JP2005090460A - Engine control device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an engine control device for optimizing the purification of exhaust gas by accurately diagnosing a lean NOx catalyst with its NOx storing performance and its O<SB>2</SB>storing performance separated from each other and reflecting the diagnosis result on the control of the lean NOx catalyst. <P>SOLUTION: The engine control device has the lean NOx catalyst for storing NOx exhausted from an engine during lean-operation and desorbing and purifying the stored NOx during rich-operation and an O<SB>2</SB>sensor arranged on the downstream side of the lean NOx catalyst. It comprises a means for computing a stored NOx amount A1 of the lean NOx catalyst and a means for computing a stored O<SB>2</SB>amount B1 of the lean NOx catalyst. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a control device for an engine equipped with a lean NOx catalyst, and is particularly suitable for application to a lean burn engine, diagnosing the performance of the lean NOx catalyst with high accuracy and optimally controlling the performance of the lean NOx catalyst. The present invention relates to an engine control device.

  In recent years, a low fuel consumption type engine has been demanded in the field of automobiles against the background of global warming and energy problems. The lean burn engine is the best of them, and in particular, the direct injection engine injects fuel directly into the combustion chamber, stratifies the mixture, and enables combustion at an air-fuel ratio of 40 or more. It is effective in reducing fuel consumption compared to engines. On the other hand, against the backdrop of environmental problems such as air pollution, it is also required to improve the exhaust gas purification performance (reducing specific components contained in the exhaust gas, etc.). It is being strengthened.

  Conventionally used three-way catalysts cannot purify NOx exhausted from the engine (combustion chamber) during lean operation. Therefore, lean burn engines use exhaust gas to purify NOx during lean operation. It is common to provide a lean NOx catalyst in the system.

The lean NOx catalyst stores NOx during lean operation of the engine, that is, in an oxidizing atmosphere, and oxidizes and purifies HC and CO. Further, in order to purify the stored NOx, when the engine is in rich operation (hereinafter rich spike or rich spike control), that is, in a reducing atmosphere, the stored NOx is desorbed and the reducing agent (H ₂ , CO ₂ ) discharged from the combustion chamber. , HC) have a function of causing a redox reaction. Thus, while the lean NOx catalyst is effective in reducing NOx emissions, it is essential to diagnose the NOx storage capacity of the catalyst.

As an example of the prior art (refer to Patent Document 1) of an engine control device equipped with this type of lean NOx catalyst, an engine that binaryly detects rich or lean with respect to the stoichiometric air-fuel ratio downstream of the lean NOx catalyst. ₂ sensors are provided, the amount of NOx stored in the lean NOx catalyst is estimated based on the period from the start of rich spike until the O ₂ sensor reaches a predetermined value or more, and the amount of reducing agent at the time of rich spike is optimized Also, there is one that detects the deterioration state of the lean NOx catalyst based on the estimated NOx storage amount. That is, during the rich spike, the reducing agent of H ₂ , CO, and HC flowing in from the lean NOx catalyst inlet and the stored NOx undergo an oxidation / reduction reaction, so that the gas atmosphere downstream of the catalyst is reduced until the stored NOx is reduced. Is due to not having a reducing atmosphere.

Further, in other prior art (see Patent Document 2), provided with an O ₂ sensor for detecting the rich or lean in a binary manner with respect to the theoretical air-fuel ratio downstream of the lean NOx catalyst, the O ₂ sensor from a rich spike start The amount of NOx stored in the lean NOx catalyst is estimated based on the period until the value exceeds the specified value, and the storage performance of the lean NOx catalyst is detected, and the next lean operation time and rich spike amount (operation time) Is disclosed. This is also based on estimating the amount of stored NOx from the oxidation / reduction reaction period between the reducing agent flowing into the catalyst during rich spike and the stored NOx.

By the way, some lean NOx catalysts have not only NOx storage performance but also O ₂ storage performance. O ₂ storage performance is exhibited by precious metal and cerium (or ceria) which is a cocatalyst. In an oxidizing atmosphere, that is, lean, the oxygen flowing in from the catalyst inlet is stored, and in a reducing atmosphere, that is, when rich. , Has the function of releasing stored O ₂ . In the present catalyst, during the lean operation, not only NOx flowing into the catalyst but also O ₂ is stored. Also, during rich spike, not only the stored NOx but also the stored O ₂ is desorbed, so that an oxidizing / reducing reaction occurs with the reducing agent of H ₂ , CO, and HC that flows in. From this, at the time of rich spike, the delay time from when the gas atmosphere upstream of the catalyst becomes the reducing atmosphere until the gas atmosphere downstream of the catalyst becomes the reducing atmosphere is not only the amount of stored NOx but also the amount of stored O ₂ . Will be included.

  In general, the performance of a lean NOx catalyst is roughly classified into a lean performance and a rich performance. Lean performance is evaluated by the amount of NOx discharged downstream of the catalyst without being stored by the lean NOx catalyst during lean operation.From the viewpoint of compliance with exhaust regulations, the amount of unstored NOx during lean operation is more than a predetermined value. It is desirable to perform a rich spike when it becomes.

In addition, since the rich time performance is evaluated by the amount of stored NOx and the purification efficiency of HC, CO, NOx flowing from the catalyst inlet, the stored NOx that is an oxidant, the stored O _2, and the inflow H ₂ , CO, that is a reducing agent It is desirable to control the amount of reducing agent (or rich time) so that HC undergoes an oxidation / reduction reaction without excess or deficiency. Therefore, the lean performance depends only on the NOx storage performance, and the rich performance depends on both the NOx storage performance and the O ₂ storage performance. However, as described above, it contributes noble metals and cerium O ₂ storage capacity (or ceria), in order to cause coagulation by heat of the exhaust gas, O ₂ storage performance deteriorates with time.

On the other hand, noble metals (mainly Pt) and alkali or alkaline earth metals that contribute to NOx storage performance also aggregate. Since the materials that contribute to the O ₂ storage performance and the NOx storage performance are different as described above, the deterioration of the NOx storage performance and the O ₂ storage performance does not necessarily have a correlation and often proceeds independently. Therefore, in a lean NOx catalyst having O ₂ storage performance, if the NOx storage performance and O ₂ storage performance are not detected separately, controllability deteriorates or misdiagnosis occurs.

From this, when the lean NOx catalyst has O ₂ storage performance, the conventional technology (Patent Document 1 and Patent Document 2) detects not only the NOx storage amount but also the O ₂ storage amount. Even if NOx storage performance does not deteriorate and only O ₂ storage performance deteriorates, it is judged that the NOx storage performance has deteriorated, so the next lean operation time is unduly shortened, or the lean NOx catalyst There is a problem that the result of misdiagnosis that the deterioration has occurred is notified to the outside.

On the other hand, other prior arts (see Patent Documents 3 and 4) disclose that the rich spike amount is determined in consideration of the O ₂ storage amount in the lean NOx catalyst. The prior art does not include an exhaust sensor such as an O ₂ sensor downstream of the lean NOx catalyst, calculates the O ₂ storage amount based on a predetermined characteristic, and not only the NOx storage amount but also the O ₂ storage amount during a rich spike. This also optimizes the amount of rich spike. However, in the method of calculating the O ₂ storage amount based on the predetermined characteristics, as described above, the O ₂ storage performance of the lean NOx catalyst changes with time, so that diagnosis and optimization control of the catalyst are realized. I can't.

Furthermore, in another prior art (see Patent Document 5), an O ₂ sensor is provided downstream of the lean NOx catalyst, and when the downstream O ₂ sensor output at the time of rich spike reverses from lean to rich, the first half of the delay time the O ₂ desorption period as the second half NOx desorption period, it is proposed to separate and detect the O ₂ storage amount and NOx storage amount.

However, the desorption of stored O _{2 and} the desorption of stored NOx of the lean NOx catalyst do not always occur in order, and it is known that desorption of NOx occurs early in the rich spike. Further, the behavior of the downstream O ₂ sensor at the time of rich spike is not necessarily a slow phenomenon that can be separated into the first half period and the second half period. It is difficult to accurately separate and detect the amount of O _{2 and the} amount of stored NOx. In this proposal, no consideration is given to the optimization control of the catalyst based on the detection result.

The present invention was made to solve the problem of as described above, it is an object to diagnose accurately separates the NOx storage capacity and O ₂ storage capacity of the lean NOx catalyst, the It is an object of the present invention to provide an engine control device that optimizes the purification of exhaust gas by reflecting the diagnosis result in the control of the lean NOx catalyst.
JP 2001-271679 A JP 2000-337130 A JP 2002-70611 A JP 2003-56388 A JP 2000-352309 A

In order to achieve the above object, the engine control device of the present invention basically stores NOx discharged from the engine during lean operation, and desorbs and purifies the stored NOx during rich operation. a catalyst, and O ₂ sensor disposed downstream of the lean NOx catalyst, and means for calculating a stored NOx amount A1 of the lean NOx catalyst, means for calculating a storage amount of O ₂ B1 of the lean NOx catalyst (See FIG. 1).

According to a specific aspect of the engine control apparatus of the present invention, the means for calculating the lean NOx catalyst storage O ₂ amount B1 is such that the air-fuel ratio of the engine becomes lean at the end of the rich operation of the lean NOx catalyst. Then, based on the period Tl until the output of the O ₂ sensor becomes a predetermined value or less or the fuel injection amount Ml during the period Tl, the stored O ₂ amount B1 is calculated, and the stored NOx amount of the lean NOx catalyst The means for calculating A1 is that the output of the O ₂ sensor is greater than or equal to a predetermined value after the air-fuel ratio of the engine becomes rich during the rich operation performed to desorb and purify the stored NOx of the lean NOx catalyst. The amount of stored NOx is calculated based on the difference between the period Tr and the period Tl or the difference between the fuel injection amount Mr and the fuel injection amount Ml. It is characterized by calculating A1 That (see Figure 2).

The lean NOx catalyst with O ₂ storage performance desorbs not only the stored NOx but also the stored O ₂ during the rich spike, so that the reducing agent of H ₂ , CO and HC that flows in and the oxidation / reduction reaction Occur. From this, at the time of rich spike, the delay time from when the gas atmosphere upstream of the lean NOx catalyst becomes the reducing atmosphere until the gas atmosphere downstream of the lean NOx catalyst becomes the reducing atmosphere is not only the amount of stored NOx The amount of stored O ₂ will also be included. During the rich operation performed to desorb and purify the stored NOx of the lean NOx catalyst, the engine air-fuel ratio becomes rich until the output of the O ₂ sensor downstream of the lean NOx catalyst reaches a predetermined value or more. The period Tr or the fuel injection amount Mr during the period Tr includes information on both the stored O ₂ amount and the stored NOx amount. On the other hand, after the end of the rich spike, the delay time from when the gas atmosphere upstream of the catalyst becomes an oxidizing atmosphere until the gas atmosphere downstream of the catalyst becomes an oxidizing atmosphere depends only on the amount of stored O ₂ . This is because the O ₂ concentration contained in the exhaust gas during lean operation is about 20% (20,000 ppm), and the NOx concentration is several hundred ppm, which is overwhelmingly higher. This is because when the stored O ₂ in the catalyst is saturated and O ₂ flows out downstream of the catalyst, the downstream O ₂ sensor reacts to it and performs lean inversion. Therefore, at the end of the rich operation, the fuel injection amount Ml during the period Tl or the period Tl from when the air-fuel ratio of the engine becomes lean until the output of the O ₂ sensor downstream of the lean NOx catalyst becomes equal to or less than a predetermined value is Since the information is only the stored O ₂ amount, the stored NOx amount can be calculated based on the difference between the period Tr and the period Tl or the difference between the fuel injection amount Mr and the fuel injection amount Ml.

According to another specific aspect of the engine control apparatus of the present invention, the means for calculating the lean NOx catalyst stored O ₂ amount B1 is the air-fuel ratio of the engine at the end of the rich operation of the lean NOx catalyst. And the stored O ₂ amount B1 is calculated based on the period Tl2 until the output of the O ₂ sensor falls below a predetermined value or the fuel injection amount Ml2 during the period Tl2, and the stored NOx of the lean NOx catalyst The means for calculating the amount A1 is such that the output of the O ₂ sensor becomes a predetermined value after the air-fuel ratio of the engine becomes rich during the rich operation performed to desorb and purify the stored NOx of the lean NOx catalyst. Calculate the period Tr until the above or the fuel injection amount Mr during the period Tr, based on the difference between the period Tr and the period Tl2 or the difference between the fuel injection amount Mr and the fuel injection amount Ml2 Specially for calculating quantity A1 Are (see Figure 3).

After the rich spike ends, the delay time from when the gas atmosphere upstream of the catalyst becomes an oxidizing atmosphere until the gas atmosphere downstream of the catalyst becomes the oxidizing atmosphere depends only on the amount of stored O ₂ . However, as described above, the O ₂ concentration during the lean operation is about 20,000 ppm, and the stored O ₂ in the lean NOx catalyst is saturated in a short time, and the delay time is short. For this reason, the sensitivity of the delay time generated at the start of the lean operation with respect to the stored O ₂ amount may be poor, and the stored O ₂ amount detection accuracy may deteriorate. Therefore, in order to increase the sensitivity of the delay time that occurs at the start of lean operation with respect to the amount of stored O ₂ , after the rich spike ends, lean operation is not performed immediately, and stoichiometric operation where the O ₂ concentration is 1% or less is performed. and the catalyst downstream O ₂ sensor at that time is for detecting a storage amount of O ₂ based on the fuel injection amount Ml2 in delay time Tl2 or Tl2 until a predetermined value or less.

Furthermore, in another specific aspect of the engine control apparatus of the present invention, the means for calculating the lean Ox catalyst storage O ₂ amount B1 is configured such that the air-fuel ratio of the engine is stoichiometric or the NOx storage performance of the lean NOx catalyst. After a predetermined time of operation at a lean air / fuel ratio that does not exert the desired value, the air / fuel ratio of the engine is made rich and during the period Tr2 or until the output of the O ₂ sensor downstream of the lean NOx catalyst becomes equal to or higher than a predetermined value The stored O ₂ amount B1 is calculated based on the fuel injection amount Mr2 (see FIG. 4).

If NOx is not stored in the lean NOx catalyst, the delay time from when the gas atmosphere upstream of the lean NOx catalyst becomes the reducing atmosphere until the gas atmosphere downstream of the lean NOx catalyst becomes the reducing atmosphere is stored. It depends only on the period during which the amount of O ₂ is desorbed. Therefore, after the engine is operated for a predetermined time at a lean air-fuel ratio that does not exhibit the NOx storage performance of the stoichiometric or lean NOx catalyst so that NOx is not stored in the lean NOx catalyst, the engine air-fuel ratio is made rich and the lean The stored O ₂ amount B1 is calculated based on the period Tr2 until the output of the O ₂ sensor downstream of the NOx catalyst reaches a predetermined value or more, or the fuel injection amount Mr2 during the period Tr2.

Furthermore, another specific aspect of the engine control apparatus of the present invention is an engine control apparatus including an O ₂ sensor or an A / F sensor upstream of the lean NOx catalyst, the O ₂ sensor or A frequency analysis is performed on the output signal of the A / F sensor to calculate a power Pa1 and a phase Ph1 of a predetermined frequency, and an output signal of an O ₂ sensor downstream of the lean NOx catalyst is frequency-analyzed to obtain a power Pa2 of a predetermined frequency. And a means for calculating the phase Ph2, and calculating the stored O ₂ amount B1 based on the relationship between the power Pa1 and the power Pa2 and the phase Ph1 and the phase Ph2 (see FIG. 5). ).

In the air-fuel ratio region richer than the lean air-fuel ratio at which the NOx storage performance of the lean NOx catalyst is not exhibited, when the air-fuel ratio upstream of the catalyst changes, the response delay time of the O ₂ sensor until the catalyst downstream is the O ₂ storage Or it depends only on the phenomenon of desorption. Therefore, an O ₂ sensor or A / F sensor in the lean NOx catalyst upstream detects the air-fuel ratio variation of the catalyst upstream, by comparing the response delay until the O ₂ sensor in the lean NOx catalyst downstream of the catalyst O ₂ Storage performance can be estimated. In addition, in order to obtain a frequency response, in this aspect, detection is performed separately for power and phase.
Furthermore, in another specific aspect of the engine control apparatus of the present invention, the start time of rich spike control performed for purifying NOx stored in the lean NOx catalyst is calculated based on the stored NOx amount A1. Means is provided (see FIG. 6).

  In terms of lean performance, from the viewpoint of complying with exhaust regulations, it is desirable to perform rich spike when the amount of unstored NOx or stored NOx during lean operation exceeds a predetermined value. Accordingly, since the rich spike start time depends only on the NOx storage performance of the lean NOx catalyst, the next rich spike start time is determined based only on the stored NOx amount A1.

Furthermore, in another specific aspect of the engine control device of the present invention, the rich spike amount or rich spike period at the time of rich spike control performed for purifying NOx stored in the lean NOx catalyst is stored in the storage. further comprising means for calculating on the basis that the NOx amount A1 in the storage amount of O ₂ B1 is characterized (see Figure 7).

The rich time performance is evaluated by the amount of stored NOx and the purification efficiency of HC, CO, and NOx flowing from the catalyst inlet. Therefore, the stored NOx that is an oxidant, the stored O _2, and the inflow H ₂ , CO, and HC that are reducing agents are It is desirable to control the amount of reducing agent (or rich time) so that the oxidation / reduction reaction can be performed without excess or deficiency. Accordingly, since the rich time performance depends on both the NOx storage performance and the O ₂ storage performance, the next rich spike amount or rich spike period is determined based on the stored NOx amount A1 and the stored O ₂ amount B1. .

Furthermore, another specific aspect of the engine control apparatus of the present invention is a lean operation performed after the rich spike based on the stored O ₂ amount B1 obtained during the rich spike or near the stoichiometric operation. A means for estimating the amount of stored O ₂ B2 stored in the lean NOx catalyst during a predetermined period of time, and a lean operation performed after the rich spike based on the stored NOx amount A1 obtained during the rich spike. Means for estimating the stored NOx amount A2 stored in the lean NOx catalyst during a predetermined period, and estimating the catalyst inlet NOx amount C flowing into the lean NOx catalyst during the predetermined period of the lean operation performed after the rich spike Means for calculating an unstored NOx amount D discharged downstream of the lean NOx catalyst without being stored from the catalyst inlet NOx amount C and the stored NOx amount A2, and the unstored Based on the NOx amount D, comprising means for calculating a start timing of the rich spike control, and means for calculating the rich spike amount or rich spike period based on the storage NOx amount A2 and the storage amount of O ₂ B2 (See FIG. 8).

The storage NOx amount A1 and the storage O ₂ amount B1 of the lean NOx catalyst are obtained at the time of rich spike or near the stoichiometric. Next, the NOx storage performance (or performance change) of the lean NOx catalyst is obtained based on the stored NOx amount A1, and based on this, the NOx amount A2 stored in the catalyst during the next lean operation is calculated. Further, the O ₂ storage performance (or performance change) of the lean NOx catalyst is obtained based on the stored O ₂ amount B1, and based on this, the O ₂ amount B2 stored in the catalyst during the next lean operation is calculated. Further, the unstored NOx amount D is obtained from the difference between the catalyst inlet NOx amount C and the stored NOx amount A2 during the lean operation, and the rich spike start time is determined based on the unstored NOx amount D. Further, based on the storage amount of O ₂ B2 and the storage NOx amount A2, it is intended to determine the next rich spike amount.

In other words, it calculates the NOx amount D to be discharged downstream of the catalyst without being a NOx amount A2 is the storage amount of O ₂ B2 and storage reservoir in the catalyst during the lean operation, based on unstored NOx amount D, the rich spike start the decision of when, based on the storage amount of O ₂ B2 and the storage NOx amount A2, determines the rich spike amount.

The NOx amount and O ₂ amount actually stored at the time of the rich spike are detected, and a change in the performance of the catalyst is detected. When performance change is observed, on the basis of the performance change allowance to adjust the next NOx amount A2 and O ₂ amount B2 is stored in the catalyst during the lean operation, the next rich spike start timing and rich spike amount Is to optimize. In this way, it is possible to appropriately cope with changes in the characteristics of the catalyst.

  Furthermore, in another specific aspect of the engine control apparatus of the present invention, means for estimating the catalyst inlet NOx amount C flowing into the lean NOx catalyst during a predetermined period of the lean operation performed after the rich spike is provided. The catalyst inlet NOx amount C is estimated by multiplying both the NOx emission basic value calculated based on the engine torque and the engine speed and the NOx emission correction value calculated based on the EGR rate and the engine air-fuel ratio. (See FIG. 9).

  Furthermore, in another specific aspect of the engine control apparatus of the present invention, the means for estimating the stored NOx amount A2 is based on the previously calculated estimated stored NOx amount A2, and the lean NOx catalyst per unit time. Means for calculating a storage rate Rn of NOx stored therein, and means for correcting the relationship between the estimated storage NOx amount A2 and the storage rate Rn based on the storage NOx amount A1, the lean NOx catalyst Multiplying the catalyst inlet NOx amount C flowing into the catalyst by the NOx storage rate Rn, the stored NOx amount per unit time is calculated, the stored NOx amount per unit time is integrated, and the estimated stored NOx amount A2 is calculated. (See FIG. 10).

  The amount of NOx stored per unit time (or unit time) is obtained by multiplying the catalyst inlet NOx amount C per unit period during lean operation by the NOx purification (storage) rate. Note that the stored NOx amount represents the NOx amount stored for a certain period T [s], and the unit is [g / T], and may be expressed as a storage rate. The total amount of NOx stored in the lean NOx catalyst is obtained by integrating the amount of NOx stored for a certain period T [s]. The NOx purification (storage) rate is determined based on the total NOx storage amount A2. For example, it represents the characteristic that the purification (storage) rate decreases as the total NOx storage amount A2 increases.

  Further, as described in claim 8 (see FIG. 8), when the NOx storage performance (or performance change) of the lean NOx catalyst is recognized in the stored NOx amount A1 of the lean NOx catalyst obtained during the rich spike, the lean Tune the parameters of the NOx purification (storage) rate calculation means of the NOx catalyst model. The relationship between total NOx storage and NOx purification (storage) rate will be changed according to performance degradation. For example, this indicates that the NOx purification (storage) rate corresponding to the total NOx storage amount is further decreased to deteriorate the storage performance.

  Furthermore, another specific aspect of the engine control apparatus of the present invention is characterized in that the rich spike control start timing calculation means starts the rich spike control when the unstored NOx amount D is a predetermined value or more. (See FIG. 11).

  From the viewpoint of conforming to exhaust regulations, it is desirable to perform a rich spike when the unstored NOx amount D during lean operation exceeds a predetermined value.

Furthermore, another specific aspect of the engine control apparatus according to the present invention is the engine control apparatus including a three-way catalyst upstream of the lean NOx catalyst, wherein the three-way catalyst has a stored O ₂ amount. Means for calculating B2, and the rich spike amount or rich spike period at the time of rich spike control performed to purify NOx stored in the lean NOx catalyst, the stored NOx amount A1, the stored O ₂ amount B1, and the storage And a means for calculating based on the O ₂ amount B2 (see FIG. 12).

A three-way catalyst generally has O ₂ storage performance. Rich spike amount from this (or rich spike time), storage O ₂ storage NOx and storage O ₂ and three-way within the catalyst in the lean NOx catalyst, H _2, CO in the exhaust gas as a reducing agent, and HC It is desirable to control the oxidation / reduction reaction without excess or deficiency.

Furthermore, another specific aspect of the engine control device of the present invention is the engine control device comprising another O ₂ sensor between the three-way catalyst and the lean NOx catalyst, The means for calculating the stored catalyst O ₂ amount B2 is the period Tr3 or the period Tr3 from when the air-fuel ratio of the engine becomes rich until the output of the other O ₂ sensor becomes a predetermined value or more. Based on the fuel injection amount Mr3, the means for calculating the stored O ₂ amount B2 in the three-way catalyst and calculating the stored O ₂ amount B1 of the lean NOx catalyst is as follows: from an output of the other of the O ₂ sensor becomes below a predetermined value, the lean NOx period until the output of the downstream O ₂ sensor of the catalyst is equal to or less than a predetermined value Tl3 or said period between the fuel injection amount Ml3 in Tl3 based on, it calculates the storage amount of _{O 2} B1, the Li Means for calculating a stored NOx amount A1 of emissions NOx catalyst, the storage NOx of the lean NOx catalyst during rich operation performed to desorption and purification, the output of the other of the O ₂ sensor becomes a predetermined value or more From this, the period Tr4 until the output of the O ₂ sensor downstream of the lean NOx catalyst reaches a predetermined value or the fuel injection amount Mr4 during the period Tr4 is calculated, and the difference between the period Tr4 and the fuel injection amount Tl3 or The stored NOx amount A1 is calculated based on the difference between the period Mr4 and the fuel injection amount Ml3.

During rich spike, the stored O ₂ in the three-way catalyst is also desorbed, so the gas atmosphere upstream of the three-way catalyst becomes the reducing atmosphere and the gas atmosphere downstream of the three-way catalyst becomes the reducing atmosphere. The delay time depends on the amount of stored O ₂ . Therefore, as described in claim 12 (refer to FIG. 12), the engine is empty during the rich operation that is performed in order to desorb and purify the stored NOx of the lean NOx catalyst by providing the O ₂ sensor downstream of the three-way catalyst. Based on the period Tr3 or the fuel injection amount Mr3 during the period Tr3 from when the fuel ratio becomes rich until the output of the O ₂ sensor downstream of the three-way catalyst reaches a predetermined value or more, the stored O ₂ in the three-way catalyst. The amount B2 is calculated.

Further, the period Tr4 until the output of the O ₂ sensor 2 downstream of the lean NOx catalyst becomes equal to or higher than the predetermined value after the output of the O ₂ sensor 1 becomes equal to or higher than the predetermined value is the lean fuel injection amount Mr4 during the period Tr4. Information on both the amount of stored O ₂ in the NOx catalyst and the amount of stored NOx is included. On the other hand, after the rich spike control is completed, when the engine performs the lean operation again, after the O ₂ sensor downstream of the three-way catalyst becomes lean, the output of the O ₂ sensor downstream of the lean NOx catalyst becomes less than a predetermined value. The fuel injection amount Ml3 during the period Tl3 or during the period Tl3 is information on only the stored O ₂ amount B1 in the lean NOx catalyst. Therefore, the stored NOx amount A1 in the lean NOx catalyst can be calculated based on the difference between the period Tr4 and the period Tl3 or the difference between the fuel injection amount Mr4 and the fuel injection amount Ml3.

  Furthermore, another aspect of the engine control apparatus of the present invention is characterized in that the engine control apparatus is mounted on an automobile.

According to the control device of the present invention, the NOx storage performance and the O ₂ storage performance of the lean NOx catalyst can be separated and accurately diagnosed, and furthermore, the diagnosis result can be reflected in the control of the lean NOx catalyst. It is possible to optimally control the engine, and it is possible to optimize the exhaust purification performance and fuel consumption performance of the engine.

Embodiment 1
FIG. 14 shows the overall system configuration of an engine to which the engine control apparatus of Embodiment 1 of the present invention is applied.
The engine 100 is configured by a multi-cylinder in-cylinder injection engine, and an air cleaner 1 and an intake manifold 4 are connected to an intake system. Air from the outside passes through the air cleaner 1 and flows into the cylinder 9 through the intake manifold 4 and the collector 5. The inflow air amount is adjusted by the electronic throttle 3, and when idling, the engine speed is controlled by adjusting the air amount by an ISC valve provided in the bypass air passage.

In the engine 100, a fuel injection valve 7 and a spark plug 8 are attached to a cylinder 9 for each cylinder.
An exhaust manifold 10 and a lean NOx catalyst 11 are connected to the exhaust system of the engine 100. The exhaust gas of the engine 100 is sent to the lean NOx catalyst 11 through the exhaust manifold 10. The exhaust gas is purified by the lean NOx catalyst 11 and discharged into the atmosphere.

The airflow sensor 2 provided in the intake manifold 4 detects the amount of inflow air, and the crank angle sensor 15 disposed in the vicinity of the crankshaft 9c of the engine 100 outputs a signal every rotation angle of the crankshaft 9c. The accelerator opening sensor 13 provided in the accelerator 6 detects the amount of depression of the accelerator 6, and thereby detects the driver's required torque. The water temperature sensor 14 detects the coolant temperature of the engine. The exhaust manifold 10 is provided with an A / F sensor 12 and an O ₂ sensor 20 to detect the air-fuel ratio.

  The signals of the accelerator opening sensor 13, the airflow sensor 2, the throttle opening sensor 17 attached to the electronic throttle 3, the crank angle sensor 15, and the water temperature sensor 14 are sent to the control unit 16 and are output from these sensor outputs. The operating state of the engine 100 is obtained, and main operation amounts of the engine 100 such as the intake air amount, fuel injection amount, ignition timing, etc. of the engine 100 are optimally calculated. The fuel injection amount calculated in the control unit 16 is converted into a valve opening pulse signal and sent to the fuel injection valve 7 attached in the cylinder 9.

  Further, a drive signal is sent to the spark plug 8 so that ignition is performed based on the ignition timing calculated by the control unit 16. The injected fuel is mixed with air from the intake manifold and flows into the combustion chamber 9a of the cylinder 9 of the engine 100 to form an air-fuel mixture.

  The air-fuel mixture explodes due to sparks generated from the spark plug 8 at a predetermined ignition timing, and the piston 9b is pushed down by the combustion pressure to become power for the engine 100. The exhaust after the explosion is sent to the lean NOx catalyst 11 through the exhaust manifold 10.

  The lean NOx catalyst 11 has a function of desorbing and purifying the stored NOx by storing NOx discharged from the engine 100 and performing rich operation during lean operation.

  The A / F sensor 12 is attached between the cylinder 9 of the engine 100 and the lean NOx catalyst 11, and is configured to have a linear output characteristic with respect to the oxygen concentration contained in the exhaust gas. Since the relationship between the oxygen concentration in the exhaust gas and the air-fuel ratio is substantially linear, the air-fuel ratio can be obtained by the A / F sensor 12 that detects the oxygen concentration. The control unit 16 calculates the air-fuel ratio upstream of the lean NOx catalyst 11 from the signal of the A / F sensor 12, and sequentially corrects the fuel injection amount or the air amount so that the air-fuel ratio of the air-fuel mixture in the cylinder 9 becomes the target air-fuel ratio. Performs F / B control.

The O ₂ sensor 20 detects almost binary whether the gas atmosphere downstream of the lean NOx catalyst 11 is rich or lean with respect to stoichiometry. The control unit 16 calculates the O ₂ concentration downstream of the lean NOx catalyst 11 from the signal of the O ₂ sensor 20, detects the performance of the lean NOx catalyst, and optimally controls the lean NOx catalyst in terms of both fuel efficiency and exhaust.

FIG. 15 shows the internal configuration of the control unit 16. In the ECU 16, sensor output values of the A / F sensor 12, the throttle valve opening sensor 17, the airflow sensor 2, the engine speed sensor 15, the water temperature sensor 14, the accelerator opening sensor 13, and the O ₂ sensor 20 are input. After the signal processing such as noise removal is performed by the input circuit 24, the signal is sent to the input / output port 25.

  The value of the input port is stored in the RAM 23 and processed in the CPU 21. A control program describing the contents of the arithmetic processing is written in the ROM 22 in advance. A value representing each actuator operation amount calculated according to the control program is stored in the RAM 23 and then sent to the output port 25.

  The operation signal of the spark plug 8 is set to an ON / OFF signal that is ON when the primary coil in the ignition output circuit 26 is energized and is OFF when the primary coil is not energized. Ignition timing is when it turns from ON to OFF. The signal for the spark plug 8 set in the input / output port 25 is amplified to a sufficient energy necessary for combustion by the ignition output circuit 26 and supplied to the spark plug 8.

  The drive signal for the fuel injection valve 7 is set to an ON / OFF signal that is ON when the valve is open and OFF when the valve is closed. The fuel injection valve drive circuit 27 amplifies the drive signal to an energy sufficient to open the fuel injection valve 7. To the fuel injection valve 7. A drive signal for realizing the target opening degree of the electronic throttle 3 is sent to the electronic throttle 3 via the electronic throttle drive circuit 28.

Hereinafter, a control program written in the ROM 22 and executed by the CPU 21 will be described.
FIG. 16 is a block diagram showing the overall control of the first embodiment, and is the main part of the air-preceding type torque base control. The control of the present embodiment includes a target torque calculator 31, a target air amount calculator 32, a target throttle opening calculator 33, an ETC (electronic throttle) controller 34, a target air-fuel ratio calculator 35, and a fuel injection amount calculator 36. The actual air amount calculating unit 37 and the lean NOx catalyst diagnosing unit 38 are included.

  In the control, the target torque is first calculated from the accelerator opening by the target torque calculating unit 31, and then the target air amount is calculated from the target torque and the target air-fuel ratio by the target air amount calculating unit 32, and the target throttle opening calculating unit is calculated. The target throttle opening for realizing the target air amount is calculated at 33, and the throttle opening is F / B controlled by the ETC control unit 34 based on the output of the opening sensor 17. The fuel injection amount is calculated from the actual air amount detected by the air flow sensor 2 and calculated by the actual air amount calculation unit 37 and the target air-fuel ratio calculated by the target air-fuel ratio calculation unit 35.

  The present embodiment is characterized by the portions related to the target air-fuel ratio calculation unit (rich spike control unit) 35 and the lean NOx catalyst diagnosis unit 38, and other calculation unit specifications are already known and have many priorities. Since technology exists, details are omitted in this embodiment. Hereinafter, details of the target air-fuel ratio calculation unit (rich spike control unit) 35 and the lean NOx catalyst diagnosis unit 38 will be described.

  FIG. 17 is a block diagram showing a target air-fuel ratio calculation unit (rich spike control unit) 35. When the target air-fuel ratio calculation unit (rich spike control unit) 35 reaches a predetermined value or more, a lean flag is set. Is set to 0, and the target air-fuel ratio is set to the rich target air-fuel ratio in order to perform rich operation. When the rich time exceeds a predetermined value, the lean flag is set to 1, and the target air-fuel ratio is set to the lean target air-fuel ratio to perform the lean operation again.

FIG. 18 is a block diagram showing the lean NOx catalyst diagnostic unit 38. The lean NOx catalyst diagnosing unit 38 and the storage NOx amount A1 calculating means 39 and the storage based on the output of the O ₂ sensor 20, the fuel injection amount and the lean flag of the target air-fuel ratio calculating unit (rich spike control unit) 35. With the O ₂ amount B1 calculating means 40, the stored NOx amount A1 and the stored O ₂ amount B1 in the lean NOx catalyst 11 are obtained.

Specifically, the storage NOx amount A1 are those multiplied by Ga in Mr-Ml, storage amount of O ₂ B1 are those multiplied by Gb to Ml. Here, Ga and Gb are coefficients indicating the relationship between the amount of stored NOx and the amount of fuel required to reduce it, the amount of stored O ₂ and the amount of fuel required to consume it, respectively. It is better to decide.

A method for calculating the fuel injection amounts Mr and Ml will be described below.
FIG. 19 is a block diagram showing the Mr calculation unit 41 of the stored NOx amount A1 calculation means 39. It is determined whether the rich operation is being performed (lean flag is 0), and during the rich operation, that is, after the air-fuel ratio of the engine 100 becomes rich, the output of the O ₂ sensor 20 downstream of the lean NOx catalyst 11 becomes rich. A value obtained by multiplying the accumulated amount Mr_0 of the fuel injection amount during the period Tr by a predetermined value GMr is defined as Mr. This period Tr corresponds to a period in which the stored NOx in the catalyst 11 and the stored O ₂ oxidize the reducing agent (H ₂ , CO, HC) flowing from the inlet of the catalyst 11, and thus this period is the amount of stored NOx and storage This is based on the fact that there is a correlation in the amount of O ₂ . A lean flag shown in FIG. 17 is used to determine whether the rich operation is being performed. GMr is an adjustment gain, and may be 1 in general.

FIG. 20 is a block diagram showing the Ml calculation unit 42 of the stored O ₂ amount B1 calculation means 40. It is determined whether the lean operation is being performed. During the lean operation, that is, after the air-fuel ratio of the engine 100 becomes lean, the fuel injection in the period Tl during which the output of the O ₂ sensor 20 downstream of the lean NOx catalyst 11 becomes lean A value obtained by multiplying the integrated quantity Ml_0 by a predetermined value GMl is defined as Ml. This period Tl corresponds to a period during which O ₂ flowing from the catalyst inlet is stored in the catalyst 11, and therefore this period utilizes the fact that the amount of stored O ₂ is correlated. For example, a lean flag shown in FIG. 17 is used to determine whether the rich operation is being performed. GMl is an adjustment gain, and may be 1 in general.

[Embodiment 2]
The second embodiment is different from the first embodiment in the configurations of the target air-fuel ratio 35 and the lean NOx catalyst diagnostic unit 38, but the other configurations are the same.

FIG. 21 is a block diagram showing the target air-fuel ratio calculation unit 35 of the second embodiment. A stoichiometric target air-fuel ratio is added to the target air-fuel ratio 35 of the first embodiment in FIG. That is, when the lean time exceeds a predetermined value, the lean flag is set to 0, and the target air-fuel ratio is set to the rich target air-fuel ratio to perform rich operation. When the rich time exceeds a predetermined value, the stoichiometric flag is set to 1, and the target air-fuel ratio is stoichiometric. Further, when the rich + stoichiometric time exceeds a predetermined value, the lean flag is set to 1, and the target air-fuel ratio is set to the lean target air-fuel ratio to perform the lean operation again. In this way, even if the rich operation is finished, the lean operation is not performed immediately, but the stoichiometric operation is performed for a predetermined time, and then the lean operation is returned. FIG. 22 is a block diagram showing the lean NOx catalyst diagnostic unit 38 of the second embodiment. Compared to the lean NOx catalyst diagnostic unit 38 of Embodiment 1 in FIG. 18, the only difference is that Ml is Ml2, and the rest is the same. In the storage NOx amount A1 calculating means 39 and the storage amount of _{O 2} B1 calculation unit 40, and requests the storage NOx amount A1 and the storage amount of _{O 2} B1 in the lean NOx catalyst 11.

FIG. 23 is a block diagram illustrating the Ml2 operation unit 42 according to the second embodiment. During the stoichiometric operation, that is, after the air-fuel ratio of the engine 100 becomes stoichiometric, the period Tl2 during which the output of the O ₂ sensor 20 downstream of the lean NOx catalyst 11 is equal to or greater than the predetermined value K3. A value obtained by multiplying the accumulated amount Ml2_0 of the fuel injection amount by a predetermined value GMl2 is defined as Ml2. This period Tl2 corresponds to a period in which a predetermined amount of O ₂ flowing from the catalyst inlet is stored in the catalyst 11, and therefore this period utilizes the fact that the amount of stored O ₂ is correlated. Compared with the first embodiment, since the air-fuel ratio is stoichiometric, the amount of O ₂ flowing from the inlet of the catalyst 11 is small, so the period TI2 is longer than the period TI in the first embodiment, and the S / N ratio is improved. A stoichiometric flag shown in FIG. 21 is used to determine whether the stoichiometric operation is being performed. GMl2 is an adjustment gain, and may normally be 1.

[Embodiment 3]
FIG. 24 is a block diagram showing the overall control of the engine control apparatus according to the third embodiment, which is the main part of the air-preceding torque base control. The third embodiment is different from the first embodiment in that the A / F sensor 12 is provided upstream of the lean NOx catalyst 11 of the exhaust manifold 10, and the output signal of the A / F sensor 12 is input to the lean NOx catalyst diagnostic unit 38. The configuration is different, but the other configurations are the same.

  The target air-fuel ratio calculation unit 35 is the same as that in FIG. 17 of the first embodiment, and the lean NOx catalyst diagnosis unit 38 is the same as that in FIG.

FIG. 25 is a block diagram showing the Ml2 calculation unit 42. The Ml2 calculating unit 42 determines whether the stoichiometric operation is being performed (stoichiometric flag = 1) and the output of the O ₂ sensor 20 is equal to or less than a predetermined value. When both conditions are satisfied, the output spectrum of the A / F sensor 12 and the O ₂ sensor 20 is calculated by using FFT (Fast Fourier Transform) to calculate power spectra Power1 and Power2 and phase spectra Phase1 and Phase2 of a predetermined frequency. . Furthermore, the difference between Phase 2 and Phase 1 and the ratio of Power 2 and Power 1 are converted into predetermined values, and both are multiplied by Ml2. That is, in the vicinity of the stoichiometry where only the O ₂ storage capacity of the lean NOx catalyst is exhibited, the transfer characteristics of the catalyst inlet air-fuel ratio and the outlet air-fuel ratio are obtained by frequency analysis, and the basic value of the stored O ₂ amount is obtained with the transfer characteristics. It is assumed that Ml2.

[Embodiment 4]
FIG. 26 is a block diagram showing the overall control of the engine control apparatus according to the fourth embodiment, which is the main part of the air-preceding torque base control.
Embodiment 3, the embodiment 1, differs configuration NOx storage amount A1, O ₂ storage amount B1 calculated by the lean NOx catalyst diagnostic unit 38 is input is output to the target air-fuel ratio calculating section 35, Other than that, the configuration is the same.

FIG. 27 is a block diagram showing the target air-fuel ratio calculation unit (rich spike control unit) 35. From the catalyst inlet NOx amount C, the stored NOx amount A1 and the air amount calculated by the catalyst inlet NOx amount C calculating unit 43 during the lean operation, the stored NOx amount A2 and the unstored NOx amount D during the lean operation are calculated by the calculating unit 44. while calculating a calculates the storage amount of _{O 2} B2 based on the storage amount of _{O 2} B1 in the storage amount of _{O 2} B2 calculation unit 45.

When the unstored NOx amount D during the stored lean operation is equal to or greater than a predetermined value, the lean flag is set to 0, and the rich operation is performed. Further, based on the stored NOx amount A2, the stored O ₂ amount B2, and the actual fuel injection amount, the rich amount control unit 46 calculates a rich control completion flag. When the rich control completion flag becomes 1, the lean flag is set to 1. In order to perform the lean operation again, the target air-fuel ratio is set to the lean target air-fuel ratio.

FIG. 28 is a block diagram showing the rich amount control unit 46 of the target air-fuel ratio calculation unit (rich spike control unit) 35 of FIG. Multiplied by the predetermined value C with respect to the storage NOx amount A2, a value obtained by adding a predetermined value D as a storage NOx purification required fuel amount is multiplied by a predetermined value E with respect to the storage amount of O ₂ B2, adding a predetermined value F The value obtained is the stored O ₂ consumption required fuel amount, and the sum of the two is the total required fuel amount during the rich spike. The actual fuel injection amount for each control cycle is subtracted from the total required fuel amount, and when it becomes 0, the rich control completion flag is set to 1.

This is based on the knowledge that the amount of stored NOx and the amount of stored O ₂ are proportional to the amount of fuel required to purify each. In addition, the rich operation may be terminated when the output of the O ₂ sensor 20 is richly inverted (predetermined value or more), but the fuel is excessively increased by the response delay from the fuel injection to the O ₂ sensor 20. Since it will be supplied, HC and CO will be deteriorated. Therefore, it is preferable to perform the feedforward control as in the fourth embodiment.

  FIG. 29 is a block diagram showing the catalyst inlet NOx amount C calculation unit 43 of the target air-fuel ratio calculation unit (rich spike control unit) 35 of FIG. The basic NOx emission value is obtained from the engine torque and engine speed. Further, a NOx emission correction value is obtained from the EGR rate and the engine air-fuel ratio. The catalyst inlet NOx amount C (concentration) is obtained by multiplying the NOx emission basic value and the NOx emission correction value.

  The engine torque may be the target torque calculated by the target torque calculation unit 31 shown in FIG. As the engine air-fuel ratio, a target air-fuel ratio calculated by a target air-fuel ratio calculation unit shown in FIG. 27 may be used. Further, the NOx emission basic value and the NOx emission correction value may be a method in which the results of actual machine tests are described in advance in a map or the like and the map is referred to.

  FIG. 30 is a block diagram showing the stored NOx amount A2 and the unstored NOx amount D calculating unit (lean NOx catalyst model) 44 of the target air-fuel ratio calculating unit (rich spike control unit) 35 of FIG.

  The catalyst upstream NOx amount (mass) calculating means 45 uses the catalyst inlet NOx amount C (concentration), the intake air amount, and the exhaust temperature, and the NOx amount upstream of the catalyst 11 per unit time (for example, every control cycle). Calculate (mass). The conversion formula from concentration to mass is not detailed here. The stored NOx amount is subtracted from the catalyst upstream NOx amount (mass) to obtain the unstored NOx amount per unit time, and the unstored NOx amount D (mass) is obtained as the integrated value.

  The conversion formula from mass to concentration is not detailed here. The NOx storage amount is obtained by multiplying the NOx amount (mass) upstream of the catalyst per unit time by NOx purification (storage rate). The amount of NOx stored here represents the amount of NOx stored in unit time, that is, a certain period T [s] as described above, and since the unit is [g / T], it may be expressed as the storage rate. . At the time of implementation, as described above, the fixed period T [s] corresponds to a control cycle. For example, when the control cycle is 10 ms, the unit of the NOx storage amount is [g / 10 ms]. The amount of NOx stored for a certain period T [s] is integrated, and the amount of stored NOx A2 stored in the catalyst during the lean time is obtained.

The NOx purification (storage) rate is obtained by the product of the basic storage rate and the basic storage rate correction. The basic storage rate is obtained from the saturation rate. That is, the NOx purification (storage) rate decreases as the stored NOx amount reference value increases. The saturation rate is determined by (storage NOx amount A2) / (maximum storage amount).
The maximum storage amount is obtained from the exhaust temperature. The basic storage rate correction is obtained from the engine air-fuel ratio. Each parameter of this model should be determined from actual machine test results according to the characteristics of the catalyst.

  FIG. 30 shows specifications for changing the relationship between the saturation rate and the basic storage rate, which are parameters of the lean NOx catalyst model. When the NOx storage capacity change flag F_cat = 1 of the lean NOx catalyst 11, as shown in FIG. 32, the relationship between the saturation ratio and the basic storage ratio of the lean NOx catalyst model is changed.

  FIG. 31 shows the NOx storage capacity change detection unit of the lean NOx catalyst 11 in the NOx storage capacity change detection unit 46 of the lean NOx catalyst of the stored NOx amount A2 and the unstored NOx amount D calculation unit (lean NOx catalyst model) 44. It is the block diagram shown. When the absolute value of the difference between the stored NOx amount A1 and the stored NOx amount A2 is equal to or greater than a predetermined value, it is assumed that the NOx storage performance of the lean NOx catalyst has changed, and Fcat = 1. That is, the NOx storage amount A2 predicted using the model at the time of lean operation is compared with the NOx amount A1 actually stored in the catalyst calculated at the time of the rich spike performed thereafter, and the difference is equal to or greater than a predetermined value. When the NOx storage performance is changed, the parameters of the NOx catalyst model are changed.

  FIG. 32 shows the saturation rate vs. basic storage rate curves fn (Rs) and fd (Rs) when new and when deteriorated. When F_cat = 1, the saturation rate vs. basic storage rate curve fm (Rs) at the time of new degradation 1 is obtained by the equation (1) in FIG.

here,
An: NOx storage during lean operation under specified conditions when the catalyst is new
Ad: NOx storage amount during lean operation under a predetermined condition at the time of deterioration 2. That is, based on the stored NOx amount A1, the internal division points of the saturation rate vs. basic storage rate curve at the time of a new product and at the time of predetermined deterioration (2 at the time of deterioration) are taken, and the internal division ratio is set to A1, Ad, An. It is decided based on.
Further, when the NOx storage amount A1 becomes equal to or less than a predetermined value, the storage NOx performance of the lean NOx catalyst 11 is out of the allowable range, and for example, a notification lamp or the like may be turned on so that the driver can be notified.

[Embodiment 5]
FIG. 33 shows the overall system configuration of an engine to which the engine control apparatus of Embodiment 5 of the present invention is applied. In the fifth embodiment, a three-way catalyst 29 is added upstream of the lean NOx catalyst 11 with respect to the system of the eleventh embodiment shown in FIG. 14, and the O ₂ sensor 30 is interposed between the three-way catalyst 29 and the lean NOx catalyst 11. Has been added. Since other than that is the same as that of Embodiment 1, detailed description is abbreviate | omitted.

FIG. 34 shows the internal configuration of the control unit 16, and an output signal of the O ₂ sensor 30 and an input signal to the input circuit 24 are added to that of the first embodiment shown in FIG. 15. Since other than that is the same as that of Embodiment 1, detailed description is abbreviate | omitted.

FIG. 35 is a block diagram showing the overall control of the engine control apparatus according to the fifth embodiment, which is the main part of the air-preceding torque base control.
In the fifth embodiment of FIG. 35, the output of the O ₂ sensor 30 is input to the lean NOx catalyst diagnostic unit 38 as compared to the first embodiment. The rest is the same as in the first embodiment of FIG. The target air-fuel ratio calculation unit 35 of the fifth embodiment in FIG. 35 is the same as the target air-fuel ratio calculation unit 35 in the first embodiment in FIG. FIG. 36 is a block diagram showing the lean NOx catalyst diagnostic unit 38 of the fifth embodiment shown in FIG. Based on the outputs of the O ₂ sensor 20 and the O ₂ sensor 30, the fuel injection amount, and the lean flag calculated by the target air-fuel ratio calculating unit 35 shown in FIG. 35, the stored NOx amount A1 calculating means 39 in the lean NOx catalyst 11 storage NOx amount A1 of the storage amount of _{O 2} in the storage amount of _{O 2} B1 calculation unit 40 B1, and calculates the storage amount of _{O 2} B2 in the three-way catalyst 29 in storage _{O 2} amount B2 calculation unit 47.

Specifically, the storage NOx amount A1 in the lean NOx catalyst 11 is multiplied by Ga4 to MR4, the storage amount of O ₂ B1 in the lean NOx catalyst is multiplied by the Gb3 on Ml 3. Further, Mr3 is multiplied by Gb4 to obtain the stored O ₂ amount B2 in the three-way catalyst 29.

The calculation method of the fuel injection amounts Mr4, Ml3 and Mr3 will be shown below.
Ga4, Gb3 and Gb4 are respectively the amount of NOx stored in the lean NOx catalyst 11 and the amount of fuel required to reduce it, the amount of O ₂ stored in the lean NOx catalyst and the amount of fuel required to consume it. Furthermore, this is a coefficient that represents the relationship between the amount of O ₂ stored in the three-way catalyst 29 and the amount of fuel required to consume it, and should be determined empirically.

FIG. 37 is a block diagram showing the Mr3 computing unit 48 of the stored O ₂ amount B2 computing unit 47. It is determined whether the rich operation is being performed (the lean flag is 0), and during the rich operation, that is, after the air-fuel ratio of the engine 100 becomes rich, the output of the O ₂ sensor 30 downstream of the lean NOx catalyst 11 becomes rich. A value obtained by multiplying the accumulated amount Mr3_0 of the fuel injection amount in the period Tr3 by a predetermined value GMr3 is defined as Mr3.

This period Tr3 corresponds to a period in which the storage O ₂ in the three-way catalyst 29 oxidizes the reducing agent (H ₂ , CO, HC) flowing from the catalyst inlet. Therefore, this period Tr3 is the storage in the three-way catalyst 29. This is based on the fact that there is a correlation in the amount of O ₂ . A lean flag shown in FIG. 17 is used to determine whether the rich operation is being performed. GMr3 is an adjustment gain, and may normally be 1.

FIG. 38 is a block diagram showing the Mr4 calculating unit 49 of the stored NOx amount A1 calculating means 39. It is determined whether the rich operation is being performed. During the rich operation, that is, after the air-fuel ratio of the engine 100 becomes rich and the output of the O ₂ sensor 30 downstream of the three-way catalyst 29 becomes a predetermined value or more, the lean A value obtained by multiplying the accumulated amount Mr4_0 of the fuel injection amount during the period Tr4 by the predetermined value GMr4 until the output of the O ₂ sensor 20 downstream of the NOx catalyst 11 becomes equal to or greater than the predetermined value (rich) is defined as Mr4. This period Tr4 corresponds to a period in which the stored NOx and the stored O ₂ in the lean NOx catalyst 11 oxidize the reducing agent (H ₂ , CO, HC) flowing from the catalyst inlet. Therefore, this period Tr4 is the lean NOx catalyst 11. The correlation between the amount of stored NOx and the amount of stored O ₂ is utilized. GMr4 is an adjustment gain, and may normally be 1.

FIG. 39 is a block diagram showing the Ml3 calculating unit 50 of the stored O ₂ amount B1 calculating means 40. It is determined whether the lean operation is being performed. During the lean operation, that is, after the air-fuel ratio of the engine 100 becomes lean and the output of the O ₂ sensor 30 downstream of the three-way catalyst 29 becomes a predetermined value or less (lean). A value obtained by multiplying the fuel injection amount integrated amount Ml3_0 during the period Tl3 during which the output of the O ₂ sensor 20 downstream of the lean NOx catalyst 11 is equal to or less than a predetermined value (lean) by a predetermined value GMl3 is defined as Ml3. This period Tl3 corresponds to a period in which O ₂ flowing from the catalyst inlet is stored in the lean NOx catalyst 11, and therefore, this period Tl3 is used to correlate with the amount of stored O ₂ in the lean NOx catalyst 11. Is. A lean flag shown in FIG. 17 is used to determine whether the rich operation is being performed. GMl3 is an adjustment gain, and may normally be 1.

The figure which is provided for description of the aspect of Claim 1 of the control apparatus of the engine of this invention. The figure which is provided for description of the aspect of Claim 2 of the control apparatus of the engine of this invention. The figure which is provided for description of the aspect of Claim 3 of the control apparatus of the engine of this invention. The figure which is provided for description of the aspect of Claim 4 of the control apparatus of the engine of this invention. The figure which is provided for description of the aspect of Claim 5 of the control apparatus of the engine of this invention. The figure which is provided for description of the aspect of Claim 6 of the control apparatus of the engine of this invention. The figure which is provided for description of the aspect of Claim 7 of the control apparatus of the engine of this invention. The figure which is provided for description of the aspect of Claim 8 of the control apparatus of the engine of this invention. The figure which is provided for description of the aspect of Claim 9 of the control apparatus of the engine of this invention. The figure which is provided for description of the aspect of Claim 10 of the control apparatus of the engine of this invention. The figure which is provided for description of the aspect of Claim 11 of the control apparatus of the engine of this invention. The figure which is provided for description of the aspect of Claim 12 of the control apparatus of the engine of this invention. The figure which is provided for description of the aspect of Claim 13 of the control apparatus of the engine of this invention. 1 is a system diagram showing the overall configuration of Embodiment 1 of an engine control apparatus according to the present invention. FIG. 3 is an internal configuration diagram of the engine control device (control unit) according to the first embodiment. FIG. 2 is an overall control block diagram of the engine control apparatus according to the first embodiment. FIG. 3 is a control block diagram illustrating a target air-fuel ratio calculation unit of the engine control apparatus according to the first embodiment. FIG. 3 is a control block diagram illustrating a lean NOx catalyst diagnostic unit of the engine control apparatus according to the first embodiment. FIG. 3 is a control block diagram showing an Mr calculation unit in the lean NOx catalyst diagnosis unit of the first embodiment. FIG. 3 is a control block diagram illustrating an Ml calculation unit in the lean NOx catalyst diagnosis unit of the first embodiment. FIG. 6 is a control block diagram illustrating a target air-fuel ratio calculation unit of the engine control apparatus according to the second embodiment. FIG. 5 is a control block diagram illustrating a lean NOx catalyst diagnostic unit of the engine control apparatus of the second embodiment. FIG. 6 is a control block diagram showing an Ml2 calculation unit in the lean NOx catalyst diagnosis unit of the second embodiment. The whole control block diagram of Embodiment 3 of the control device of the engine concerning the present invention. FIG. 6 is a control block diagram showing an Ml2 calculation unit in a lean NOx catalyst diagnosis unit of Embodiment 3. The whole control block diagram of Embodiment 4 of the control device of the engine concerning the present invention. FIG. 6 is a control block diagram illustrating a target air-fuel ratio calculation unit according to a fourth embodiment. FIG. 6 is a control block diagram showing a rich amount control unit in a target air-fuel ratio calculation unit of Embodiment 4. FIG. 6 is a control block diagram showing a catalyst inlet NOx amount C calculation unit in a target air-fuel ratio calculation unit of Embodiment 4. FIG. 10 is a control block diagram illustrating a stored NOx amount A3 and an unstored NOxD calculation unit in a target air-fuel ratio calculation unit according to a fourth embodiment. FIG. 6 is a control block diagram illustrating a NOx storage capacity change detection unit of a lean NOx catalyst of a target air-fuel ratio calculation unit according to a fourth embodiment. The figure which shows the update method of the relationship between the saturation rate of the target air fuel ratio calculating part of Embodiment 4, and a basic storage rate. The system figure which shows the whole structure of Embodiment 5 of the control apparatus of the engine which concerns on this invention. FIG. 10 is an internal configuration diagram of an engine control device (control unit) according to a fifth embodiment. FIG. 10 is an overall control block diagram of an engine control apparatus according to a fifth embodiment. FIG. 10 is a control block diagram illustrating a lean NOx catalyst diagnostic unit according to a fifth embodiment. FIG. 10 is a control block diagram showing an Mr3 calculation unit in the lean NOx catalyst diagnosis unit of the fifth embodiment. FIG. 10 is a control block diagram showing an Mr4 calculation unit in the lean NOx catalyst diagnosis unit of the fifth embodiment. FIG. 10 is a control block diagram illustrating an Ml3 calculation unit in the lean NOx catalyst diagnosis unit of the fifth embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Air cleaner 2 Air flow sensor 3 Electronic throttle 4 Intake pipe 5 Collector 6 Accelerator 7 In-cylinder fuel injection valve 8 Spark plug 9 Cylinder 10 Exhaust pipe 11 Lean NOx catalyst 12 A / F sensor 13 Accelerator opening sensor 14 Water temperature sensor 15 Engine Rotational speed sensor 16 Control unit (control device)
17 Throttle opening sensor 18 Exhaust gas recirculation pipe 19 Exhaust gas recirculation control valve 20 O ₂ sensor downstream of lean NOx catalyst 29 Three-way catalyst 30 O ₂ sensor downstream of three-way catalyst 100 Engine

Claims (14)

An engine comprising: a lean NOx catalyst that stores NOx discharged from the engine during lean operation, and desorbs and purifies the stored NOx during rich operation; and an O ₂ sensor disposed downstream of the lean NOx catalyst. A control device,
The engine control apparatus characterized by comprising means for calculating a stored NOx amount A1 of the lean NOx catalyst, means for calculating a storage amount of O ₂ B1 of the lean NOx catalyst. The means for calculating the lean NOx catalyst storage O ₂ amount B1 is such that, at the end of the rich operation of the lean NOx catalyst, after the air-fuel ratio of the engine becomes lean, the output of the O ₂ sensor becomes a predetermined value or less. Based on the time period Tl until or the fuel injection amount Ml during the time period Tl, the stored O ₂ amount B1 is calculated,
The means for calculating the stored NOx amount A1 of the lean NOx catalyst is obtained when the air-fuel ratio of the engine becomes rich during the rich operation performed to desorb and purify the stored NOx of the lean NOx catalyst. The period Tr until the output of the _two sensors becomes a predetermined value or more or the fuel injection amount Mr during the period Tr is calculated, and the difference between the period Tr and the period Tl or the fuel injection amount Mr and the fuel injection amount Ml The engine control apparatus according to claim 1, wherein the stored NOx amount A1 is calculated based on the difference. The means for calculating the stored NO ₂ amount B1 of the lean NOx catalyst is such that, at the end of the rich operation of the lean NOx catalyst, the air-fuel ratio of the engine is stoichiometric and the output of the O ₂ sensor becomes a predetermined value or less. Based on the period Tl2 or the fuel injection amount Ml2 during the period Tl2, the stored O ₂ amount B1 is calculated,
The means for calculating the stored NOx amount A1 of the lean NOx catalyst is obtained when the air-fuel ratio of the engine becomes rich during the rich operation performed to desorb and purify the stored NOx of the lean NOx catalyst. The period Tr until the output of the _two sensors becomes a predetermined value or more or the fuel injection amount Mr during the period Tr is calculated, and the difference between the period Tr and the period Tl2 or the fuel injection amount Mr and the fuel injection amount Ml2 The engine control apparatus according to claim 1, wherein the stored NOx amount A1 is calculated based on the difference. The means for calculating the stored O ₂ amount B1 of the lean NOx catalyst is operated after a predetermined time of operation at a lean air-fuel ratio at which the air-fuel ratio of the engine is stoichiometric or the NOx storage performance of the lean NOx catalyst is not exhibited. The air-fuel ratio is made rich, and the stored O ₂ amount B1 is calculated based on the period Tr2 or the fuel injection amount Mr2 during the period Tr2 until the output of the O ₂ sensor downstream of the lean NOx catalyst becomes a predetermined value or more. The engine control device according to claim 1. An engine control device provided with an O ₂ sensor or an A / F sensor upstream of the lean NOx catalyst,
A frequency analysis of the output signal of the O ₂ sensor or the A / F sensor, and a frequency analysis of the output signal of the O ₂ sensor downstream of the lean NOx catalyst, means for calculating a power Pa1 and a phase Ph1 of a predetermined frequency, Means for calculating power Pa2 and phase Ph2 of a predetermined frequency,
_2. The engine control device according to claim 1, wherein a stored O ₂ amount B < _b > 1 is calculated based on a relationship among the power Pa <b> 1, the power Pa < _b > 2, and the phase Ph < _b > 1 and the phase Ph < _b > 2.   The engine control apparatus according to claim 1, further comprising means for calculating a start time of rich spike control performed to purify NOx stored in the lean NOx catalyst based on a stored NOx amount A1. . Means for calculating a rich spike amount or rich spike period during rich spike control performed to purify NOx stored in the lean NOx catalyst based on the stored NOx amount A1 and the stored O ₂ amount B1 The engine control device according to claim 1. Based on the stored O ₂ amount B1 obtained during the rich spike or during the operation near the stoichiometric condition, the stored O ₂ amount B2 stored in the lean NOx catalyst during a predetermined period of the lean operation performed after the rich spike is calculated. Means to estimate;
Means for estimating a stored NOx amount A2 stored in the lean NOx catalyst during a predetermined period of a lean operation performed after the rich spike based on the stored NOx amount A1 obtained during the rich spike;
Means for estimating a catalyst inlet NOx amount C flowing into the lean NOx catalyst during a predetermined period of lean operation performed after the rich spike;
Means for calculating an unstored NOx amount D discharged downstream of the lean NOx catalyst without being stored from the catalyst inlet NOx amount C and the stored NOx amount A2.
Means for calculating a start time of rich spike control based on the unstored NOx amount D;
The engine control apparatus according to claim 1, characterized by comprising, means for calculating the rich spike amount or rich spike period based on the storage NOx amount A2 and the storage amount of O ₂ B2.   The means for estimating the catalyst inlet NOx amount C flowing into the lean NOx catalyst during a predetermined period of the lean operation performed after the rich spike is a basic value of NOx emission calculated based on the engine torque and the engine speed. 9. The engine control apparatus according to claim 8, wherein the catalyst inlet NOx amount C is estimated by multiplying both the EGR rate and the NOx emission correction value calculated based on the engine air-fuel ratio. The means for estimating the stored NOx amount A2 is a means for calculating a storage rate Rn of NOx stored in the lean NOx catalyst per unit time based on the estimated stored NOx amount A2 calculated last time, and the stored NOx. A means for correcting the relationship between the estimated storage NOx amount A2 and the storage rate Rn based on the amount A1,
The amount of stored NOx per unit time is calculated by multiplying the catalyst inlet NOx amount C flowing into the lean NOx catalyst by the NOx storage rate Rn, and the stored NOx amount per unit time is integrated, and the estimated stored NOx amount A2 The engine control device according to claim 8, wherein:   9. The engine control device according to claim 8, wherein the rich spike control start timing calculation means starts the rich spike control when the unstored NOx amount D is equal to or greater than a predetermined value. The engine control device comprising a three-way catalyst upstream of the lean NOx catalyst,
The means for calculating the stored O ₂ amount B2 of the three-way catalyst, and the rich spike amount or rich spike period at the time of rich spike control performed to purify NOx stored in the lean NOx catalyst, and the stored NOx amount A1 The engine control device according to claim 1, further comprising means for calculating based on the stored O ₂ amount B1 and the stored O ₂ amount B2. The engine control device comprising another O ₂ sensor between the three-way catalyst and the lean NOx catalyst,
The means for calculating the stored O ₂ amount B2 of the three-way catalyst is the period Tr3 or the period Tr3 from when the air-fuel ratio of the engine becomes rich until the output of the other O ₂ sensor becomes a predetermined value or more. Based on the fuel injection amount Mr3 in the storage, the stored O ₂ amount B2 in the three-way catalyst is calculated,
The means for calculating the stored O ₂ amount B1 of the lean NOx catalyst is the downstream of the lean NOx catalyst after the output of the other O ₂ sensor becomes a predetermined value or less at the end of the rich operation of the lean NOx catalyst. Based on the period Tl3 until the output of the O ₂ sensor becomes equal to or less than the predetermined value or the fuel injection amount Ml3 during the period Tl3, the stored O ₂ amount B1 is calculated,
The means for calculating the stored NOx amount A1 of the lean NOx catalyst is such that the output of the other O ₂ sensor becomes a predetermined value or more during the rich operation performed to desorb and purify the stored NOx of the lean NOx catalyst. After that, a period Tr4 until the output of the O ₂ sensor downstream of the lean NOx catalyst becomes equal to or greater than a predetermined value is calculated, or a fuel injection amount Mr4 during the period Tr4 is calculated, and a difference between the period Tr4 and the fuel injection amount Tl3 is calculated. The engine control device according to claim 12, wherein the stored NOx amount A1 is calculated based on a difference between the period Mr4 and the fuel injection amount Ml3. An automobile equipped with the engine control device according to claim 1.

Images