convergence in the first stage. However, although at a low speed, the oxygen concentration converges to the second target value T2 with high accuracy by the effect of feedback control (the integrator in the PI controller 90 when the control system of FIG. 8 is used).

  When it is detected in S310 that the oxygen concentration has been reduced to the second target value, the processing in FIG. 4 is terminated. As shown in FIG. 7, the oxygen concentration can be reduced only to T1 in the first stage without the learning of the prior art, while the present invention with learning greatly reduces the oxygen concentration to T1 ′ in the first stage. ing. Therefore, in the case of the present invention, the amount that must be adjusted by feedback (F / B) can be reduced.

  Therefore, since the gentle convergence by F / B can be made in a shorter time in the present invention, the oxygen concentration reaches the second target value T2 at time t1. On the other hand, in the prior art, the oxygen concentration finally reaches the second target value T2 at time t2. Therefore, the settling time from the start time t0 of the oxygen concentration reduction process until reaching the second target value is shorter in the present invention with learning than in the conventional technique without learning. That is, in the present invention, the oxygen concentration can be quickly reduced.

  As a result, as shown in the lower part of FIG. 7, in the present invention, an increase in the internal temperature of the DPF 6 is also suppressed, and an excessive temperature rise can be avoided. On the other hand, in the prior art, overheating occurs while it takes time to reduce the oxygen concentration.

  Next, a second embodiment of the present invention will be described. In the second embodiment, FIG. 2 in the first embodiment is changed to FIG. 5, and the rest is not changed. Only the changed part will be described below.

  In the processing procedure of FIG. 5, the procedure S30 in FIG. 2 is changed to the procedure S35, and the other steps are the same as those in FIG. In S35, it is determined whether or not the engine 2 is in an idle operation state. If it is in the idling operation state (S35: YES), the process proceeds to S40, and if it is not the idling operation state (S35: NO), the process returns to S10.

  That is, in FIG. 5, the deviation between the command value and the measured value of the oxygen concentration is learned only in the case of idle operation (further, the vehicle travels a predetermined distance from the previous learning and the PM accumulation amount is small). As a result, learning is performed in an idle operation state in which it is considered that the oxygen concentration is most likely to fluctuate due to a difference in the intake air amount or the injection amount because both the intake air amount and the fuel injection amount are small. Can be expected. In the second embodiment, the correction of the command value in S230 of FIG. 4 may be a fixed correction regardless of the operating state.

  Next, Example 3 will be described. In the third embodiment, FIG. 4 in the first embodiment is changed to FIG. Further, as described above, the oxygen concentration sensor 42 is equipped downstream of the DPF 6 and the oxygen concentration sensor 41 upstream of the DPF 6 is not equipped. The rest is the same as in Example 1. Only the changed part from Example 1 is demonstrated below.

  In the third embodiment, an oxygen concentration sensor 42 is installed downstream of the DPF 6. In general, since oxygen is consumed for the combustion reaction in the DPF 6, the oxygen concentration downstream of the DPF 6 is lower than the oxygen concentration in the exhaust gas flowing into the DPF 6. Therefore, even when the oxygen concentration downstream of the DPF 6 is sufficiently reduced and it is determined that there is no possibility of excessive temperature rise based only on the numerical value, the oxygen concentration in the exhaust gas flowing into the DPF 6 is high. In actuality, overheating may occur.

  Therefore, in the third embodiment, such a problem is dealt with by calculating the oxygen concentration upstream of the DPF 6 from the measured value of the oxygen concentration of the DPF 6. 6 is the same as FIG. 4 except that steps S250 to S270 and S300 are added to FIG.

  In S250, the output of the oxygen concentration sensor 42 downstream of the DPF 6 is read. Next, in S260, the oxygen consumption amount in the DPF 6 is calculated. As a method of calculating the oxygen consumption amount in the DPF 6, a known method of calculating from a mathematical model of oxygen consumption in the DPF 6 and various measured values may be used. In S270, the oxygen concentration in the exhaust upstream of the DPF 6 (inflow into the DPF 6) is calculated from the oxygen concentration downstream of the DPF 6 acquired in S250 and the oxygen consumption in the DPF 6 calculated in S260.

  In S300, the oxygen concentration upstream of the DPF 6 is also calculated in the second stage by the feedback control started in S290, as in S270 described above. When it is detected in S310 that the oxygen concentration upstream of the DPF 6 has reached the first target value, the processing in FIG. 6 is terminated.

  As described above, in Example 3, even when the oxygen concentration sensor 42 is installed downstream of the DPF 6, the oxygen concentration upstream of the DPF 6 is calculated from the measured value, and the oxygen concentration in the exhaust gas flowing into the DPF 6 is overheated. It can reduce rapidly to the 2nd target value which can be controlled. The above is the third embodiment. In the third embodiment, since the oxygen concentration sensor 42 is installed downstream of the DPF 6, even in the learning process of FIGS. 2 and 5, if the upstream value is calculated from the downstream oxygen concentration measurement value and the upstream deviation is obtained. Good.

  Note that all the measured values of the oxygen concentration sensor 41 or 42 may be numerical values corrected by the ECU 7 (correction means). The correction here refers to, for example, obtaining the error of the oxygen concentration sensor 41 or 42 by comparing the measured value of the oxygen concentration sensor 41 or 42 when the fuel injection is cut with the numerical value (known) of the oxygen concentration in the atmosphere. Thus, the correction may be made to cancel the error.

  In all of the above, the oxygen concentration sensors 41 and 42 may be changed to A / F sensors. In that case, the oxygen concentration may be replaced with the air / fuel ratio in the above discussion, or the air / fuel ratio measured by the A / F sensor may be converted into the oxygen concentration by the ECU 7. In the above embodiment, the fuel injection amount, the intake air amount, and the EGR amount are manipulated when the oxygen concentration is reduced (input amount). You may transform into an argument using at least one. Moreover, although the diesel engine was used as the internal combustion engine, this may not be a diesel engine, for example, a lean burn gasoline engine.

  The procedure from S110 to S140 in the above embodiment constitutes the detection means. The procedure of S240 constitutes a first reduction means. The procedure from S10 to S80 and S230 constitutes learning means and first learning means. The procedure of S220 constitutes setting means. The procedure of S290 constitutes the second reduction means. The procedure of S270 and S300 constitutes a calculation means.

1 Exhaust gas purification device 2 Diesel engine (internal combustion engine)
3 Intake pipe 4 Exhaust pipe (exhaust passage)
5 EGR pipe 6 Diesel particulate filter (DPF, filter)
7 Electronic control unit (ECU)
21 Injector 22 Engine Speed Sensor 31 Air Flow Meter 32 Intake Throttle 51 EGR Valve 41 Oxygen Concentration Sensor 42 Oxygen Concentration Sensor (Downstream Measuring Unit)
61, 62 Exhaust temperature sensor 63 Differential pressure sensor

Claims (9)

A filter disposed in the exhaust passage of the internal combustion engine for collecting particulate matter;
Regenerating means for regenerating the filter by burning particulate matter collected in the filter;
Detecting means for detecting that the filter is in an overheatable state in which the temperature of the filter may be excessively increased during regeneration by the regeneration means;
An input amount including a fuel injection amount and an intake air amount so that the oxygen concentration in the exhaust gas flowing into the filter is reduced to the target value for rapid reduction when the detection means detects that the temperature rise is possible. First reduction means for operating
By acquiring a deviation between the command value and the measured value of the oxygen concentration when the input amount is manipulated, the accuracy of convergence of the oxygen concentration to the target value for rapid reduction is improved. Learning means to correct the competence,
When the learning unit does not correct the input amount, the learning unit corrects the input amount compared to a target value for rapid reduction that is set to avoid excessive reduction of the oxygen concentration. Setting means for setting the target value for rapid reduction used in the first reduction means to be low when
An exhaust emission control device for an internal combustion engine, comprising:   2. The internal combustion engine according to claim 1, wherein the learning unit includes a first learning unit that operates the input amount and acquires the deviation in a state where the detection unit does not detect that the overheating is possible. Exhaust purification equipment.   The exhaust gas purification apparatus for an internal combustion engine according to claim 2, wherein the first learning means acquires the deviation for each operating condition of the internal combustion engine.   The exhaust gas purification apparatus for an internal combustion engine according to claim 2, wherein the first learning means acquires the deviation during idle operation of the internal combustion engine. The internal combustion engine is mounted on an automobile;
The exhaust gas purification apparatus for an internal combustion engine according to any one of claims 1 to 4, wherein the learning means updates the value of the deviation for each predetermined travel distance of the automobile or for each predetermined time.   Following the control of bringing the oxygen concentration in the exhaust gas flowing into the filter closer to the rapid reduction target value by the first reduction means, the oxygen concentration in the exhaust gas flowing into the filter is lower than the rapid reduction target value. The exhaust emission control device for an internal combustion engine according to any one of claims 1 to 5, further comprising second reduction means for executing control to bring the target value for avoiding excessive temperature rise closer.   The exhaust purification device for an internal combustion engine according to claim 6, wherein the second reduction means performs control to feed back a measured value of oxygen concentration in the exhaust gas flowing into the filter. Downstream measurement means for measuring the oxygen concentration in the exhaust passage downstream of the filter;
Calculating means for calculating oxygen concentration in the exhaust gas flowing into the filter from a measurement value by the downstream measuring means,
The exhaust emission control device for an internal combustion engine according to any one of claims 1 to 7, wherein the measured value of the oxygen concentration is replaced with an oxygen concentration value calculated by the calculating means.   The measured value of the oxygen concentration is a value corrected by a correction unit that corrects the oxygen concentration value by using a difference between the measured value of the oxygen concentration when the fuel injection is stopped and the oxygen concentration in the atmosphere. The exhaust gas purification apparatus for an internal combustion engine according to any one of the above.

Images