JP5981827B2 - Air-fuel ratio control device - Google Patents

Description

  The present invention relates to an air-fuel ratio control apparatus, for example, an air-fuel ratio control apparatus suitable for use in a vehicle (such as a motorcycle) equipped with an internal combustion engine.

  For example, in a system in which exhaust gas from an internal combustion engine (hereinafter referred to as an engine) such as an automobile is purified by a catalyst device and released, the air-fuel ratio of the engine exhaust gas is improved in the exhaust gas purification capability of the catalyst device. Control to an appropriate air-fuel ratio is desired from the viewpoint of environmental protection.

  As an apparatus that performs such air-fuel ratio control, for example, there is an air-fuel ratio control apparatus described in Patent Document 1.

  This Patent Document 1 discloses a target air-fuel ratio of a fuel injection amount obtained from a fuel injection amount map (engine speed, throttle opening, negative pressure, etc. are parameters) for determining a fuel injection amount in an engine. An air-fuel ratio control apparatus having a configuration in which a correction coefficient is superimposed on the fuel injection amount is disclosed.

  Specifically, a LAF sensor (a sensor that converts a signal of a level proportional to the oxygen concentration (air-fuel ratio) of the exhaust gas over a wide range of oxygen concentration (air-fuel ratio)) upstream of a catalyst device (purifier) disposed in the exhaust pipe of the engine. And an oxygen sensor (air-fuel ratio sensor) downstream of the catalyst device. Then, the predicted value of the air-fuel ratio after the catalyst is obtained using the detection value of the LAF sensor, and the correction coefficient is obtained using, for example, a sliding mode controller using the predicted value.

Japanese Patent No. 3373724

  By the way, since the LAF sensor is expensive, there is a demand to abolish the LAF sensor installed on the upstream side of the catalyst device due to the cost reduction of the system and the limitation of the arrangement space in a motorcycle or the like. .

  However, the output value (SVO2) of the oxygen sensor, which is the emission target value, is set to the target value based on the output value (SVO2) that is the input value of the sliding mode controller (SMC) modeled on engine intake and exhaust. If the LAF sensor is not installed upstream of the catalyst device, the air-fuel ratio before the catalyst cannot be measured. Therefore, the engine tolerance and aging deterioration in the engine model, the fuel injection valve injection error The prediction of the predicted value of the output value (SVO2) is expanded, and it may take time for the sliding mode controller (SMC) to converge to the target value.

  In addition, since the convergence gain of the sliding mode controller (SMC) also has a limit of adjustment, the output value (SVO2) cannot be converged to the target value without losing the prediction error of the predicted value of the output value (SVO2). Is also possible.

  Depending on the specifications of the internal combustion engine, a device for introducing secondary air may be installed in the exhaust pipe upstream from the catalyst in order to reduce exhaust gas emissions. When secondary air is introduced, the pre-catalyst air-fuel ratio is shifted, which may affect the entire control system of the sliding mode controller. In general, the secondary air is set so as to supply sufficient air so that the residual fuel in the exhaust pipe can be surely burned unless special feedback is performed. For this reason, when secondary air is introduced, the air-fuel ratio upstream of the catalyst tends to become lean (excess oxygen). For this reason, when the secondary air is introduced, the oxygen storage amount of the catalyst is likely to increase immediately after the introduction of the secondary air, and as a result, the NOx purification performance may be temporarily lowered.

  In addition to introducing secondary air, when the fuel injection is cut when the throttle is fully closed, air flows through the catalyst, which increases the amount of oxygen stored in the catalyst and is the same problem as introducing secondary air. Occurs. Similarly, the amount of oxygen stored in the catalyst also increases when deceleration leaning is performed without performing fuel cut from the viewpoint of catalyst protection.

  The present invention has been made in consideration of such problems, and it is possible to optimize the air-fuel ratio without installing a LAF sensor upstream of the catalyst device, thereby reducing system costs, motorcycles, and the like. The application of air-fuel ratio control to the air-fuel ratio can be promoted, and even when secondary air is introduced, fuel injection cut or deceleration leaning is performed, the air-fuel ratio is promptly returned from the control state to the normal state. It is an object of the present invention to provide an air-fuel ratio control device that can be returned to control and can maintain high emission performance.

First feature: at least the engine (28), the catalyst (50) installed in the exhaust pipe (32) of the engine (28), and secondary air in the exhaust path upstream of the catalyst (50). The secondary air introduction device (1000) to be introduced is installed in a vehicle (12), and the fuel injection amount to the engine (28) is controlled so that the air-fuel ratio of the exhaust gas of the engine (28) is appropriate. An air-fuel ratio control device for controlling the air-fuel ratio, comprising: an air-fuel ratio detection means (52) provided downstream of the catalyst (50 ) for detecting an actual air-fuel ratio (SVO2) downstream of the catalyst (50); , basic fuel injection quantity calculator for determining a basic fuel injection amount to the engine (28) based on at least a preset basic fuel injection map (118) and the engine rotational speed and the throttle opening parameters (116), less An air-fuel ratio control unit main body (100) for determining a target air-fuel ratio (KO2) with respect to the basic fuel injection amount based on the actual air-fuel ratio (SVO2) from the air-fuel ratio detection means (52), and the secondary An air introduction device (1008A, 1008B) that operates in accordance with the introduction of the secondary air by the air introduction device (1000), and the air-fuel ratio control unit body (100) includes at least the actual air-fuel ratio (SVO2). ) On the downstream side of the catalyst (50) after a dead time (dt) corresponding to the distance from the fuel injection valve (40) of the engine (28) to the air-fuel ratio detection means (52). predict fuel ratio, based on the predicted air fuel ratio predicting means for outputting as (DVPRE) (102), the predicted air from the air-fuel ratio estimating means (102) (DVPRE), the target air A correction coefficient calculating means for determining a first correction coefficient for obtaining the ratio (KO2) a (DKO2OP) by sliding mode control (104), wherein the predicted air from an air-fuel ratio estimating means (102) to (DVPRE) the A second correction coefficient (KTIMB) for making the prediction error (ERPRE), which is the difference between the output (DVPRE (k−dt)) delayed by the dead time (dt) and the actual air-fuel ratio (SVO2), zero. produced and a adaptive model correcting means (122), the air-fuel ratio estimating means (102), and the actual air-fuel ratio (SVO2) from the air-fuel ratio detecting means (52), wherein the correction coefficient calculating means (104 The corrected air-fuel ratio obtained using the first correction coefficient (DKO2OP) from) and the second correction coefficient (KTIMB) from the adaptive model correcting means (122) Based on the above, the air-fuel ratio on the downstream side of the catalyst (50) is predicted, and the air introduction corresponding part (1008A, 1008B) is the correction coefficient calculating means ( 100) when the introduction of the secondary air is started. and means (1010) to suspend the sliding mode control in 104), immediately after the introduction of the secondary air is completed, and means for the rich control of open-loop air-fuel ratio control (1012), the actual air-fuel ratio ( And means (1014) for resuming the sliding mode control in the correction coefficient calculation means (104 ) when SVO2) becomes rich.

Second feature: at least the engine (28), a catalyst (50) installed in the exhaust pipe (32) of the engine (28), and a throttle valve (installed in the intake pipe (30) of the engine (28)). 38) is installed in a vehicle (12) having a fuel cut control means (1004) for controlling the stop of fuel injection during the closed period, and controls the fuel injection amount to the engine (28), An air-fuel ratio control apparatus for controlling an air-fuel ratio of exhaust gas of the engine (28) to an appropriate air-fuel ratio, provided downstream of the catalyst (50), and an actual air-fuel ratio ( downstream of the catalyst (50) ( and air-fuel ratio detecting means for detecting the SVO2) (52), based on the engine (28) based on at least a preset basic fuel injection map (the 118) and the engine rotational speed and the throttle opening degree parameter Based on the basic fuel injection amount calculation unit (116) for determining the fuel injection amount and at least the actual air-fuel ratio (SVO2) from the air-fuel ratio detection means (52), the target air-fuel ratio (KO2) with respect to the basic fuel injection amount An air-fuel ratio control unit main body (100) for determining the air-fuel ratio, and an air introduction corresponding unit (1008A, 1008B) operating in accordance with fuel injection stop control by the fuel cut control means (1004). The main body (100) has a dead time (dt) corresponding to the distance from the fuel injection valve (40) of the engine (28) to the air-fuel ratio detection means (52) based on at least the actual air-fuel ratio (SVO2). ) the predicted fuel ratio downstream of the catalyst (50) after the lapse of a air-fuel ratio predicting means for outputting a predicted air (DVPRE) (102), the air-fuel ratio predicted hand Wherein from (102) based on the predicted air (DVPRE), a first correction coefficient for obtaining the target air-fuel ratio (KO2) correction coefficient calculating means for determining the sliding mode control (DKO2OP) (104), This is a difference between an output (DPRE (k−dt)) obtained by delaying the predicted air / fuel ratio (DVPRE) from the air / fuel ratio predicting means (102) by the dead time (dt) and the actual air / fuel ratio (SVO2). and a second correction coefficient for a prediction error (ERPRE) to zero adaptive model correction means (122) for generating (KTIMB), the air-fuel ratio estimating means (102), the air-fuel ratio detecting means (52 wherein the actual air-fuel ratio (SVO2) from), the correction factor the first correction factor from the calculation means (104) (DKO2OP) and said adaptive model correcting means (12 2), the air-fuel ratio on the downstream side of the catalyst (50) is predicted based on the corrected air-fuel ratio obtained using the second correction coefficient (KTIMB) from 2), and the air introduction corresponding portions (1008A, 1008B) ), the when the stop control of the fuel injection is started by the fuel cut control means (1004), and means (1010) to suspend the sliding mode control in the correction coefficient calculating means (104), said fuel Immediately after the stop control of the fuel injection by the cut control means (1004) is completed, the means (1012) for setting the air-fuel ratio control to the open loop rich control and the stage where the actual air-fuel ratio (SVO2) is on the rich side. And means (1014) for restarting the sliding mode control in the correction coefficient calculation means (104) .

Third feature: installed in a vehicle (12) having at least an engine (28), a catalyst (50) installed in an exhaust pipe (32) of the engine (28), and a deceleration lean control means (1005). And an air-fuel ratio control device for controlling the air-fuel ratio of the exhaust gas of the engine (28) to an appropriate air-fuel ratio by controlling the fuel injection amount to the engine (28), and downstream of the catalyst (50) provided, the air-fuel ratio detecting means for detecting an actual air-fuel ratio (SVO2) of the downstream side of the catalyst (50) (52), at least a preset basic fuel injection map (the 118) engine speed and the throttle opening A basic fuel injection amount calculation unit (116) for determining a basic fuel injection amount for the engine (28) based on the degree parameter, and at least the actual air-fuel ratio (SV) from the air-fuel ratio detection means (52) 2), the air-fuel ratio control unit body (100) for determining the target air-fuel ratio (KO2) with respect to the basic fuel injection amount, and the introduction of air containing oxygen by the deceleration lean control means (1005). The air-fuel ratio control unit main body (100) is operated based on at least the actual air-fuel ratio (SVO2), and the fuel injection valve (40) of the engine (28). ) To the air-fuel ratio detecting means (52) , the air-fuel ratio downstream of the catalyst (50) after the dead time (dt) has elapsed is predicted, and the air-fuel ratio is output as the predicted air-fuel ratio (DVPRE) and predicting means (102), wherein from an air-fuel ratio estimating means (102) based on the predicted air (DVPRE), the first correction coefficient for obtaining the target air-fuel ratio (KO2) (DKO2 A correction coefficient calculating means for determining the sliding mode control P) (104), the said predicted air from an air-fuel ratio estimating means (102) to (DVPRE) is delayed by the dead time (dt) Output (DVPRE ( k-dt)) and the a second correction coefficient for the zero prediction error (ERPRE) which is the difference between the actual air-fuel ratio (SVO2) (adaptive model correcting means (122 for generating a KTIMB)) The air-fuel ratio predicting means (102) includes the actual air-fuel ratio (SVO2) from the air-fuel ratio detecting means (52), the first correction coefficient (DKO2OP) from the correction coefficient calculating means (104), and the Based on the corrected air-fuel ratio obtained using the second correction coefficient (KTIMB) from the adaptive model correcting means (122), the air-fuel ratio downstream of the catalyst (50) is preliminarily estimated. The air / fuel ratio predicting means (102) is based on the actual air / fuel ratio (SVO2) from the air / fuel ratio detecting means (52) and the corrected air / fuel ratio from the multiplier (136). The air-fuel ratio on the downstream side of the catalyst (50) is predicted, and the air introduction corresponding part (1008A, 1008B) is adapted to calculate the correction coefficient in accordance with the introduction of air containing oxygen by the deceleration lean control means (1005). (104) and means (1010) to suspend the sliding mode control, the immediately after the introduction of the deceleration leaning control means air containing oxygen by (1005) is completed, the rich control of the open loop air-fuel ratio control and means (1012), said at a stage where the actual air-fuel ratio (SVO2) becomes richer, to resume the sliding mode control in the correction coefficient calculating means (104) And having a means (1014).

Fourth feature: In any one of the first to third features, the air-fuel ratio control unit body (100) determines a prediction accuracy based on a PID control unit (124) and the prediction error (ERPRE). A prediction accuracy determination unit (146), and the PID control unit (124) determines a decrease in prediction accuracy by the prediction accuracy determination unit (146) before resuming the sliding mode control. In this case, PID control is performed so that an error between the actual air-fuel ratio (SVO2) and a preset target value becomes zero, and the air-fuel ratio on the downstream side of the catalyst (50) converges to the target value. It is characterized by controlling as follows .

5th characteristic; 4th characteristic WHEREIN: The said air-fuel ratio control part main body (100) WHEREIN: When the determination which shows that the prediction accuracy was ensured in the said prediction accuracy determination means (146), the said The PID control is switched to the sliding mode control.

Sixth feature: In any one of the first to third features, the air-fuel ratio control unit body (100) includes a PID control unit (124), at least the correction coefficient calculation means (104), and the PID control unit. A control unit (126 ) for controlling (124), and the adaptive model correcting means (122) has a prediction accuracy determining means (146) for determining a prediction accuracy based on the prediction error (ERPRE), When the sliding accuracy control by the correction coefficient calculating unit (104) is resumed, the control unit (126), when the prediction accuracy determining unit (146) determines that the prediction accuracy is reduced, pause the sliding mode control, the start PID control section (124), the PID controller (124), said without the use of air-fuel ratio estimating means (102), prior to An error between a preset target value and the actual air-fuel ratio (SVO2) and the PID control so as to zero the air-fuel ratio on the downstream side of the catalyst (50) is controlled to converge to the target value Features.

The seventh aspect; in any of the first to third aspect, the air introducing corresponding portion (1008B) further comprises P ID control unit (1020), prior to resuming the sliding mode control When the error between the actual air-fuel ratio (SVO2) and a preset target value exceeds a preset threshold value, the PID control unit (1020) causes the PID to be zeroed by the PID control unit (1020). And controlling so that the air-fuel ratio on the downstream side of the catalyst (50) converges to the target value .

(1) According to the first feature, even in a system in which the secondary air introduction device is installed, after the secondary air introduction, the feedback control is once interrupted and the introduction of the secondary air is completed. By performing the rich injection control, the amount of oxygen stored in the catalyst by introducing the secondary air can be reduced at an early stage. As a result, it is possible to quickly return to the interrupted feedback control, and to maintain high emission performance.

(2) When the fuel injection is stopped and controlled by the fuel cut control means by closing the throttle opening, air containing oxygen itself flows into the catalyst, so that the amount of oxygen stored in the catalyst increases. However, according to the second feature, when the fuel injection stop control by the fuel cut control means is started, the feedback control is temporarily interrupted, and when the fuel injection stop control is completed, the rich injection control is performed. Thus, the amount of oxygen stored in the catalyst due to the introduction of air containing oxygen accompanying the stop control of fuel injection can be reduced at an early stage. As a result, it is possible to quickly return to the interrupted feedback control, and to maintain high emission performance.

(3) When deceleration leaning by the deceleration lean control means is started, air containing oxygen itself flows into the catalyst as in the case of the fuel cut control described above, so that the amount of oxygen stored in the catalyst increases. become. However, according to the third feature, when the deceleration leaning by the deceleration lean control means is started, the feedback control is temporarily interrupted, and the rich injection control is performed at the stage where the deceleration leaning is completed. As a result, the amount of oxygen stored in the catalyst due to the introduction of air containing oxygen accompanying the deceleration leaning can be reduced at an early stage. As a result, it is possible to quickly return to the interrupted feedback control, and to maintain high emission performance.

(4) According to the fourth feature, it is predicted that the convergence of the sliding mode control takes time due to the introduction of secondary air, or the introduction of air containing oxygen accompanying fuel cut control or deceleration leaning. In this case, when the introduction of secondary air is completed, or when fuel cut control or deceleration leaning is completed, the PID control is performed first, so that the convergence of the feedback control can be accelerated and the emission performance is improved. Can be maintained.

(5) According to the fifth feature, if the prediction error is equal to or less than the threshold value, the convergence of the sliding mode control is ensured. Therefore, the emission performance can be maintained high by restarting the feedback control. it can.

(6) According to the sixth feature, when the decrease in prediction accuracy is determined when the feedback control is resumed, the actual air-fuel ratio and the preset target value are used without using the air-fuel ratio predicting means. Since the PID control is performed so that the error is zero, the time until the prediction accuracy is ensured can be shortened and the emission performance is higher than when the air-fuel ratio prediction means is used. Can be maintained.

(7) According to the seventh feature, prior to restarting the feedback control, the dedicated PID control unit is used to perform PID control so that the difference (error) between the actual air-fuel ratio and the target value becomes zero. As a result, the error can be converged below the threshold earlier than when the prediction error of the air-fuel ratio control unit body is used, and normal sliding mode control (control by the first sliding mode control unit) can be performed. It can be restarted sooner. This leads to maintaining high emission performance.

1 is a perspective view showing an example of a motorcycle on which an air-fuel ratio control apparatus according to an embodiment is installed. It is a block diagram showing an example of a control system of an engine of a motorcycle. It is a functional block diagram which shows the structure of the air fuel ratio control part which has a 1st air introduction corresponding | compatible part. It is a control block diagram which shows the structure of the air fuel ratio control apparatus (air fuel ratio control part) which concerns on this Embodiment. It is a control block diagram which shows the structure of the air fuel ratio control part main body which concerns on a comparative example. It is explanatory drawing which shows the prediction model by a predictor. It is explanatory drawing which shows the operation | movement concept of sliding mode control. It is a block diagram which shows the structure of an adaptive model modifier. It is a block diagram which shows the specific structure of an adaptive model modifier. FIG. 10A is a characteristic diagram showing a change in the output of the oxygen sensor with respect to the air-fuel ratio A / F, and FIG. 10B is a characteristic diagram showing a change in the first weighting component with respect to the actual air-fuel ratio. FIG. 11A is a characteristic diagram showing a change in the basic fuel injection amount with respect to the throttle opening, and FIG. 11B is a characteristic diagram showing a change in the second weighting component with respect to the throttle opening. 12A is a characteristic diagram showing a weighting function for the engine speed NE, and FIG. 12B is a characteristic diagram showing a weighting function for the throttle opening TH. It is a figure for demonstrating the principle which calculates | requires a correction coefficient from prediction error correction amount. It is a control block diagram which shows the structure of the air fuel ratio control part main body which concerns on a 1st modification. It is a control block diagram which shows the structure of the air fuel ratio control part main body which concerns on a 2nd modification. It is a control block diagram which shows the structure of the air fuel ratio control part main body which concerns on a 3rd modification. It is a control block diagram which shows the structure of the air fuel ratio control part main body which concerns on a 4th modification. It is a control block diagram which shows the structure of the air fuel ratio control part main body which concerns on a 5th modification. It is a flowchart which shows the processing operation of a 1st air introduction corresponding | compatible part. Implementation period of secondary air introduction or FC control or deceleration leaning, rich spike control period (residual oxygen treatment period of the catalyst device), change in fuel injection amount per unit time in rich spike control, catalyst device 6 is a timing chart showing a change in the oxygen storage amount and a change in the output value (SVO2) of the oxygen sensor. It is a functional block diagram which shows the structure of the air fuel ratio control part which has a 2nd air introduction corresponding | compatible part. It is a flowchart which shows the processing operation of a 2nd air introduction corresponding | compatible part.

  Hereinafter, an embodiment in which the air-fuel ratio control apparatus according to the present invention is applied to, for example, a motorcycle will be described with reference to FIGS.

  First, a motorcycle 12 equipped with an air-fuel ratio control apparatus 10 according to the present embodiment will be described with reference to FIG.

  As shown in FIG. 1, the motorcycle 12 is configured such that a vehicle body front portion 14 and a vehicle body rear portion 16 are connected via a low floor portion 18. The vehicle body front portion 14 has a handle 20 rotatably attached to an upper portion thereof, and a front wheel 22 supported on a lower portion thereof. The vehicle body rear portion 16 has a seat 24 attached to the upper portion thereof, and a rear wheel 26 supported on the lower portion thereof.

  As schematically shown in FIG. 2, the engine 28 of the motorcycle 12 is provided with an intake pipe 30 and an exhaust pipe 32, and the intake pipe 30 is piped between the engine 28 and the air cleaner 34. The throttle body 36 provided in the intake pipe 30 is provided with a throttle valve 38. A fuel injection valve 40 is provided between the engine 28 and the throttle body 36 on the intake pipe 30.

  The throttle valve 38 is rotated in accordance with the rotation operation of the throttle grip 42 (see FIG. 1), and the rotation amount (the opening degree of the throttle valve 38) is detected by the throttle sensor 44. The amount of air supplied to the engine 28 is made variable by opening and closing the throttle valve 38 according to the driver's operation of the throttle grip 42.

  The engine 28 is provided with a water temperature sensor 46 that detects the engine cooling water temperature, and the intake pipe 30 is provided with a PB sensor 48 that detects intake air pressure (intake negative pressure). A secondary air introducing device 1000 provided upstream of the catalyst device 50 installed in the exhaust pipe of the engine 28 and introducing air from the air cleaner 34 into the exhaust pipe as secondary air, and installed in the exhaust pipe of the engine 28 An oxygen sensor 52 (air-fuel ratio detection means) that is provided downstream of the catalyst device 50 and detects an air-fuel ratio downstream of the catalyst device 50 is provided. The oxygen concentration detected by the oxygen sensor 52 corresponds to the actual air-fuel ratio of the exhaust gas after passing through the catalyst device 50. Further, the engine 28 is provided with a vehicle speed sensor 56 that detects the vehicle speed from the rotational speed of the output gear of the speed reduction mechanism 54. The starter switch 58 is a switch for starting the engine 28 by operating an ignition key. Further, an atmospheric pressure sensor 60 is provided at a position far from the intake pipe 30 of the air cleaner 34.

  As shown in FIG. 3, the engine control device (engine control unit: ECU 62) includes an AI (secondary air introduction) control unit 1002, an FC (fuel cut) control unit 1004, and a deceleration lean control unit 1005. And an air-fuel ratio control unit 1006 that functions as the air-fuel ratio control apparatus 10 according to the present embodiment.

  The AI control unit 1002 drives the secondary air introduction device 1000 while the predetermined condition for introducing the secondary air is satisfied, and uses the air from the air cleaner 34 as the secondary air. And introduced upstream of the catalyst device 50. The AI control unit 1002 outputs an AI start signal Sais prior to the start of introduction of secondary air or at the start of introduction, and outputs an AI end signal Saie when the introduction of secondary air is completed.

  The FC control unit 1004 performs fuel cut control for interrupting fuel injection during a period in which a predetermined condition for performing FC control such as zero throttle opening (fully closed) is satisfied. The FC control unit 1004 outputs the FC control start signal Sfcs prior to the start of execution of the fuel cut control or at the start of execution, and outputs the FC control end signal Sfce when the fuel cut control ends.

  The deceleration lean control unit 1005 performs deceleration leaning such as reducing the basic injection pulse width during a period in which a predetermined condition for performing deceleration leaning is established based on the amount of decrease in throttle opening and the amount of change in intake pressure. Run. The deceleration lean control unit 1005 outputs a deceleration leaning start signal Srss prior to the start of execution of deceleration leaning or at the start of execution, and outputs a deceleration leaning completion signal Srse when deceleration leaning ends.

  The air-fuel ratio control unit 1006 includes an air-fuel ratio control unit main body 100, a basic fuel injection amount calculation unit 116, and a first air introduction corresponding unit 1008A. The air-fuel ratio control unit main body 100 is a control unit main body using sliding mode control, which will be described in detail later.

  The basic fuel injection amount calculation unit 116 obtains a reference fuel injection amount defined from the engine speed NE, the throttle opening TH, and the intake air pressure PB using the basic fuel injection map 118, and calculates the reference fuel injection amount. The basic fuel injection amount TIMB is calculated by correcting according to the effective opening area of the throttle valve 38.

The basic fuel injection map 118 includes a first basic fuel injection map 118a based on the engine speed NE and the throttle opening TH, and a second basic fuel injection map 118b based on the engine speed NE and the intake air pressure PB. Therefore, in the air-fuel ratio control unit 1006, of the first basic fuel injection map 118a and the second basic fuel injection map 118b, an index of the basic fuel injection map to be used based on the engine speed NE and the throttle opening TH. Has a map selection unit 142 for selecting and instructing a basic fuel injection map to be used from the selection map 140 arranged. As shown in FIG. 8 , the selection map 140 includes an area where the first basic fuel injection map 118a is to be used and an area where the second basic fuel injection map 118b is to be used. The map selection unit 142 selects a basic fuel injection map to be used from the selection map 140 based on the input engine speed NE and the throttle opening TH, and outputs the selection result Sa. If the engine speed NE is low, the probability that the first basic fuel injection map 118a will be selected increases, and if the engine speed NE is high, the probability that the second basic fuel injection map 118b will be selected increases.

  Therefore, the basic fuel injection amount calculation unit 116 uses the basic fuel injection map selected by the map selection unit 142 for the reference fuel injection amount defined by the engine speed NE, the throttle opening TH, and the intake air pressure PB. The basic fuel injection amount TIMB is calculated by correcting the reference fuel injection amount according to the effective opening area of the throttle valve 38. The basic fuel injection amount TIMB is corrected by the target air-fuel ratio KO2 (k) from the air-fuel ratio control unit main body 100 and the environment correction coefficient KECO including the water temperature, the intake air temperature, the atmospheric pressure, etc., and is output as the fuel injection time Tout. The

  The first air introduction handling unit 1008A includes an air-fuel ratio control stop request unit 1010, a rich spike control unit 1012, and a restart determination unit 1014.

  The air-fuel ratio control stop request unit 1010 receives the AI start signal Sais from the AI control unit 1002, the FC control start signal Sfcs from the FC control unit 1004, or the deceleration lean start signal Srss from the deceleration lean control unit 1005. Based on this, a stop request signal Sg is output to the air-fuel ratio control unit main body 100. The control unit 126 of the air-fuel ratio control unit main body 100 temporarily stops the air-fuel ratio control based on the input of the stop request signal Sg from the air-fuel ratio control stop request unit 1010.

  The rich spike control unit 1012 is based on the input of the AI end signal Saie from the AI control unit 1002, the FC control end signal Sfce from the FC control unit 1004, or the deceleration leaning end signal Srse from the deceleration lean control unit 1005. Then, the fuel injection valve 40 is controlled so that the rich spike in the combustion chamber of the engine 28 (for the purpose of reducing the NOx adsorbed to the catalyst device, the fuel injection amount is temporarily set to a richer air-fuel ratio than usual). Perform rich injection). In this rich spike control, the current output value (SVO2) of the oxygen sensor 52 is referred to while referring to a rich spike map 1016 in which the fuel supply amount per unit time corresponding to the output value (SVO2) of the oxygen sensor 52 is registered. The amount of fuel supplied in accordance with is supplied to the combustion chamber of the engine 28 per unit time. The rich spike control unit 1012 ends the rich spike control when the output value (SVO2) of the oxygen sensor 52 is on the rich side.

  The restart determination unit 1014 outputs a control restart signal Si to the air-fuel ratio control unit main body 100 when the output value (SVO2) of the oxygen sensor 52 becomes rich. The air-fuel ratio control unit main body 100 restarts the air-fuel ratio control based on the input of the control restart signal Si. The resumption of air-fuel ratio control in the air-fuel ratio control unit main body 100 will be described later.

  Next, the air-fuel ratio control unit main body 100 will be described with reference to FIGS.

  As shown in FIG. 4, the air-fuel ratio control unit main body 100 includes a predictor 102 (air-fuel ratio predicting means) that predicts an air-fuel ratio downstream of the catalyst device 50, and a predicted air-fuel ratio DVPRE from the predictor 102. A first sliding mode control unit 104 (correction coefficient calculating means) for determining a first correction coefficient DKO2OP (k) for the fuel injection amount based on the identifier, and an identifier for identifying parameters of the first sliding mode control unit 104 and the predictor 102 106 and an air-fuel ratio reference value calculation unit 108 for calculating an air-fuel ratio reference value.

  Here, the operations of the predictor 102, the first sliding mode control unit 104, the identifier 106, and the air-fuel ratio reference value calculation unit 108 are similar to the comparative example of FIG. 5 (the air-fuel ratio control device described in Patent Document 1). The description will be made in comparison with the air-fuel ratio control unit main body 300).

  First, in the air-fuel ratio control unit main body 300 according to the comparative example of FIG. 5, the LAF sensor 110 (see the broken line block in FIG. 2) is installed on the upstream side of the catalyst device 50, and the pre-catalyst air-fuel ratio from the LAF sensor 110. It is assumed that A / F (k) is input.

  The predictor 102 corresponds to the dead time dt (the distance from the fuel injection valve 40 to the oxygen sensor 52) from the current time (k) in order to determine the fuel injection amount (target air-fuel ratio) on the downstream side of the catalyst device 50. The air-fuel ratio (VO2) after the dead time has elapsed is predicted.

  The prediction model by the predictor 102 is as follows when the current time is k and the air-fuel ratio φin before the catalyst from the time point ta to the time point tb and the output Vout of the oxygen sensor 52 are known as shown in FIG. The output Vout (k + dt) = Vpre (k) at the time point k + dt can be predicted from the relational expression (1).

  However, since φin of j = 1 to (dt−d−1) cannot be observed at time k, the target value (φop) is substituted. Here, Vout ′ (K) is the deviation between the output of the oxygen sensor 52 and the target value at time k, and Vout ′ (K−1) is one unit time (constant time period) before time k. The deviation between the output of the oxygen sensor 52 and the target value is shown. α1, α2, and βj are parameters determined by the identifier 106.

  The first sliding mode control unit 104 calculates the injection amount according to the model error (predicted air / fuel ratio−target value). Normally, in the sliding mode control, as shown in FIG. 7, a switching straight line represented by a linear function having a plurality of state quantities to be controlled as variables is constructed in advance, and these state quantities are subjected to high gain control. To converge on the switching line at high speed (arrival mode), and further, by so-called equivalent control input, the state quantity is constrained on the switching line and converged to a required equilibrium point (convergence point) on the switching line ( This is a variable structure type feedback control method.

  In such sliding mode control, if a plurality of state quantities to be controlled converge on the switching line, the state quantity is stably converged to the equilibrium point on the switching line without being affected by disturbances or the like. It has excellent properties that it can.

  When determining the correction amount of the air-fuel ratio of the engine 28 so that the concentration of the specific component such as the oxygen concentration of the exhaust gas downstream of the catalyst device 50 is set to a predetermined appropriate value, for example, the exhaust gas downstream of the catalyst device 50 The concentration value of the specific component of the gas and its rate of change are defined as the state quantities of the exhaust system to be controlled, and these state quantities are respectively converted to equilibrium points (concentration values and changes thereof) on the switching line using sliding mode control. A correction amount of the air-fuel ratio is obtained so that the speed converges to a predetermined appropriate value and “0”. If the correction amount of the air-fuel ratio is obtained using the sliding mode control, the concentration of the specific component of the exhaust gas on the downstream side of the catalyst can be accurately set to a predetermined appropriate value as compared with the conventional PID control or the like. .

  The switching function and the control input arithmetic expression in the sliding mode control are as follows.

  Here, Ueq (k) is an equivalent law input, Urch (k) is a reaching law input, and Uadp (k) is an adaptive law input, which are calculated by the following equations. Here, Vout ′ (k) and Vout ′ (k−1) indicate model errors, and Vout ′ (k) is a deviation between the predicted air-fuel ratio and the target value at time k, and Vout ′ (k k-1) indicates the deviation between the predicted air-fuel ratio and the target value one unit time (constant time period) before time k.

  Krch and Kadp are feedback gains, and S is a switching function setting parameter.

  The identifier 106 corrects the prediction accuracy of the predictor 102 by correcting the model parameter of the predictor 102. For the first sliding mode control unit 104, a model equation by adjusting the convergence speed (feedback gain) of σ (k) to the switching straight line according to the model error.

The parameters a1 (k), a2 (k), and b1 (k) are adjusted so that the deviation of Vout ′ (k + 1) calculated by the above is minimized. This corrects the correspondence relationship between Vout with respect to the pre-catalyst air-fuel ratio φin and the target air-fuel ratio φop by correcting the model parameter of the prediction formula.

  The air-fuel ratio reference value calculation unit 108 obtains the air-fuel ratio reference value of the engine 28 defined by the adaptive law input Uadp (k) from the first sliding mode control unit 104 using a preset map.

  The output from the first sliding mode control unit 104, that is, the control input Uop (= DKO2OP (k)) to the exhaust system is added to the air-fuel ratio reference value from the air-fuel ratio reference value calculation unit 108 by the adder 112. Thus, the target air-fuel ratio KO2 (k) is obtained. This target air-fuel ratio KO2 (k) is input to the adaptive control unit 114 at the subsequent stage. This adaptive control unit 114 detects changes in the operating state and characteristics of the engine 28 from the detected air-fuel ratio φin (= A / F (k)) of the LAF sensor 110 and the target air-fuel ratio φop (KO2 (k)). This is a recurrence type controller that adaptively obtains the feedback correction coefficient KAF in consideration of a change in the environment.

  Then, the basic fuel injection amount calculation unit 116 obtains a reference fuel injection amount defined by the engine speed NE, the throttle opening TH, and the intake air pressure PB using a preset basic fuel injection map 118, and The basic fuel injection amount TIMB is calculated by correcting the reference fuel injection amount according to the effective opening area of the throttle valve. The basic fuel injection amount TIMB is supplied to the multiplier 120, and is corrected by the feedback correction coefficient KAF from the adaptive control unit 114 and the environmental correction coefficient KECO including the water temperature, the intake air temperature, the atmospheric pressure, etc., as the fuel injection time Tout. Is output.

  Since the air-fuel ratio control unit main body 300 according to the comparative example as described above uses the expensive LAF sensor 110, there is a problem that it cannot be applied to a motorcycle or the like having a limited system cost and a limited arrangement space. . Therefore, in the air-fuel ratio control unit main body 300 according to the comparative example, when the LAF sensor 110 is not installed upstream of the catalyst device 50, the air-fuel ratio φin before the catalyst cannot be measured. If the engine 28 and the fuel injection valve 40 have a large deviation from the stoichiometric air-fuel ratio due to variations in characteristics of the engine 28 and the fuel injection valve 40, deterioration of the air-fuel ratio, etc. It is expected that it will be difficult to achieve this.

  Therefore, as shown in FIG. 4, the air-fuel ratio control unit main body 100 according to the present embodiment calculates a deviation between the actual air-fuel ratio SVO2 (k) and the predicted air-fuel ratio DVPRE (k−dt) as a prediction error ERPRE (k). And the prediction accuracy in the predictor 102 decreases with the adaptive model corrector 122 (adaptive model correcting means) that superimposes the second correction coefficient KTIMB on the first correction coefficient DKO2OP (k) so as to make this zero. At this stage, the second sliding mode / PID control unit 124 that feeds back the error between the actual air-fuel ratio SVO2 (k) and the preset target value to zero, and at least the first sliding mode control unit 104 are adapted. Based on an instruction from the control unit 126 that controls the model corrector 122 and the control unit 126, the output on the first sliding mode control unit 104 side and the second slide And a switching unit 128 for switching the output of the i Recording mode / PID controller 124 side. The switching unit 128 normally selects the output on the first sliding mode control unit 104 side and switches to the output on the second sliding mode / PID control unit 124 side based on the switching instruction signal Sd from the control unit 126.

  The second sliding mode / PID control unit 124 includes a second sliding mode control unit and a PID control unit, and one of the control units is selected according to an instruction from the control unit 126. For example, in normal operation other than special situations such as secondary air introduction, fuel cut control, and deceleration leaning, the difference between the moving average of the prediction error ERPRE (k) after filtering and a preset threshold value In response, either the second sliding mode control unit or the PID control unit is selected. If the difference is greater than a preset selection reference value, the PID control unit is selected, and if the difference is less than or equal to the selection reference value, the second sliding mode control unit is selected. The selection reference value may be determined as follows, for example. That is, grasp the convergence (convergence time, etc.) by the second sliding mode control unit with respect to the difference in advance through experiments, simulations, etc., and further grasp the range of the difference where the convergence does not affect other control systems, For example, the selection reference value is determined from the grasped difference range.

  As a result, when the difference is equal to or less than the selection criterion, the second sliding mode control unit is selected and feedback control is performed so that the error between the actual air-fuel ratio and the preset target value becomes zero. It is possible to ensure the prediction accuracy. In particular, when the difference exceeds the selection criterion, the PID control unit is selected. Therefore, even when the difference is large, the time until the prediction accuracy is ensured can be further shortened.

  Further, the control unit 126 activates the PID control unit based on the input of the control restart signal Si output from the restart determination unit 1014 when the secondary air introduction is completed, the fuel cut control is completed, or the deceleration leaning is completed. select. Thereby, even if the said difference is large, time until prediction accuracy is ensured can be shortened more. That is, the suspended sliding mode control in the air-fuel ratio control unit main body 100 that has been interrupted can be returned early, and the emission performance can be maintained high.

  Further, the air-fuel ratio control unit main body 100 includes a time adjustment unit 130 that delays the predicted air-fuel ratio DVPRE (k) from the predictor 102 by a dead time dt, and an output DVPRE (k−dt) from the time adjustment unit 130. A subtractor 132 that takes a difference from the actual air-fuel ratio SVO2 (k) from the oxygen sensor 52 to obtain a prediction error ERPRE (k). The prediction error ERPRE (k) from the subtracter 132 is adaptive model correction. Is supplied to the vessel 122. The adder 134 adds 1 to the second correction coefficient KTIMB output from the adaptive model corrector 122. The output of the adder 134 and the target air-fuel ratio KO2 (k) are multiplied by the multiplier 136 and output as a corrected air-fuel ratio in which the second correction coefficient KTIMB is superimposed on the target air-fuel ratio KO2 (k). The corrected air-fuel ratio is subtracted from the air-fuel ratio reference value by the subtractor 138 and input to the predictor 102 and the identifier 106.

  Then, as shown in FIG. 8, the adaptive model corrector 122 is based on the filter processing unit 144 that performs various filter processes on the prediction error ERPRE (k) in the first stage, and the prediction error ERPRE (k) after the filter process. A prediction accuracy determination unit 146 (prediction accuracy determination means) that determines the prediction accuracy, a first correction amount calculation unit 148a and a first correction coefficient calculation unit 150a corresponding to the first basic fuel injection map 118a, and a second basic fuel A second correction amount calculation unit 148b and a second correction coefficient calculation unit 150b corresponding to the injection map 118b are provided.

  When the first basic fuel injection map 118a is selected by the map selection unit 142, the first correction amount calculation unit 148a reflects the weight component due to the engine speed NE and the throttle opening TH in a certain period of time. The prediction error correction amount θth (i, j) is fed back so that the prediction error ERPRE (k) becomes zero. For example, the calculation is started before the dead time dt of the time point k, that is, from the time point (k−dt), the calculation is performed at a constant time period, and the prediction error correction amount θthIJ (k) is output at the time point k.

  Specifically, as shown in FIG. 9, the first weighting component WSO2S (k) in which the sensitivity to the air-fuel ratio of the oxygen sensor 52 is reflected with respect to the prediction error ERPRE (k) and the engine in a certain period of time. The second weighting component Wtha (k-dt) reflecting the change in the value of the first basic fuel injection map 118a with respect to the change in the rotational speed NE and the throttle opening TH, and the first basic fuel injection map 118a are used as the engine speed NE. And a third weighting component WthIJ (k−dt) corresponding to a plurality of areas divided based on the throttle opening TH and weighting to obtain a correction model error EwIJ (k) corresponding to the plurality of areas. Unit 152 and a prediction error corresponding to the plurality of regions so that the correction model error EwIJ (k) corresponding to the plurality of regions is set to zero in a certain period of time. Correction amount θthIJ a (k), respectively and a sliding mode controller 154 for feedback.

  The first weighting component WSO2S (k) will be described. As shown in FIG. 10A, the output Vout of the oxygen sensor 52 has a nonlinear characteristic with respect to the air-fuel ratio A / F. In the regions Za and Zc, the output Vout of the oxygen sensor 52 hardly changes even when the air-fuel ratio changes. On the other hand, in the region Zb, the output Vout of the oxygen sensor 52 changes greatly with a slight change in the air-fuel ratio A / F. In FIG. 10A, the solid line La indicates the characteristics after the new catalyst, and the broken line Lb indicates the characteristics after the aged catalyst. If such characteristics are directly reflected in the correction model error EwIJ (k), a sudden change in the region Zb is input to the sliding mode control unit 154, and the correction model error EwIJ (k) is set to zero. There is a problem that it takes time. Therefore, as shown in FIG. 10B, the weighting value is set to be smaller in the region Zb so that the rapid change is alleviated.

  Explaining the second weighting component Wtha, the output value SVO2 of the oxygen sensor 52 indicates that the probability that the prediction error ERPRE is caused by the detection error of the throttle opening TH is based on the change in the throttle opening TH as shown in FIG. 11A. The higher the inclination of the basic fuel injection amount Tibs, the higher. When a detection error occurs and the reference point of the value of the basic fuel injection amount in the basic fuel injection map is shifted, the amount of change in the air-fuel ratio increases as “change amount accompanying shift / value at reference point” increases. Therefore, “(inclination of basic fuel injection amount Tibs relative to change in throttle opening TH) ÷ (value of basic fuel injection amount Tibs)” is set for each engine speed NE. As a result, as shown in FIG. 11B, when the engine speed NE is high, the second weighting component Wtha is substantially the same over the throttle opening TH from fully closed to fully open, but the engine speed NE is low. As the throttle opening TH decreases, the second weighting component Wtha increases.

  For example, as shown in FIG. 12A, the third weighting component WthIJ has a peak at the engine speed NE when the engine speed NE is 1000, 2000, 3000, 4500 (rpm). The function is such that the weighting value decreases linearly from the adjacent vertex to the adjacent vertex. However, in FIG. 12A, the weighting value is constant at an engine speed of 1000 rpm or less and 4500 rpm or more. Similarly, as shown in FIG. 12B, when the weighting function for the throttle opening TH of 1 °, 3 °, 5 °, and 8 ° is viewed, the throttle opening TH is a vertex, and the vertexes adjacent from each vertex The weighting value is a function that decreases linearly toward. However, in FIG. 12B, the weighting value is constant when the throttle opening is 1 ° or less and 8 ° or more.

  The third weighting component WthIJ is obtained by multiplying the weighting Wthn (i) based on the engine speed NE by the weighting Wtht (j) based on the throttle opening TH.

  Note that the sliding mode control unit 154 feeds back the prediction error correction amount θthIJ so that the correction model error EwIJ is zero for the region where the third weighting component WthIJ> WthIJ> 0, and the third weighting component WthIJ is WthIJ. For the region where = 0, the operation amount = 0, so that the prediction error correction amount θthIJ is not updated.

  The first correction coefficient calculation unit 150a superimposes a third weighting component WthIJ corresponding to each of the plurality of regions on the prediction error correction amount θthIJ (k) corresponding to the plurality of regions at a predetermined timing. The correction coefficient KTITHIJ corresponding to is obtained, and all the correction coefficients are added to obtain the second correction coefficient KTIMB. Here, since all the correction coefficients are added, the third weighting component WthIJ is on the region including the point determined by the engine speed NE and the throttle opening TH in the first basic fuel injection map 118a. The weighting according to the position of the point is shown. Therefore, as shown in FIG. 13, a plurality of regions having grid points of engine speeds of 1000, 2000, 3000, 4500 (rpm) and throttle openings of 1 °, 3 °, 5 °, 8 ° are formed, When the point determined by the input engine speed NE and the throttle opening TH is the point A, the correction coefficient corresponding to the point A is complemented by the correction coefficients of the four surrounding points.

  On the other hand, when the second basic fuel injection map 118b is selected by the map selection unit 142, the second correction amount calculation unit 148b reflects the weight component due to the engine speed NE and the intake air pressure PB in a certain period of time. The prediction error correction amount is fed back so that the predicted error becomes zero. For example, the calculation is started before the dead time dt of the time point k, that is, from the time point (k−dt), and the calculation is performed at a constant time period, and the prediction error correction amount θpbIJ (k) is output at the time point k. The specific configuration of the second correction amount calculation unit 148b is substantially the same as that of the first correction amount calculation unit 148a shown in FIG.

  The second correction coefficient calculation unit 150b superimposes a third weighting component corresponding to each of the plurality of regions on the prediction error correction amount θpbIJ (k) corresponding to the plurality of regions at a predetermined timing, thereby overlapping the plurality of regions. A corresponding correction coefficient is obtained, and all the correction coefficients are added to obtain a second correction coefficient KTIMB. The specific configuration of the second correction coefficient calculation unit 150b is also substantially the same as that of the first correction coefficient calculation unit 150a shown in FIG.

  The prediction accuracy determination unit 146 determines that the prediction accuracy has decreased when a state in which the moving average of the prediction error ERPRE (k) after the filtering process is higher than a preset threshold value continues for a set number of times. A decrease signal Sb is output. When the moving average of the prediction error after the filtering process continues for a set number of times or more, the prediction accuracy ensuring signal Sc is output assuming that the prediction accuracy is ensured. The prediction accuracy decrease signal Sb and the prediction accuracy ensuring signal Sc are supplied to the control unit 126.

  As shown in FIG. 4, the control unit 126 temporarily stops the processing by the first sliding mode control unit 104 and temporarily stops the identifier 106 based on the input of the prediction accuracy decrease signal Sb (see FIG. 8). In the meantime, the activation cycle of the adaptive model corrector 122 is shortened. That is, the fixed time period for activating the first correction amount calculation unit 148a and the second correction amount calculation unit 148b is shortened.

  Further, the control unit 126 outputs a switching instruction signal Sd to the switching unit 128 based on the input of the prediction accuracy decrease signal Sb. The switching unit 128 switches to the output on the second sliding mode / PID control unit 124 side based on the input of the switching instruction signal Sd. The control unit 126 also starts processing in the second sliding mode / PID control unit 124 based on the input of the prediction accuracy lowering signal Sb. In this case, the predicted air-fuel ratio from the predictor 102 is not used. Further, as described above, the control unit 126 determines whether the second sliding mode control unit and the PID control unit correspond to the difference between the moving average of the prediction error ERPRE (k) after the filter process and a preset threshold value. Select one of the following. If the difference is larger than a preset selection reference value, the PID control unit is selected, and if the difference is equal to or less than the selection reference value, the second sliding mode control unit is selected. In particular, the control unit 126 activates the PID control unit based on the input of the control restart signal Si output from the restart determination unit 1014 when the secondary air introduction ends, the fuel cut control ends, or the deceleration leaning ends. Force selection. The second sliding mode / PID control unit 124 performs feedback so that an error between the actual air-fuel ratio (SVO2) and a preset target value (for example, a fixed value indicating a stoichiometric region) becomes zero. The output from the second sliding mode / PID control unit 124 is supplied to the multiplier 120 via the switching unit 128. The basic fuel injection amount calculation unit 116 selects a reference fuel injection amount defined from the engine speed NE, the throttle opening TH, and the intake air pressure PB by a preset basic fuel injection map or the map selection unit 142. The basic fuel injection amount TIMB is calculated by correcting the reference fuel injection amount in accordance with the effective opening area of the throttle valve 38. This basic fuel injection amount TIMB is corrected by an output from the switching unit 128 (target air-fuel ratio KO2 (k)) and an environmental correction coefficient KECO including water temperature, intake air temperature, atmospheric pressure, etc., and is output as a fuel injection time Tout. The

  The temporary suspension of the first sliding mode control unit 104 and the identifier 106 may be canceled by the output of the prediction accuracy ensuring signal Sc from the prediction accuracy determining unit 146, or a predetermined time (prediction accuracy is ensured). It is also possible to cancel after the elapse of the expected time). In this case, since the supply of the switching instruction signal Sd from the control unit 126 to the switching unit 128 is stopped, the switching unit 128 switches to the output on the first sliding mode control unit 104 side. In addition, the control unit 126 restores the fixed time period for activating the first correction amount calculation unit 148a and the second correction amount calculation unit 148b in the adaptive model corrector 122. Further, the control unit 126 releases the temporary suspension of the first sliding mode control unit 104 and resets the parameters of the identifier 106 to initial values.

  Thus, in the air-fuel ratio control apparatus 10 (air-fuel ratio control unit 1006) according to the present embodiment, the second correction coefficient KTIMB is superimposed on the target air-fuel ratio KO2 (k) in the predictor 102 and the identifier 106. A value obtained by subtracting the air-fuel ratio reference value from the value is input. That is, since the predictor 102 outputs the predicted air-fuel ratio DVPRE (k) after the dead time dt based on the actual air-fuel ratio SVO2 (k), the predicted air-fuel ratio DVPRE (k) is delayed by the dead time dt. Thus, the difference between the actual air-fuel ratio SVO2 (k) and the predicted air-fuel ratio DVPRE (k−dT) that are matched in time is input to the adaptive model corrector 122 as the prediction error ERPRE (k). The adaptive model modifier 122 superimposes the second correction coefficient KTIMB on the first correction coefficient DKO2OP (k) so that the prediction error ERPRE (k) becomes zero, and the value is applied to the predictor 102 and the identifier 106. This is input and reflected in the processing in the predictor 102.

  That is, the first correction coefficient DKO2OP (k) obtained by feeding back the deviation between the predicted air-fuel ratio DVPRE (k) from the predictor 102 and the target air-fuel ratio KO2 (k) to zero, and the prediction error ERPRE The second correction coefficient KTIMB obtained by feedback so that (k) is set to zero is superimposed and input to the predictor 102. Therefore, even if the LAF sensor 110 that has been conventionally installed upstream of the catalyst device 50 is abolished, the accuracy of predicting the air-fuel ratio on the downstream side of the catalyst device 50 can be ensured. The air-fuel ratio can be converged to an appropriate value, and as a result, the purification performance of the catalyst device 50 can be ensured. Moreover, even when an air-fuel ratio error occurs due to variations in characteristics of the engine 28, the fuel injection valve 40, etc., deterioration over time, etc., a decrease in prediction accuracy can be avoided. As described above, since the LAF sensor 110 can be omitted, the harness related to the LAF sensor 110 and the interface circuit of the ECU 62 can be omitted, thereby reducing the cost of the system, saving the arrangement space, and the like. And can be easily applied to a vehicle having a small arrangement space such as the motorcycle 12. Normally, the LAF sensor 110 needs to maintain a constant temperature with a heater in order to ensure good operating characteristics. However, in this embodiment, the heater for the LAF sensor 110 can also be omitted. Electricity can be reduced and fuel consumption can be improved.

  Furthermore, in the present embodiment, since the processing by the first sliding mode control unit 104 is temporarily stopped based on the input of the prediction accuracy reduction signal Sb, the restriction on the period in the adaptive model modifier 122 is eliminated. The fixed time period for activating the first correction amount calculation unit 148a and the second correction amount calculation unit 148b can be shortened. Therefore, it is possible to shorten the time until the prediction error ERPRE (k) is set to zero.

  In addition, the process in the second sliding mode / PID control unit 124 is started without using the predicted air-fuel ratio DVPRE (k) from the predictor 102 based on the input of the prediction accuracy decrease signal Sb. The fuel injection amount is controlled so that the actual air-fuel ratio SVO2 (k) is directed to the predetermined target value, and the prediction accuracy can be ensured in a short time.

By such processing operation, even in the cases shown in the following (a) to (c), the air-fuel ratio on the downstream side of the catalyst device 50 can be converged to an appropriate value at an early stage, and the catalyst device 50 Emission deterioration due to the continued state in which the air-fuel ratio of the exhaust gas on the downstream side cannot converge to an appropriate value can be eliminated.
(A) When there is a large prediction error exceeding the adjustable range of the predictor 102 by the identifier 106 due to the occurrence of an air-fuel ratio error due to variations in characteristics of the engine 28, fuel injection valve 40, etc., deterioration over time, etc. (B) When the dynamic characteristics of the controlled object changes suddenly (exhaust volume change due to changes in operating conditions, use of ethanol mixed fuel, etc.)
(C) When in the dead zone of the oxygen sensor 52 (a region where the output of the oxygen sensor 52 hardly changes even if the air-fuel ratio changes).

  In the present embodiment, when it is determined that the prediction accuracy is ensured, the activation cycle of the adaptive model corrector 122 is returned to the original state, and the suspension of the first sliding mode control unit 104 is canceled. Since the generation of the first correction coefficient DKO2OP (k) by the first sliding mode control unit 104 is resumed when the prediction accuracy is ensured, the prediction accuracy is further improved, and the air-fuel ratio downstream of the catalyst device 50 is increased. Optimization can be accelerated.

  In this case, since the parameter of the identifier 106 is reset to the initial value, when the prediction accuracy is ensured or when the prediction accuracy is expected to be ensured, the prediction accuracy is reduced as the identification parameter. By using the initial value without using the identification parameter, it is possible to maintain the prediction accuracy, and it is possible to accelerate the optimization of the air-fuel ratio downstream of the catalyst device 50.

  Further, in the first correction amount calculation unit 148a of the adaptive model corrector 122, the prediction error in which the weight component due to the engine speed NE and the throttle opening TH with respect to the first basic fuel injection map 118a is reflected at a constant time period is zero. The prediction error correction amount θthIJ is fed back so that the second correction coefficient KTIMB is obtained based on the prediction error correction amount θthIJ at a predetermined timing in the first correction coefficient calculation unit 150a. Even if the LAF sensor 110 installed upstream is eliminated, the air-fuel ratio downstream of the catalyst device 50 can be optimized.

  In particular, the prediction error correction amount θthIJ corresponding to a plurality of regions divided from the first basic fuel injection map 118a on the basis of the engine speed NE and the throttle opening TH is set to zero in a certain period of time. A prediction error correction amount θthIJ corresponding to a plurality of regions is fed back, a correction coefficient KTITHIJ corresponding to the plurality of regions is obtained based on a prediction error correction amount θthIJ corresponding to the plurality of regions at a predetermined timing, and all correction coefficients are obtained. Since the second correction coefficient KTIMB is obtained by addition, the second correction coefficient KTIMB corrects the map value to be used with the correction coefficients KTITHIJ of a plurality of regions so that the prediction error ERPRE (k) becomes zero. Value. Therefore, the air-fuel ratio downstream of the catalyst device 50 can be optimized by superimposing the second correction coefficient KTIMB having such characteristics on the first correction coefficient DKO2OP.

  The same applies to the second correction amount calculation unit 148b and the second correction coefficient calculation unit 150b corresponding to the second basic fuel injection map 118b.

  In the above-described example, the processing of the first sliding mode control unit 104 and the identifier 106 is temporarily stopped at the stage when the prediction accuracy is determined to be lowered, and the switching unit 128 causes the second sliding mode / PID control unit 124 to For example, based on the input of a signal Se indicating that the air-fuel ratio feedback condition is satisfied from the ECU 62, the processing of the first sliding mode control unit 104 and the identifier 106 is temporarily stopped, and the switching unit At 128, the output from the second sliding mode / PID control unit 124 may be switched. In this case, when a prediction error has occurred due to operating conditions or the like before the air-fuel ratio feedback condition is satisfied, the prediction error can be eliminated at an early stage from the time when the air-fuel ratio feedback condition is satisfied. . It should be noted that the above-described pause is canceled after a predetermined time (a time when the prediction accuracy is expected to be ensured) has elapsed since the input time of the signal Se indicating that the air-fuel ratio feedback condition has been established. May be.

  In addition, after a predetermined time (predetermined time) has elapsed after the determination that the prediction accuracy has decreased, the activation cycle of the adaptive model corrector 122 is returned to the original state, and the first sliding mode control unit 104 is temporarily stopped. In the case of canceling, the generation of the first correction coefficient DKO2OP (k) by the first sliding mode control unit 104 is resumed when the prediction accuracy is secured after a predetermined time of one or more times has passed. Therefore, the prediction accuracy is further improved, and the optimization of the air-fuel ratio downstream of the catalyst device 50 can be accelerated. By setting the predetermined time as a time at which the prediction accuracy is expected to be ensured, the prediction accuracy is ensured when two predetermined times have passed.

  Further, instead of temporarily suspending the processing of the first sliding mode control unit 104 and the identifier 106 and shortening the activation cycle of the adaptive model corrector 122, the operation gain of the correction coefficient by the adaptive model corrector 122 is made higher than usual. The same effect can be obtained even if the value is increased.

  Next, modified examples of the above-described air-fuel ratio control unit main body 100 will be described with reference to FIGS.

  As shown in FIG. 14, the air-fuel ratio control unit main body 100a according to the first modification has substantially the same configuration as the air-fuel ratio control unit main body 100, but adapts to the target air-fuel ratio KO2 (k) from the adder 112. The difference is that the adder 160 adds the second correction coefficient KTIMB from the model corrector 122. Also in this case, a value obtained by adding the first correction coefficient DKO2OP (k) and the second correction coefficient KTIMB is input to the predictor 102 and the identifier 106. Therefore, the same effect as that of the air-fuel ratio control unit main body 100 can be obtained.

  As shown in FIG. 15, the air-fuel ratio control unit main body 100b according to the second modification has substantially the same configuration as the air-fuel ratio control unit main body 100, but the second correction coefficient KTIMB is given to the predictor 102 and the identifier 106. Is not reflected, and the output from the adder 112 (the value obtained by adding the first correction coefficient DKO2OP (k) from the first sliding mode control unit 104 and the air-fuel ratio reference value from the air-fuel ratio reference value calculation unit 108 (KO2OP (K)) and the output from the adder 134 (the value obtained by adding 1 to the second correction coefficient KTIMB) are multiplied by the multiplier 162 to obtain the target air-fuel ratio KO2 (k). Since the two correction coefficient KTIMB is reflected in the output of the basic fuel injection amount calculation unit 116, the same effect as that of the air-fuel ratio control unit main body 100 according to the present embodiment can be obtained.

  As shown in FIG. 16, the air-fuel ratio control unit body 100 c according to the third modification has substantially the same configuration as the air-fuel ratio control unit body 100 b according to the second modification, but the output KO2SL ( k) and the second correction coefficient KTIMB from the adaptive model corrector 122 are added by an adder 164 to obtain a target air-fuel ratio KO2 (k). Also in this case, since the second correction coefficient KTIMB is reflected in the output of the basic fuel injection amount calculation unit 116, the same effect as the air-fuel ratio control unit main body 100 according to the present embodiment can be obtained.

  As shown in FIG. 17, the air-fuel ratio control unit main body 100d according to the fourth modification has substantially the same configuration as the air-fuel ratio control unit main body 100 according to the present embodiment, but the predictor 102 and the first sliding mode A first switching unit 128 a is installed between the control unit 104 and a second switching unit 128 b is installed on the output side of the first sliding mode control unit 104. In normal times, the first switching unit 128a selects the predictor 102, and the second switching unit 128b selects the output to the adder 112. As a result, the predicted air-fuel ratio DVPRE (k) from the predictor 102 is input to the first sliding mode control unit 104, so that the first correction coefficient DKO2OP (k) from the first sliding mode control unit 104 is an adder. At 112, the air-fuel ratio reference value is added and output as the target air-fuel ratio KO2 (k). On the other hand, when the switching instruction signal Sd is output from the control unit 126, the first switching unit 128a selects the input of the actual air-fuel ratio SVO2 (k), and the second switching unit 128b selects the output to the multiplier 120. . As a result, the first sliding mode control unit 104 feeds back such that the error between the actual air-fuel ratio (SVO2) and a preset target value (for example, a fixed value indicating the stoichiometric region) becomes zero. The output from the first sliding mode control unit 104 is supplied to the multiplier 120 via the second switching unit 128b. Therefore, also in the fourth modified example, the same effect as that of the air-fuel ratio control unit main body 100 according to the present embodiment can be obtained. In particular, according to the fourth modification, the second sliding mode / PID control unit 124 can be omitted, and the configuration can be simplified.

  As shown in FIG. 18, the air-fuel ratio control unit main body 100e according to the fifth modified example has substantially the same configuration as the air-fuel ratio control unit main body 100 according to the present embodiment. The difference is that a sensor 110 is installed and the detected air-fuel ratio A / F (k) from the LAF sensor 110 is used. In this case, the adaptive control unit 114 is installed between the switching unit 128 and the multiplier 120.

  By using the LAF sensor 110, it is possible to quickly resolve a decrease in prediction accuracy due to insufficient accuracy of the basic fuel injection map. Of course, in the air-fuel ratio control unit main body 100 according to the present embodiment and the air-fuel ratio control unit main body 100d according to the first modification to the air-fuel ratio control unit main body 100d according to the fourth modification, the first sliding mode control unit 104 is provided. Is superimposed on the first correction coefficient DKO2OP (k) from the adaptive model corrector 122 and input to the predictor 102 and the identifier 106, so that the prediction accuracy can be reduced early. Although it can be eliminated, by using the LAF sensor 110, it is possible to quickly resolve a decrease in prediction accuracy due to insufficient accuracy of the basic fuel injection map 118.

  In the air-fuel ratio control unit main body 100 and various modifications according to the present embodiment described above, not only the air-fuel ratio control of the engine but also the transport delay time from the control input to the output is long, and the predictor 102 needs to be configured. Application to a certain control system is possible.

  Next, the processing operation of the first air introduction corresponding unit 1008A will be described with reference to the flowchart of FIG.

  First, in step S1 of FIG. 19, the air-fuel ratio control stop request unit 1010 determines whether or not introduction of secondary air has started. This determination is made based on whether or not an AI start signal Sais is input from the AI control unit 1002.

  When the AI start signal Sais is input, the process proceeds to the next step S2, and the air-fuel ratio control stop request unit 1010 outputs the stop request signal Sg to the air-fuel ratio control unit body 100. The air-fuel ratio control unit main body 100 temporarily stops the air-fuel ratio control based on the input of the stop request signal Sg.

  Thereafter, in step S3, the rich spike control unit 1012 waits for the end of the introduction of the secondary air. That is, it waits until an AI end signal Saie is input from the AI control unit 1002.

  On the other hand, if it is determined in step S1 that the AI start signal Sais is not input, in step S4, the air-fuel ratio control stop request unit 1010 determines whether or not FC control is started. This determination is made based on whether or not an FC control start signal Sfcs is input from the FC control unit 1004.

  When the FC control start signal Sfcs is input, the process proceeds to the next step S5, and the air-fuel ratio control stop request unit 1010 outputs the stop request signal Sg to the air-fuel ratio control unit body 100. The air-fuel ratio control unit main body 100 temporarily stops the air-fuel ratio control based on the input of the stop request signal Sg.

  Thereafter, in step S6, the rich spike control unit 1012 waits for the end of the FC control. That is, it waits until the FC control end signal Sfce is input from the FC control unit 1004.

  If it is determined in step S4 that the FC control start signal Sfcs is not input, in step S7, the air-fuel ratio control stop request unit 1010 determines whether deceleration leaning has started. This determination is made based on whether or not a deceleration leaning start signal Srss is input from the deceleration lean control unit 1005.

  When the deceleration lean start signal Srss is input, the process proceeds to the next step S8, and the air-fuel ratio control stop request unit 1010 outputs the stop request signal Sg to the air-fuel ratio control unit main body 100. The air-fuel ratio control unit main body 100 temporarily stops the air-fuel ratio control based on the input of the stop request signal Sg.

  Thereafter, in step S9, the rich spike control unit 1012 waits for the end of deceleration leaning. That is, it waits until the deceleration leaning end signal Srse is input from the deceleration lean control unit 1005.

  When the AI end signal Saie is input in step S3, the FC control end signal Sfce is input in step S6, or the deceleration leaning end signal Srse is input in step S9, the following In step S 10, the rich spike control unit 1012 controls the fuel injection valve 40 to perform rich spike control on the combustion chamber of the engine 28. In this rich spike control, the fuel supply amount corresponding to the current output value (SVO2) of the oxygen sensor 52 is supplied to the combustion chamber of the engine 28 per unit time while referring to the rich spike map 1016. This rich spike control is performed until the output value (SVO2) of the oxygen sensor 52 becomes rich.

  In step S11, the restart determination unit 1014 determines whether the output value (SVO2) of the oxygen sensor 52 has become rich, and when the output value (SVO2) of the oxygen sensor 52 has become rich, the air-fuel ratio. A control restart signal Si is output to the control unit main body 100.

  In step S12, the air-fuel ratio control unit main body 100 restarts the air-fuel ratio control based on the input of the control restart signal Si.

  Here, the processing from step S1 to step S12 will be described based on the timing chart of FIG. FIG. 20 shows the execution period Ta of introduction of secondary air or FC control or deceleration lean, rich spike control period Tb (residual oxygen treatment period of the catalyst device 50), and fuel injection per unit time in rich spike control. A change in amount, a change in oxygen storage amount in the catalyst device 50 (see solid line), and a change in the output value (SVO2) of the oxygen sensor (see broken line) are shown.

  First, at the time t1, when the introduction of secondary air (AI), FC control, or deceleration leaning is started, the actual air-fuel ratio SVO2 indicated by the output value of the oxygen sensor 52 shifts to the lean side, Along with this, the oxygen storage amount of the catalyst device 50 gradually increases.

  At the stage where the oxygen storage amount of the catalyst reaches the maximum, the introduction of secondary air (AI), FC control, or deceleration leaning is completed, and at this end time t2, rich spike control, that is, the catalyst device 50 Residual oxygen treatment is performed. In this rich spike control, from time t2 to time t3, normally, fuel at a level that is a very rich air-fuel ratio with respect to stoichiometry is injected in a combustible region (high concentration injection period Tb1). Therefore, in this high concentration injection period Tb1, the oxygen storage amount of the catalyst device 50 decreases rapidly. When the output value SVO2 of the oxygen sensor 52 starts to gradually increase toward the stoichiometric level, the fuel injection amount per unit time is gradually reduced. From time t3 to time t4, normally, fuel at a level that is higher than the stoichiometric air-fuel ratio is injected within a range that does not affect the output (low concentration injection period Tb2). Therefore, in this low concentration injection period Tb2, the oxygen storage amount of the catalyst device 50 gradually decreases compared to the high concentration injection period Tb1. When the output value SVO2 of the oxygen sensor 52 gradually approaches the stoichiometric level, the fuel injection amount per unit time is gradually reduced.

  Then, at the stage where the output value SVO2 of the oxygen sensor 52 has shifted to the rich side, the rich spike control ends, and the feedback control in the air-fuel ratio control unit main body 100 is resumed.

  The prediction accuracy determination unit 146 (see FIG. 8) of the air-fuel ratio control unit main body 100 that has resumed control determines whether or not the prediction accuracy in the predictor 102 has decreased in step S13 of FIG. That is, it is determined whether or not the state in which the moving average of the prediction error ERPRE (k) after the filtering process is higher than a preset threshold value continues for a set number of times. If the prediction accuracy in the predictor 102 is lowered, the process proceeds to step S14, where the control unit 126 temporarily stops the control in the first sliding mode control unit 104 and includes the second sliding mode / PID control unit 124. , Select the PID controller. At this time, the switching unit 128 switches to the output on the second sliding mode / PID control unit 124 side by supplying the switching instruction signal Sd from the control unit 126. The PID controller feeds back so that an error between the actual air-fuel ratio (SVO2) and a preset target value becomes zero. As a result, the output from the second sliding mode / PID control unit 124 is supplied to the multiplier 120 via the switching unit 128. As a result, the basic fuel injection amount TIMB is corrected by the output from the second sliding mode / PID control unit 124 (in this case, the PID control unit) and the environment correction coefficient KECO including the water temperature, the intake air temperature, the atmospheric pressure, and the like. The fuel injection time Tout is output and the actual air-fuel ratio SVO2 changes toward the target value.

  In step S13, when it is determined that the prediction accuracy is secured, the process proceeds to step S15, and the control unit 126 resumes the control in the first sliding mode control unit 104. That is, the suspension of the first sliding mode control unit 104 is released, and the parameters of the identifier 106 are reset to initial values. Further, the supply of the switching instruction signal Sd to the switching unit 128 is stopped. The switching unit 128 switches to the output on the first sliding mode control unit 104 side. Thus, the basic fuel injection amount TIMB is corrected by the output KO2 (k) from the first sliding mode control unit 104 and the environmental correction coefficient KECO including the water temperature, the intake air temperature, the atmospheric pressure, etc., and is output as the fuel injection time Tout. Is done.

  Thereafter, when the feedback control in the first sliding mode control unit 104 is restarted in the above-described step S15, or when it is determined in step S7 that the vehicle is not decelerating and leaning (that is, fuel cut control even when secondary air is introduced) However, if the engine is not decelerating and leaning), the process proceeds to step S16, and it is determined whether or not there is a request for termination of the air-fuel ratio control device 10 (power cut, maintenance request, etc.). If there is no termination request, the processing from step S1 is repeated, and the processing operation in the air-fuel ratio control device 10 is terminated when the termination request is made.

  As described above, in the air-fuel ratio control apparatus 10 according to the present embodiment, even in the system in which the secondary air introduction apparatus 1000 is installed, after the secondary air introduction, the air-fuel ratio control unit main body 100 once. By interrupting the sliding mode control and performing rich injection control at the stage where the introduction of the secondary air is completed, the amount of oxygen stored in the catalyst due to the introduction of the secondary air can be reduced at an early stage. As a result, it is possible to quickly return to the interrupted sliding mode control and to maintain high emission performance.

  By the way, when the fuel injection is stopped and controlled by the FC control unit 1004 by closing the throttle opening, air containing oxygen itself flows into the catalyst, so that the oxygen storage amount of the catalyst device 50 increases. . However, in the air-fuel ratio control apparatus 10 according to this embodiment, after the fuel injection stop control by the FC control unit 1004, the sliding mode control in the air-fuel ratio control unit main body 100 is temporarily interrupted to stop the fuel injection. Since rich injection control is performed at the stage when the control is completed, the amount of oxygen stored in the catalyst device 50 due to the introduction of air containing oxygen accompanying the stop control of fuel injection can be reduced early. As a result, it is possible to quickly return to the interrupted sliding mode control and to maintain high emission performance.

  Further, when the deceleration leaning is started, the oxygen storage amount of the catalyst increases because the oxygen-containing air itself flows into the catalyst as in the case of the fuel cut control described above. However, according to the air-fuel ratio control apparatus 10 according to this embodiment, when the deceleration lean control is performed by the deceleration lean control unit 1005, the feedback control is temporarily interrupted, and the rich injection control is performed at the stage where the deceleration lean conversion is completed. Therefore, the amount of oxygen stored in the catalyst by introducing the air containing oxygen accompanying the deceleration and leaning can be reduced at an early stage. As a result, it is possible to quickly return to the interrupted feedback control, and to maintain high emission performance.

  In addition, when it is predicted that it will take time for convergence of the sliding mode control due to the introduction of secondary air or the introduction of air containing oxygen accompanying fuel cut control or deceleration leaning, the introduction of secondary air The convergence of the feedback control can be accelerated by first performing the PID control at the stage where the fuel cut control or deceleration leaning is finished.

  If the prediction error is equal to or less than the threshold value, the convergence of the sliding mode control is ensured. Therefore, the emission performance can be maintained high by restarting the feedback control.

  In particular, when a decrease in prediction accuracy is determined when the control in the air-fuel ratio control unit main body 100 is resumed, an error between the actual air-fuel ratio and a preset target value without using the predictor 102 is determined. Since the PID control is performed so as to make zero, the time until the prediction accuracy is ensured can be shortened and the emission performance can be maintained higher than when the predictor 102 is used. it can.

  Next, the second air introduction corresponding portion 1008B will be described with reference to FIGS.

  As shown in FIG. 21, the second air introduction support unit 1008B has substantially the same configuration as the first air introduction support unit 1008A described above, but further includes an error calculation unit 1018, a PID control unit 1020, and a switching function. It differs in having a portion 1022. In addition, the processes in the air-fuel ratio control stop request unit 1010 and the restart determination unit 1014 described above are partially different.

  That is, the air-fuel ratio control stop request unit 1010 receives the AI start signal Sais from the AI control unit 1002, the FC control start signal Sfcs from the FC control unit 1004, or the deceleration lean start signal Srss from the deceleration lean control unit 1005. Based on the input, a stop request signal Sg is output to the air-fuel ratio control unit main body 100, and a first switching signal Sh1 is output to the switching unit 1022. The air-fuel ratio control unit main body 100 temporarily stops the air-fuel ratio control based on the input of the stop request signal Sg from the air-fuel ratio control stop request unit 1010. The switching unit 1022 switches to the output from the PID control unit 1020 based on the input of the first switching signal Sh1.

  The error calculator 1018 calculates an error ERR between the current output value (SVO2) of the oxygen sensor 52 and a preset target value.

  The PID control unit 1020 performs PID control (feedback control) so that the error ERR calculated by the error calculation unit 1018 becomes 0 (zero). The output of the PID control unit 1020 is input to the multiplier 120 via the switching unit 1022 as the target air-fuel ratio KO2 (k). Thus, the basic fuel injection amount TIMB is corrected by the output from the PID control unit 1020 and the environment correction coefficient KECO including the water temperature, the intake air temperature, the atmospheric pressure, etc., and is output as the fuel injection time Tout, and the output of the oxygen sensor 52 The value SVO2 changes toward the target value.

The restart determination unit 1014 compares the error ERR from the error calculation unit 1018 with a preset threshold value, and when the error ERR becomes equal to or less than the threshold value, the restart determination unit 1014 sends a control restart signal Si to the air-fuel ratio control unit main body 100. And the second switching signal Sh2 is output to the switching unit 1022. The threshold value may be the same as the threshold value used in the prediction accuracy determination unit 146 of the air-fuel ratio control unit main body 100. The air-fuel ratio control unit main body 100 restarts the air-fuel ratio control based on the input of the control restart signal Si from the restart determination unit 1014 , and the switching unit 1022 is based on the input of the second switching signal Sh2. Switch to 100 output. As a result, the output from the air-fuel ratio control unit main body 100 is input to the multiplier 120 via the switching unit 1022 as the target air-fuel ratio KO2 (k). That is, the basic fuel injection amount TIMB is corrected by the output from the air-fuel ratio control unit main body 100 and the environment correction coefficient KECO including the water temperature, the intake air temperature, the atmospheric pressure, etc., and is output as the fuel injection time Tout.

  Next, the processing operation of the second air introduction corresponding unit 1008B will be described with reference to the flowchart of FIG.

  First, in step S101 of FIG. 22, the air-fuel ratio control stop request unit 1010 determines whether or not introduction of secondary air has been started. This determination is made based on whether or not an AI start signal Sais is input from the AI control unit 1002.

  When the AI start signal Sais is input, the process proceeds to the next step S102, where the air-fuel ratio control stop request unit 1010 outputs a stop request signal Sg to the air-fuel ratio control unit body 100, and the switching unit 1022 performs the first switching. The signal Sh1 is output. The air-fuel ratio control unit main body 100 temporarily stops the air-fuel ratio control based on the input of the stop request signal Sg. The switching unit 1022 switches to the output from the PID control unit 1020 based on the input of the first switching signal Sh1.

  Thereafter, in step S103, the rich spike control unit 1012 waits for the end of the introduction of the secondary air. That is, it waits until an AI end signal Saie is input from the AI control unit 1002.

  On the other hand, if it is determined in step S101 that there is no input of the AI start signal Sais, in step S104, the air-fuel ratio control stop request unit 1010 determines whether FC control has been started. This determination is made based on whether or not an FC control start signal Sfcs is input from the FC control unit 1004.

  When the FC control start signal Sfcs is input, the process proceeds to the next step S105, where the air-fuel ratio control stop request unit 1010 outputs the stop request signal Sg to the air-fuel ratio control unit body 100, and the switching unit 1022 receives the first request. A switching signal Sh1 is output.

  Thereafter, in step S106, the rich spike control unit 1012 waits for the end of the FC control. That is, it waits until the FC control end signal Sfce is input from the FC control unit 1004.

  If it is determined in step S104 that the FC control start signal Sfcs is not input, the air-fuel ratio control stop request unit 1010 determines in step S107 whether deceleration leaning has started. This determination is made based on whether or not a deceleration leaning start signal Srss is input from the deceleration lean control unit 1005.

  When the deceleration lean start signal Srss is input, the process proceeds to the next step S108, and the air-fuel ratio control stop request unit 1010 outputs the stop request signal Sg to the air-fuel ratio control unit body 100. The air-fuel ratio control unit main body 100 temporarily stops the air-fuel ratio control based on the input of the stop request signal Sg.

  Thereafter, in step S109, the rich spike control unit 1012 waits for completion of deceleration leaning. That is, it waits until the deceleration leaning end signal Srse is input from the deceleration lean control unit 1005.

  If the AI end signal Saie is input in step S103, the FC control end signal Sfce is input in step S106, or the deceleration leaning end signal Srse is input in step S109, In step S110, the rich spike control unit 1012 controls the fuel injection valve 40 to perform a rich spike in the combustion chamber of the engine 28. In this rich spike control, the fuel supply amount corresponding to the current output value (SVO2) of the oxygen sensor 52 is supplied to the combustion chamber of the engine 28 per unit time while referring to the rich spike map 1016. This rich spike control is performed until it is determined in step S111 that the output value (SVO2) of the oxygen sensor 52 has become rich.

  Then, when the output value of the oxygen sensor 52 has shifted to the rich side, the process proceeds to step S112, and the error calculation unit 1018 obtains the difference (error) between the current output value SVO2 of the oxygen sensor 52 and the target value.

  Thereafter, in step S113, the restart determination unit 1014 determines whether or not to restart the control in the air-fuel ratio control unit main body 100. This determination is made as to whether or not error ≦ threshold. If the error exceeds the threshold value, the process proceeds to step S114, and the PID control unit 1020 performs PID control so that the error becomes 0 (zero). The output of the PID control unit 1020 is input to the multiplier 120 via the switching unit 1022 as the target air-fuel ratio KO2 (k). Thus, the basic fuel injection amount TIMB is corrected by the output from the PID control unit 1020 and the environment correction coefficient KECO including the water temperature, the intake air temperature, the atmospheric pressure, etc., and is output as the fuel injection time Tout, and the output of the oxygen sensor 52 The value SVO2 changes toward the target value.

  In step S113, when it is determined that the error is equal to or less than the threshold value, the process proceeds to the next step S115, and the restart determination unit 1010 outputs a control restart signal Si to the air-fuel ratio control unit main body 100, and the switching unit 1022 The second switching signal Sh2 is output. The air-fuel ratio control unit main body 100 restarts the air-fuel ratio control based on the input of the control restart signal Si from the restart determination unit 1010, and the switching unit 1022 is based on the input of the second switching signal Sh2. Switch to 100 output. As a result, the output from the air-fuel ratio control unit main body 100 is input to the multiplier 120 via the switching unit 1022 as the target air-fuel ratio KO2 (k). That is, the basic fuel injection amount TIMB is corrected by the output from the air-fuel ratio control unit main body 100 and the environment correction coefficient KECO including the water temperature, the intake air temperature, the atmospheric pressure, etc., and is output as the fuel injection time Tout.

  After that, when the feedback control in the first sliding mode control unit 104 is resumed in the above-described step S115, or when it is determined in step S107 that the deceleration leaning is not achieved (that is, fuel cut control even when secondary air is introduced). However, if the engine is not decelerating and leaning), the process proceeds to step S116, and it is determined whether or not there is a request for termination of the air-fuel ratio control device 10 (power cut, maintenance request, etc.). If there is no termination request, the processing from step S101 is repeated, and the processing operation in the air-fuel ratio control device 10 is terminated at the stage where the termination request is made.

  Thus, in the air-fuel ratio control apparatus 10 having the second air introduction corresponding unit 1008B, the dedicated PID control unit 1020 is used without using the second sliding mode / PID control unit 124 of the air-fuel ratio control unit main body 100. Since the PID control is performed so that the difference (error) between the output value SVO2 of the oxygen sensor 52 and the target value becomes zero, a prediction accuracy determination routine (predictor 102 → Without passing through the time adjustment unit 130 → the filter processing unit 144 in the adaptive model corrector 122 → the prediction accuracy determination unit 146), the error can be converged below the threshold value early, and the first air introduction corresponding unit 1008A The normal sliding mode control (control by the first sliding mode control unit 104) can be resumed earlier than when using the It is possible to maintain the emissions performance high.

  Further, the second air introduction corresponding unit 1008B can be applied to a configuration in which the second sliding mode / PID control unit 124 is not used, like the air-fuel ratio control unit main body 100d according to the fourth modification shown in FIG. It is versatile and can be applied to various specifications.

  In addition, the air-fuel ratio control apparatus according to the present invention is not limited to the above-described embodiment, and it is needless to say that various configurations can be adopted without departing from the gist of the present invention.

DESCRIPTION OF SYMBOLS 10 ... Air-fuel ratio control apparatus 12 ... Motorcycle 28 ... Engine 30 ... Intake pipe 32 ... Exhaust pipe 38 ... Throttle valve 40 ... Fuel injection valve 44 ... Throttle sensor 48 ... PB sensor 50 ... Catalyst apparatus 52 ... Oxygen sensor 62 ... ECU
DESCRIPTION OF SYMBOLS 100 ... Air-fuel ratio control part main body 102 ... Predictor 104 ... 1st sliding mode control part 106 ... Identifier 108 ... Air-fuel ratio reference value calculation part 110 ... LAF sensor 116 ... Basic fuel injection amount calculation part 118 ... Basic fuel injection map 118a ... first basic fuel injection map 118b ... second basic fuel injection map 122 ... adaptive model corrector 124 ... second sliding mode / PID control unit 126 ... control unit 128 ... switching unit 140 ... selection map 142 ... map selection unit 144 ... Filter processing unit 146 ... Prediction accuracy determination unit 148a ... First correction amount calculation unit 148b ... Second correction amount calculation unit 150a ... First correction coefficient calculation unit 150b ... Second correction coefficient calculation unit 152 ... Weighting unit 154 ... Sliding mode Control unit 1000 ... secondary air introduction device 1002 ... AI control unit 1004 ... FC control Unit 1005 ... deceleration lean control unit 1006 ... air-fuel ratio control unit 1008A ... first air introduction corresponding unit 1008B ... second air introduction corresponding unit 1010 ... air-fuel ratio control stop request unit 1012 ... rich spike control unit 1014 ... restart determination unit 1018 ... Error calculation unit 1020... PID control unit

Claims (7)

Secondary air that introduces secondary air into at least the engine (28), the catalyst (50) installed in the exhaust pipe (32) of the engine (28), and the exhaust path upstream of the catalyst (50). Installed in a vehicle (12) having an introduction device (1000) and controlling the fuel injection amount to the engine (28) to control the air-fuel ratio of the exhaust gas of the engine (28) to an appropriate air-fuel ratio. An air-fuel ratio control device,
An air-fuel ratio detecting means (52) provided downstream of the catalyst (50 ) and detecting an actual air-fuel ratio (SVO2) downstream of the catalyst (50);
A basic fuel injection amount calculation unit (116) for determining a basic fuel injection amount for the engine (28) based on at least a preset basic fuel injection map (118) and parameters of engine speed and throttle opening;
An air-fuel ratio control unit main body (100) for determining a target air-fuel ratio (KO2) for the basic fuel injection amount based on at least the actual air-fuel ratio (SVO2 ) from the air-fuel ratio detection means (52);
An air introduction corresponding part (1008A, 1008B) operating in accordance with the introduction of the secondary air by the secondary air introduction device (1000),
The air-fuel ratio control unit body (100)
Based on at least the actual air-fuel ratio (SVO2), the catalyst after a dead time (dt) corresponding to the distance from the fuel injection valve (40) of the engine (28) to the air-fuel ratio detection means (52) ( 50) an air-fuel ratio prediction means (102) for predicting an air-fuel ratio downstream of the air-fuel ratio and outputting it as a predicted air-fuel ratio (DVPRE);
Correction coefficient calculation means for determining a first correction coefficient (DKO2OP) for obtaining the target air-fuel ratio (KO2) based on the predicted air-fuel ratio (DVPRE) from the air-fuel ratio prediction means (102) by sliding mode control. (104)
This is a difference between an output (DPRE (k−dt)) obtained by delaying the predicted air / fuel ratio (DVPRE) from the air / fuel ratio predicting means (102) by the dead time (dt) and the actual air / fuel ratio (SVO2). Adaptive model correction means (122) for generating a second correction coefficient (KTIMB) for making the prediction error (ERPRE) zero ,
The air-fuel ratio predicting means (102) includes the actual air-fuel ratio (SVO2) from the air-fuel ratio detecting means (52), the first correction coefficient (DKO2OP) from the correction coefficient calculating means (104), and the adaptation. Based on the corrected air-fuel ratio obtained using the second correction coefficient (KTIMB) from the model correcting means (122) , the air-fuel ratio downstream of the catalyst (50) is predicted,
The air introduction corresponding part (1008A, 1008B)
Means (1010) for interrupting the sliding mode control in the correction coefficient calculation means (104) when the introduction of the secondary air is started;
Immediately after the introduction of the secondary air is finished, means (1012) for making the air-fuel ratio control open loop rich control;
An air-fuel ratio control apparatus comprising means (1014) for restarting the sliding mode control in the correction coefficient calculation means (104) when the actual air-fuel ratio (SVO2) becomes rich. At least the engine (28), the catalyst (50) installed in the exhaust pipe (32) of the engine (28), and the throttle valve (38) installed in the intake pipe (30) of the engine (28) are closed. And a fuel cut control means (1004) for stopping control of fuel injection during a period of time, and controlling the fuel injection amount to the engine (28) to control the engine (28) An air-fuel ratio control apparatus for controlling the air-fuel ratio of the exhaust gas to an appropriate air-fuel ratio,
An air-fuel ratio detecting means (52) provided downstream of the catalyst (50 ) and detecting an actual air-fuel ratio (SVO2) downstream of the catalyst (50);
A basic fuel injection amount calculation unit (116) for determining a basic fuel injection amount for the engine (28) based on at least a preset basic fuel injection map (118) and parameters of engine speed and throttle opening;
An air-fuel ratio control unit main body (100) for determining a target air-fuel ratio (KO2) for the basic fuel injection amount based on at least the actual air-fuel ratio (SVO2 ) from the air-fuel ratio detection means (52);
An air introduction corresponding part (1008A, 1008B) that operates in accordance with the fuel injection stop control by the fuel cut control means (1004),
The air-fuel ratio control unit body (100)
Based on at least the actual air-fuel ratio (SVO2), the catalyst after a dead time (dt) corresponding to the distance from the fuel injection valve (40) of the engine (28) to the air-fuel ratio detection means (52) ( 50) an air-fuel ratio prediction means (102) for predicting an air-fuel ratio downstream of the air-fuel ratio and outputting it as a predicted air-fuel ratio (DVPRE);
Correction coefficient calculation means for determining a first correction coefficient (DKO2OP) for obtaining the target air-fuel ratio (KO2) based on the predicted air-fuel ratio (DVPRE) from the air-fuel ratio prediction means (102) by sliding mode control. (104)
This is a difference between an output (DPRE (k−dt)) obtained by delaying the predicted air / fuel ratio (DVPRE) from the air / fuel ratio predicting means (102) by the dead time (dt) and the actual air / fuel ratio (SVO2). Adaptive model correction means (122) for generating a second correction coefficient (KTIMB) for making the prediction error (ERPRE) zero ,
The air-fuel ratio predicting means (102) includes the actual air-fuel ratio (SVO2) from the air-fuel ratio detecting means (52), the first correction coefficient (DKO2OP) from the correction coefficient calculating means (104), and the adaptation. Based on the corrected air-fuel ratio obtained using the second correction coefficient (KTIMB) from the model correcting means (122) , the air-fuel ratio downstream of the catalyst (50) is predicted,
The air introduction corresponding part (1008A, 1008B)
Means (1010) for interrupting the sliding mode control in the correction coefficient calculation means (104) when the fuel injection stop control by the fuel cut control means (1004) is started;
Immediately after the fuel injection stop control by the fuel cut control means (1004) is completed, means (1012) for making the air-fuel ratio control open loop rich control;
An air-fuel ratio control apparatus comprising means (1014) for restarting the sliding mode control in the correction coefficient calculation means (104) when the actual air-fuel ratio (SVO2) becomes rich. Installed in a vehicle (12) having at least an engine (28), a catalyst (50) installed in an exhaust pipe (32) of the engine (28), and a deceleration lean control means (1005). 28) an air-fuel ratio control device for controlling the air-fuel ratio of the exhaust gas of the engine (28) to an appropriate air-fuel ratio by controlling the fuel injection amount to 28),
An air-fuel ratio detecting means (52) provided downstream of the catalyst (50 ) and detecting an actual air-fuel ratio (SVO2) downstream of the catalyst (50);
A basic fuel injection amount calculation unit (116) for determining a basic fuel injection amount for the engine (28) based on at least a preset basic fuel injection map (118) and parameters of engine speed and throttle opening;
An air-fuel ratio control unit main body (100) for determining a target air-fuel ratio (KO2) for the basic fuel injection amount based on at least the actual air-fuel ratio (SVO2 ) from the air-fuel ratio detection means (52);
An air introduction corresponding portion (1008A, 1008B) that operates in accordance with the introduction of air containing oxygen by the deceleration lean control means (1005),
The air-fuel ratio control unit body (100)
Based on at least the actual air-fuel ratio (SVO2), the catalyst after a dead time (dt) corresponding to the distance from the fuel injection valve (40) of the engine (28) to the air-fuel ratio detection means (52) ( 50) an air-fuel ratio prediction means (102) for predicting an air-fuel ratio downstream of the air-fuel ratio and outputting it as a predicted air-fuel ratio (DVPRE);
Correction coefficient calculation means for determining a first correction coefficient (DKO2OP) for obtaining the target air-fuel ratio (KO2) based on the predicted air-fuel ratio (DVPRE) from the air-fuel ratio prediction means (102) by sliding mode control. (104)
This is a difference between an output (DPRE (k−dt)) obtained by delaying the predicted air / fuel ratio (DVPRE) from the air / fuel ratio predicting means (102) by the dead time (dt) and the actual air / fuel ratio (SVO2). Adaptive model correction means (122) for generating a second correction coefficient (KTIMB) for making the prediction error (ERPRE) zero ,
The air-fuel ratio predicting means (102) includes the actual air-fuel ratio (SVO2) from the air-fuel ratio detecting means (52), the first correction coefficient (DKO2OP) from the correction coefficient calculating means (104), and the adaptation. Based on the corrected air-fuel ratio obtained using the second correction coefficient (KTIMB) from the model correcting means (122) , the air-fuel ratio downstream of the catalyst (50) is predicted,
The air introduction corresponding part (1008A, 1008B)
Means (1010) for interrupting the sliding mode control in the correction coefficient calculation means (104) with the introduction of air containing oxygen by the deceleration lean control means (1005);
Immediately after the introduction of the oxygen-containing air by the deceleration lean control means (1005) is completed, means (1012) for making the air-fuel ratio control open loop rich control;
An air-fuel ratio control apparatus comprising means (1014) for restarting the sliding mode control in the correction coefficient calculation means (104) when the actual air-fuel ratio (SVO2) becomes rich. The air-fuel ratio control apparatus according to any one of claims 1 to 3,
The air-fuel ratio control unit body (100) includes a PID control unit (124), and prediction accuracy determination means (146) that determines prediction accuracy based on the prediction error (ERPRE).
The PID control unit (124) sets the actual air-fuel ratio (SVO2) when the prediction accuracy determining means (146) determines that the prediction accuracy is lowered prior to resuming the sliding mode control. An air-fuel ratio control apparatus that performs PID control so that an error from a preset target value becomes zero and controls the air-fuel ratio on the downstream side of the catalyst (50) to converge to the target value. . The air-fuel ratio control apparatus according to claim 4,
The air-fuel ratio control unit body (100)
An air-fuel ratio control apparatus that switches from the PID control to the sliding mode control when the prediction accuracy determination means (146) determines that the prediction accuracy is secured. The air-fuel ratio control apparatus according to any one of claims 1 to 3,
The air-fuel ratio control unit body (100) includes a PID control unit (124), and a control unit (126) that controls at least the correction coefficient calculation means (104) and the PID control unit (124).
The adaptive model correction means (122) has a prediction accuracy determination means (146) for determining a prediction accuracy based on the prediction error (ERPRE),
The control unit (126)
When the sliding mode control by the correction coefficient calculating unit (104) is resumed, the sliding mode control is temporarily stopped when the prediction accuracy determining unit (146) determines that the prediction accuracy is lowered. And activate the PID control unit (124),
The PID control unit (124)
Without using the air-fuel ratio predicting means (102), PID control is performed so that the error between the actual air-fuel ratio (SVO2) and a preset target value becomes zero, and the downstream side of the catalyst (50) The air-fuel ratio control apparatus controls the air-fuel ratio of the engine so as to converge to the target value. The air-fuel ratio control apparatus according to any one of claims 1 to 3,
The air introduction corresponding part (1008B)
Furthermore, it has a PID control unit (1020),
Prior to resuming the sliding mode control, if the error between the actual air-fuel ratio (SVO2) and a preset target value exceeds a preset threshold, the PID control unit (1020) Then, PID control is performed so that the error becomes zero, and the air-fuel ratio on the downstream side of the catalyst (50) is controlled to converge to the target value.

Images