JP2007009710A - Deterioration diagnostic device for linear air-fuel ratio sensor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To certainly detect one-sided deterioration of a linear air-fuel ratio sensor. <P>SOLUTION: When a predetermined diagnostic condition is satisfied, impulse disturbances are alternately output to the lean and rich sides of a feedback control system a predetermined number of times, and determining parameter T<SB>LR</SB>, T<SB>RL</SB>for deterioration determination are calculated based on the differential value of output of the linear air-fuel ratio sensor. Next, deterioration of the linear air-fuel ratio sensor is determined based on the difference DF<SB>T</SB>and sum AD<SB>T</SB>between/of rich side determining parameter T<SB>LR</SB>, T<SB>RL</SB>and lean side determining parameter T<SB>RL</SB>. The difference DF<SB>T</SB>and sum AD<SB>T</SB>are multiplied respectively by corresponding weighting factors α, β, and an evaluation value ES is calculated. The deterioration of the linear air-fuel ratio sensor is determined based on the evaluation value ES, thereby, so-called one-sided deterioration is detected. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a deterioration diagnosis apparatus for a linear air-fuel ratio sensor, and more particularly, to a linear air-fuel ratio sensor that is provided in an engine exhaust system and that detects deterioration of a linear air-fuel ratio sensor that outputs a value proportional to the oxygen concentration in exhaust gas. The present invention relates to a deterioration diagnosis device for a fuel ratio sensor.

Conventionally, as this type of linear air-fuel ratio sensor deterioration diagnosis device, for example, there is a technique disclosed in Patent Document 1. In the technique disclosed in Patent Document 1, during normal operation, air-fuel ratio feedback control is executed by PID operation, and during diagnosis, D operation of the feedback control system is prohibited and switched to PI operation. The output fluctuation of the linear air-fuel ratio sensor is enlarged, and the response deterioration of the linear air-fuel ratio sensor is enlarged and detected based on the fact that the response cycle becomes longer as the degree of sensor deterioration is larger.
Japanese Patent No. 3377336

  By the way, there are two modes of response deterioration of the linear air-fuel ratio sensor: double-side deterioration and single-side deterioration. Both-side deterioration refers to response deterioration in which the responsiveness of sensor output changes to both rich and lean changes deteriorates. One-side deterioration refers to an air-fuel ratio from the lean side to the rich side, or from the rich side to the lean side. Is a response deterioration in which the responsiveness of the sensor output change is deteriorated with respect to only one of them.

  In both-side deterioration, the response of the output change of the linear air-fuel ratio sensor is deteriorated to both the lean side and the rich side. Therefore, although the response of the entire feedback is delayed, the average air-fuel ratio does not deviate from the target air-fuel ratio. On the other hand, when feedback control is performed on the air-fuel ratio based on the output of the linear air-fuel ratio sensor that has deteriorated on one side, there is a problem that the average air-fuel ratio deviates from the target air-fuel ratio.

  In this regard, in the apparatus disclosed in Patent Document 1, since the output fluctuation of the linear air-fuel ratio sensor is enlarged, it is easy to determine the deterioration of the linear air-fuel ratio sensor for both-side deterioration. However, it has not been possible to determine whether or not the linear air-fuel ratio sensor has deteriorated on one side simply by increasing the output fluctuation. For this reason, if the feedback control of the air-fuel ratio is continued based on the deterioration diagnosis of Patent Document 1, the problem that the average air-fuel ratio deviates from the original target air-fuel ratio (center air-fuel ratio) cannot be avoided.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a linear air-fuel ratio sensor deterioration diagnosis device that can reliably detect one-side deterioration of a linear air-fuel ratio sensor.

  In order to solve the above-described problems, the present invention provides a feedback control system that executes air-fuel ratio feedback control based on the oxygen concentration in the exhaust gas of an engine, and the feedback control system. A linear air-fuel ratio sensor that outputs a value proportional to the oxygen concentration, and a disturbance generating means that outputs an impulse-like disturbance alternately to the lean side and the rich side to the feedback control system a predetermined number of times when a predetermined diagnosis condition is satisfied And at least one of a dead time and a time constant of the linear air-fuel ratio sensor based on the differential value output by the differentiating means based on the differential value output from the linear air-fuel ratio sensor after the disturbance output by the disturbance generating means. As a determination parameter for deterioration determination, determination parameter calculation means for calculating for each of the rich side and the lean side, and the calculated rich side determination Deterioration diagnosis of a linear air-fuel ratio sensor, comprising: an emission deterioration estimating means for estimating an emission deterioration associated with deterioration of a linear air-fuel ratio sensor based on a difference and a sum of determination parameters on a lean side and a lean side Device. In this aspect, when a predetermined diagnostic condition is satisfied, the disturbance generating means outputs an impulse-like disturbance to the feedback control system, so that the actual air-fuel ratio corresponding to the disturbance is fed back from the linear air-fuel ratio sensor, At least one of a dead time or a time constant is detected from the differential value of the output of the linear air-fuel ratio sensor, and a deterioration diagnosis is executed by detecting a change in a transient characteristic (dead time or first-order delay) of the linear air-fuel ratio sensor. It becomes possible. For this reason, it becomes possible to analyze not only the degradation of the linear air-fuel ratio sensor itself but also the degradation mode in detail. Moreover, since the deterioration of the linear air-fuel ratio sensor is determined based on the difference and sum of the rich-side determination parameter and the lean-side determination parameter, the linear air-fuel ratio sensor is deteriorated not only on both sides but also on one side. As a result, it is possible to determine whether or not the air-fuel ratio is being controlled. As a result, the average air-fuel ratio does not deviate from the target air-fuel ratio. That is, in the case of one-side deterioration of the linear air-fuel ratio sensor, the difference between the rich-side determination parameter and the lean-side determination parameter increases. Therefore, if this difference is large, it means that the center air-fuel ratio has shifted to the lean side or the rich side, and it can be estimated that the emission has deteriorated due to the shift. On the other hand, when both sides of the linear air-fuel ratio sensor deteriorate, the sum of both determination parameters becomes large. Therefore, if this sum is large, it means that an air-fuel ratio shift has occurred due to the divergence of the feedback control system due to the extension of the period of the feedback control system, and that the emission has deteriorated due to the shift. Can be estimated.

  In a preferred aspect, the emission deterioration estimating means obtains a correction parameter by multiplying the difference and the sum by a predetermined weighting factor, respectively, and an evaluation value for obtaining an evaluation value by calculating the sum of the correction parameters A calculation step and a determination step for comparing the calculated evaluation value with a threshold value to determine pass / fail are executed. Normally, the gain of the feedback control system may be changed according to the operating state, but if the feedback gain is changed, the period of the feedback control system is affected. For example, if the feedback gain is reduced, the feedback responsiveness decreases, and even if the linear air-fuel ratio sensor is normal, the period of the feedback control system is extended. Will be estimated incorrectly. However, in this aspect, it is possible to easily execute a deterioration evaluation that comprehensively considers the center air-fuel ratio shift due to one-side deterioration of the linear air-fuel ratio sensor and the emission deterioration due to divergence of feedback control accompanying both-side deterioration.

  In a preferred aspect, the emission deterioration estimation means changes the sum weight coefficient so that the emission deterioration criterion decreases as the gain of the feedback control system decreases. In this aspect, it is possible to set an appropriate weighting factor according to the difference in the feedback gain and evaluate the emission based on the deterioration diagnosis.

  In a preferred aspect, the linear air-fuel ratio sensor is disposed upstream of a catalyst that purifies the exhaust gas of the engine, and the feedback control system receives the output of the linear air-fuel ratio sensor as well as the output thereof. A main control element that outputs an operation amount based on the difference between the target air-fuel ratio and the target value, and a main control element that is provided in the main control element and that performs learning for correcting a steady deviation between the target air-fuel ratio and the actual air-fuel ratio. Learning means, and the emission deterioration estimating means changes the weighting factor of the difference in the correction parameter so that the standard of emission deterioration is lowered as the learning degree of the main learning means progresses. Normally, in an air-fuel ratio feedback control system, based on the feedback correction amount calculated by the main control element, based on the output of the linear air-fuel ratio sensor, in order to correct the center air-fuel ratio deviation due to variations in fuel system parts and aging. Thus, a learning coefficient (learning value) is calculated to correct the deviation of the center air-fuel ratio. The learning coefficient is gradually updated from the viewpoint of preventing erroneous learning, and the difference in the degree of learning update affects the center air-fuel ratio shift. For example, when the learning update rate is low, the center air-fuel ratio shift cannot be sufficiently corrected, and even if the linear air-fuel ratio sensor is normal, the center air-fuel ratio shift becomes large. Otherwise, the emission deterioration will be estimated incorrectly. However, in this aspect, it is possible to set an appropriate weighting factor according to the degree of learning by the main learning means provided in the feedback control system, and to evaluate the emission based on the deterioration diagnosis.

  In a preferred aspect, the linear air-fuel ratio sensor is disposed upstream of a catalyst that purifies the exhaust gas of the engine, and the feedback control system is disposed downstream of the catalyst and purified by the catalyst. An oxygen concentration sensor that outputs the oxygen concentration in the exhaust gas as a detection value; a detection value from the oxygen concentration sensor; and a sub-control element that corrects a target value based on a sub-correction amount based on the detection value; A sub-learning unit that is provided in the sub-control element and executes learning for correcting a steady deviation between the target air-fuel ratio and the actual air-fuel ratio, and the emission deterioration estimating unit includes: The weighting factor of the difference in the correction parameter is changed so that the emission deterioration standard is lowered as the learning degree is advanced. Normally, in an air-fuel ratio feedback control system, a sub-control element is used based on the output of an oxygen concentration sensor provided downstream of the catalyst in order to correct a variation in fuel system parts and a center air-fuel ratio shift due to aging. Based on the calculated feedback correction amount, a learning coefficient (learning value) is calculated to correct the center air-fuel ratio shift. The learning coefficient is gradually updated from the viewpoint of preventing erroneous learning, and the difference in the degree of updating of the learning affects the center air-fuel ratio shift. For example, when the learning update rate is low, the center air-fuel ratio shift cannot be sufficiently corrected, and even if the linear air-fuel ratio sensor is normal, the center air-fuel ratio shift becomes large. Otherwise, the emission deterioration will be estimated incorrectly. However, even in this aspect, it is possible to set an appropriate weighting factor according to the degree of learning by the sub-learning means provided in the sub-control element of the feedback control system, and to evaluate the emission based on the deterioration diagnosis.

  As described above, in the present invention, the deterioration of the linear air-fuel ratio sensor is determined based on the difference and the sum of the determination parameter on the rich side and the determination parameter on the lean side. As a result of determining whether or not the linear air-fuel ratio sensor has deteriorated on one side, it is possible to prevent the deviation of the average air-fuel ratio from the target air-fuel ratio.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.

  FIG. 1 is a system diagram of an engine 10 according to an embodiment of the present invention.

  Referring to FIG. 1, an engine 10 according to the degradation determination device 1 of the present embodiment is provided with a plurality of cylinders 11, and pistons 12 connected to a crankshaft (not shown) are provided in each cylinder 11. Is inserted, and the combustion chamber 14 is formed above it. The engine 10 is provided with a rotation angle sensor SW1 that detects an engine rotation speed Ne of the crankshaft.

  In the cylinder head of the engine 10, an intake port 15 and an exhaust port 16 that open toward the combustion chamber 14 are formed for each cylinder 11, and an intake valve 17 and an exhaust port are provided in these ports 15 and 16, respectively. Each valve 18 is equipped.

  The intake port 15 is provided with an intake system 20, and the exhaust port 16 is provided with an exhaust system 30.

  The intake system 20 includes an air cleaner 21 for purifying intake air at the upstream end. An element 22 is built in the air cleaner 21. A throttle body 23 is provided on the downstream side of the air cleaner 21. The throttle body 23 is provided with a throttle valve 24 that adjusts the intake air amount Qa flowing through the intake system 20. An intake manifold 25 is provided on the downstream side of the throttle body 23, and a branch intake passage 26 provided at the downstream end of the intake manifold 25 is connected to the intake port 15 of the corresponding cylinder 11. In the illustrated example, a fuel injection valve 27 is provided in the branch intake passage 26. In the intake system 20, an air flow sensor SW <b> 2 is disposed between the air cleaner 21 and the throttle body 23. The air flow sensor SW2 outputs the intake air amount Qa of the intake air filtered by the element 22. Further, the throttle body 23 is provided with a throttle sensor SW3 for detecting the throttle opening TVO of the throttle valve 24.

  The exhaust system 30 includes an exhaust manifold 31 connected to the exhaust port 16, and a three-way catalyst 32 that is disposed downstream of the exhaust manifold 31 and purifies the burned gas discharged into the exhaust manifold 31. It has been. The exhaust system 30 is provided with a linear air-fuel ratio sensor SW4 disposed upstream of the three-way catalyst 32 and an oxygen concentration sensor SW5 disposed downstream. The linear air-fuel ratio sensor SW4 is for outputting a signal that is approximately proportional to the oxygen concentration from the burned gas. The oxygen concentration sensor SW5 is configured so that the output voltage changes suddenly at an oxygen concentration corresponding to the stoichiometric air-fuel ratio, and by detecting whether the oxygen concentration is higher or lower than the stoichiometric air-fuel ratio on and off, the air-fuel ratio is detected. The feedback control is executed. The linear air-fuel ratio sensor SW4 calculates an output corresponding to the actual air-fuel ratio of feedback control, whereas the oxygen concentration sensor SW5 calculates a detection value corresponding to the oxygen concentration of the burned gas after purification. Is. In this embodiment, the target air-fuel ratio of the engine is set to the theoretical air-fuel ratio (λ = 1) in principle.

  Each of the sensors SW1 to SW5 and the fuel injection valve 27 described above are connected to the control unit 100 to constitute an air-fuel ratio feedback control system.

  FIG. 2 is a control circuit block diagram of the degradation determination apparatus 1 according to the present embodiment, and FIG. 3 is a block diagram of the degradation determination apparatus 1 realized by the control circuit of FIG.

  First, referring to FIG. 2, the control unit 100 includes a CPU 101, an auxiliary storage device 102 embodied by ROM, and a main storage device 103 embodied by RAM. Each of the sensors SW1 to SW5 described above is connected to the CPU 101 and is configured to output the corresponding signals Ne, Qa, TVO, PF, and SF to the CPU 101, respectively.

  The CPU 101 processes the signals Ne, Qa, TVO, PF, SF output from the sensors SW1 to SW5 based on the program stored in the auxiliary storage device 102, and controls the fuel injection valve 27 to set the air-fuel ratio. It is configured to perform feedback control.

  The auxiliary storage device 102 stores a deterioration diagnosis program to be described later in detail.

  In the process of executing the program stored in the auxiliary storage device 102, the main storage device 103 receives signals Ne, Qa, TVO, PF, SF output from the sensors SW1 to SW5 and calculated values based on the signals Ne, Qa, TVO, PF, SF. It is configured to memorize.

  Referring to FIG. 3, control unit 100 constitutes feedback control system 110 shown in FIG. The feedback control system 110 includes a reference input element 111 having a target air-fuel ratio (λ = 1) as a target value DV, a BIAS correction element 112 for correcting the reference input IP output from the reference input element 111, and a BIAS correction element. And a main control element 114 that determines an operation amount OV to the engine 10 (more specifically, the fuel injection valve 27) based on the operation signal AS corrected to 112.

An output PF corresponding to the actual air-fuel ratio detected by the linear air-fuel ratio sensor SW4 is input between the BIAS correction element 112 and the main control element 114, and the main control element 114 receives the reference input. subtracting the correction amount SS of BIAS correction element 112 from a reference input IP elements 111, further receives the operation signal AS obtained by subtracting the output PF of the linear air-fuel ratio sensor SW4, predetermined transfer comprising a gain K _P function G _P (S ), The operation amount OV is output. In the present embodiment, the transfer function G _P (S) is determined so that the control operation of the main control element 114 is at least a PI operation.

Next, a sub control element 115 is connected between the reference input element 111 and the BIAS correction element 112. The sub control element 115 is configured to receive the detection value SF from the oxygen concentration sensor SW5 and output the sub correction amount SbS based on a predetermined transfer function G _S (S) including the gain K _S. . Therefore, the operation signal AS from which the sub correction amount SbS has been subtracted is input to the main control element 114. In this embodiment, the transfer function G _S (S) is determined so that the control operation of the sub control element 115 is at least a PI operation.

  Each control element 114, 115 is provided with a main learning means 114a and a sub learning means 115a, respectively.

  The main learning means 114a receives the operation signal AS and the operation amount OV output from the main control element by a low-pass filter, and learns the ratio between the base fuel amount based on the operation signal AS and the corrected fuel amount based on the operation amount OV. And a learning coefficient StP based on this learning value is output between the reference input element 111 and the main control element 114. The main learning means 114a is provided with an unillustrated SRAM that stores a learning coefficient StP output at a predetermined timing (for example, a predetermined number of times of fuel injection). In this embodiment, the SRAM captures the learning coefficient StP. At this time, the state quantity of the low-pass filter is cleared, and the I term of the main control element 114 is corrected according to the cleared state quantity of the low-pass filter. As a result, each time the operation amount OV of the main control element 114 is output, the main learning means 114a learns a steady deviation between the operation signal AS and the operation amount OV and reflects it in the main control element 114 by the learning coefficient StP. Is configured to do.

  Further, the sub-learning means 115a takes in the sub-correction amount SbS of the sub-control element 115, cuts the high-frequency component with a low-pass filter, and further uses the sub-learning amount integrated using an integration element such as SRAM as the reference input element 111 and the main input. It is comprised so that it may output between control elements 114. At this time, in the present embodiment, when the sub correction amount SbS is taken into the integration element from the low-pass filter, the state quantity of the low-pass filter is cleared, and the sub-control element according to the cleared state quantity of the low-pass filter. The I term of 115 is corrected. As a result, every time the sub correction amount SbS of the sub control element 115 is output, the sub learning means 115a learns a steady deviation between the operation signal AS and the sub correction amount SbS, and the sub control element 115 is determined by the learning coefficient StS. It is configured to reflect.

  Therefore, the CPU 101 constituting the feedback control system 110 can detect the learning degree of the corresponding control elements 114 and 115 by the learning coefficients StP and StS output from the learning units 114a and 115a.

  The transfer function of the low-pass filter of each learning means 114a, 115a may be, for example, a first-order lag element or a dead time + first-order lag element.

Further, the feedback control system 110 according to the present embodiment is functionally configured with disturbance generating means 116 that alternately generates disturbances LR and RL. This disturbance generating means 116 operates when a deterioration diagnosis of the linear air-fuel ratio sensor SW4 described below is performed by executing a program stored in the auxiliary storage device 102. The disturbance generating means 116 is configured to transiently change the air-fuel ratio to the rich side or the lean side by giving an impulse-like disturbance to the fuel injection amount. In the following description, the disturbance when changing the air-fuel ratio to the rich side is expressed as LR, and the disturbance when changing the air-fuel ratio to the lean side is expressed as RL. The number of occurrences N _LR and N _RL of the disturbances LR and RL output from the disturbance generation means 116 is stored in the main storage device 103, respectively. Then, the same number of disturbances LR and RL are alternately output for the number of output times N _END set in advance in the deterioration diagnosis program. As a result, the air-fuel ratio intentionally changed by the diagnosis is neutralized, and the air-fuel ratio controlled by the main control element 114 is prevented from being disturbed more than necessary, thereby preventing the emission from being lowered.

  4 and 5 are flowcharts of the deterioration diagnosis program in the present embodiment. FIG. 6 is a timing chart of signals obtained by executing the flowcharts of FIGS.

First, referring to FIG. 4 and FIG. 6, when the deterioration diagnosis program is executed, the CPU 101 waits for the diagnosis condition to be satisfied (step S1). Here, diagnostic conditions are
(1) The change amount of the engine rotation speed Ne detected by the rotation angle sensor SW1 is equal to or less than a predetermined change amount,
(2) The change amount of the throttle opening TVO detected by the throttle sensor SW3 is not more than a predetermined change amount, and (3) The change amount of the charging efficiency CE calculated by CE = Qa / Ne is not more than the predetermined change amount. It means that it is a so-called steady operation that satisfies all of the conditions.

  If the diagnostic conditions are not met, such as during acceleration, the system waits until the diagnostic conditions are met.If the diagnostic conditions are met, the process proceeds to the convergence determination threshold setting subroutine to determine the convergence threshold. Values ThC and dThC are set (step S2).

When the convergence determination threshold values ThC and dThC are set, the convergence determination is executed based on the convergence determination threshold values ThC and dThC (step S3). In this convergence determination, the fluctuation range OP of the output PF of the linear air-fuel ratio sensor SW4 is compared with the threshold value ThC as shown in FIG. 6 in a state where the disturbances RL and LR by the disturbance generating means 116 are reset. At the same time, the fluctuation range dOP of the differential value D _O2 and the threshold value dThC are compared, and it is determined that the air-fuel ratio has converged when the fluctuation ranges OP and dOP are both less than the corresponding threshold values ThC and dThC. By executing step S3, the CPU 101 functionally constitutes a convergence determination unit.

If it is determined that the air-fuel ratio has converged, the CPU 101 starts a timer (step S4), and waits for the timer time to become zero due to the countdown of the timer (step S5). Then, after starting the diagnosis after the timer time reaches zero, the disturbances LR and RL occurrence times N _LR and N _RL are compared (step S6). If N _RL > N _LR is satisfied, the disturbance LR Is output to the feedback control system 110 (step S7), and if not established, the disturbance RL is output to the feedback control system 110 (step S8). As a result, for example, as shown in FIG. 6, the disturbance LR is first output, which changes the output of the linear air-fuel ratio sensor SW4.

When the disturbance generating means 116 outputs the disturbance LR (or RL), the CPU 101 calculates a differential value D _O2 of the output PF of the linear air-fuel ratio sensor SW4 (step S9). Thereby, it becomes possible to grasp how the output PF of the linear air-fuel ratio sensor SW4 changes due to the disturbance LR (or RL). As described above, the CPU 101 of the present embodiment functionally configures a differentiating unit that outputs a differential value D _O2 obtained by differentiating the output PF of the linear air-fuel ratio sensor SW4 after the disturbance output by the disturbance generating unit 116.

  Here, the detected value of the linear air-fuel ratio sensor SW4 usually has a predetermined dead time L and a time constant τ.

  FIG. 7 is a graph showing the step response characteristics of the linear air-fuel ratio sensor SW4.

Referring to FIG. 7, between the input x (t) and the output y (t) of the linear air-fuel ratio sensor SW4,
y (t) = x (t−L) (L ≧ 0) (1)
The linear air-fuel ratio sensor SW4 serving as a dead time element follows a transfer function of the following equation.

G (s) = Y (s) / X (s) = e ^−LS (2)
However, Y (s): Complex function of output, X (s): Complex function of input Therefore, the step response h (t) of the linear air-fuel ratio sensor SW4 as a dead time element is expressed by the following equation by inverse Laplace transform. As shown in FIG.

h (t) = u (t−L) (3)
Where u (t): unit step function (1 (t ≧ 0) and 0 (t <0))
Therefore, in this embodiment, the system waits for the dead time L to elapse, detects the end of the dead time L, and resets the disturbances LR and RL. In order to execute such control, as shown in FIG. 4, the CPU 101 compares the calculated differential value ± D _O2 with a predetermined reset threshold value ± ThD, and + D _O2 > + ThD or −D _O2 <−. The system waits for the establishment of ThD (step S10), and if it is established, calculates a dead time L of the linear air-fuel ratio sensor SW4 (step S11) and resets the disturbance caused by the disturbance generating means 116 (step S12).

Next, the time constant τ is the transfer function of the first-order lag element
G (s) = Y (s) / X (s) = K / (1-τ) (4)
Is a constant.

Step response h (t) obtained by inverse Laplace transform from this equation (4)
h (t) = K (1−e ^{−T / M} ) (5)
However, from K and M: constants,
h (t) | _{τ = M} = 0.632K (6)
Therefore, if the characteristic of the dead time L is added to this, the time constant τ and the step response h (t) of the linear air-fuel ratio sensor SW4 have the relationship shown in FIG.

  As is apparent from FIG. 7, the larger the time constant τ, the slower the rise of the step response waveform, and the longer it takes to reach the final value. As the deterioration of the linear air-fuel ratio sensor SW4 progresses, the time constant τ becomes longer.

Referring to FIG. 5, in this embodiment, in order to incorporate the time constant τ as an element of deterioration diagnosis, the differential peak value of the output of the linear air-fuel ratio sensor SW4 (the peak value of the differential value ± D _O2 output by the CPU 101). ) D _O2PK is calculated (step S14), and the time constant τ of the linear air-fuel ratio sensor SW4 is calculated from the differential peak value D _O2PK (step S15).

When the calculation of the time constant τ is completed, the CPU 101 determines whether the disturbance generated by the disturbance generating means 116 is to reduce or increase the fuel (step S16). In the case of an increase, the determination parameters (calculated time delay L, time constant τ) are stored in the main storage device 103 as _LR (steps S17 and S18), and the number of occurrences N _LR stored in the main storage device 103 is stored. , N _RL is incremented (steps S19 and S20).

Thereafter, for each disturbance LR, RL, it is determined whether or not the required number of times N _END has been completed (steps S21 and S22), and any of the number of occurrences N _LR and N _RL is less than the required number of times N _END. In step S1, the process is repeated, and when both occurrences N _LR and N _RL have been completed, the process proceeds to the deterioration determination process.

  FIG. 8 is a flowchart showing details of the deterioration determination process.

With reference to FIG. 8, in this embodiment, in order to perform deterioration determination, the sum of the dead time L and the time constant τ is defined as the transition time T, and when the processing up to step S22 is completed, the CPU 101 calculates a difference DF _T the mean transit time T _LR, T _RL according to the disturbance RL disturbance LR and the lean side of the rich side (step S211). Here in this embodiment, the relationship between the deterioration degree of the computed difference DF _T and emissions, focusing on that there is a predetermined correlation according to the learning degree of the control elements 114 and 115, each learning means Based on the learning coefficients StP and StS output from 114a and 115a, the learning states of the main and sub control elements 114 and 115 are detected (step S212), and parameters for determining the deterioration determination of the linear air-fuel ratio sensor SW4 from the learning state The weighting coefficient α is corrected (step S213).

9, the average transition time period T _LR, is a graph showing the relationship between deterioration degree of the difference DF _T and emissions T _RL.

Referring to FIG. 9, when the learning degree of the control elements 114 and 115 is low, while the deterioration degree of the emission from a relatively small value difference DF _T increases proportionally, if the learning degree is high , deterioration degree of the emission will not increase unless become considerable large difference DF _T. Therefore, in the present embodiment, the learning state of the main and sub control elements 114 and 115 is detected in step S212, and the weighting coefficient α is calculated in step S213 according to the learning degree. Specifically, the weighting factor α is set so that the deterioration determination of the linear air-fuel ratio sensor SW4 becomes loose as the learning degree increases. As a result, in the present embodiment, it is possible to evaluate the emission based on the deterioration diagnosis according to the learning degree by the learning means 114a and 115a provided in the feedback control system 110.

Next, the CPU 101 calculates the sum AD _T of the average transition times T _LR and T _RL (step S214). Here in this embodiment, the relationship between the deterioration degree of the computed sum AD _T and emissions, focusing on that there is a predetermined correlation according to the gain K _P, K _S of each control element 114 The gains K _P and K _S of the control elements 114 and 115 are acquired (step S215), and the weighting factor β that is a parameter for determining the deterioration determination of the linear air-fuel ratio sensor SW4 is corrected according to the gains K _P and K _S. (Step S216).

FIG. 10 is a graph showing the relationship between the sum AD _T of the average transient times T _LR and T _RL and the degree of emission deterioration.

Referring to FIG. 10, when the gains K _P and K _{S of the} control elements 114 and 115 are large, the degree of emission deterioration increases from a relatively small value of the sum AD _T , whereas the gains K _P and K _S when is small, deterioration degree of emission if not sum AD _T is considerably larger will not increase. Therefore, in the present embodiment, the gains K _P and K _S of the main and sub control elements 114 and 115 are acquired in step S215, and the weighting coefficient β is calculated in step S216 according to the values. Specifically, the weighting coefficient β is set so that the deterioration determination of the linear air-fuel ratio sensor becomes gentle as the gains K _P and K _S become small. As a result, in the present embodiment, it is possible to evaluate the emission based on the deterioration diagnosis according to the difference between the feedback gains K _P and K _S.

After determining the average transition time T _LR, a differential DF _T and the sum AD _T of T _RL, as described above, CPU 101 is a calculation of the evaluation value ES executes the following equation (step S217).

ES = DF _T * α + AF _T * β (7)
Next, the evaluation value ES is compared with a predetermined threshold value ThES (step S218). If the evaluation value ES is equal to or less than the threshold value ThES, the determination is normal (step S219). If the evaluation value ES exceeds the threshold value ThES, It is configured to make a deterioration determination of the linear air-fuel ratio sensor SW, that is, an emission deterioration determination (step S220).

  FIG. 11 is a flowchart showing an example of the convergence determination threshold value setting subroutine (step S2) in FIG.

  Referring to FIG. 11, in this example, in this example, a map 220 for obtaining threshold values ThC and dThC from intake air amount Qa is stored in main memory 103 in advance, and is detected from airflow sensor SW2. The threshold values ThC and dThC are indexed from the intake air amount Qa (step S230), and the convergence determination threshold values ThC and dThC are set (step S231). In this embodiment, as the intake air amount Qa decreases, the convergence determination threshold values ThC and dThC can be set smaller to make the convergence determination stricter. As described above, in the embodiment of FIG. 11, the air flow sensor SW2 for detecting the intake air amount Qa of the engine 10 is provided, and the convergence determining means functionally configured by the CPU 101 makes the convergence condition stricter as the intake air amount Qa is smaller. As a result of correcting the determination criterion of the time constant τ accompanying the change in the intake air amount Qa, it is possible to avoid erroneous determination.

As described above, in the present embodiment, when a predetermined diagnosis condition is satisfied, the disturbance generating means 116 outputs the impulse-like disturbances LR and RL to the feedback control system 110, thereby corresponding to the disturbances LR and RL. results actual air-fuel ratio is fed back from the linear sensor SW4, and detecting at least one of the dead time L or time constant τ from the differential value D _O2 of the output PF of the linear air-fuel ratio sensor SW4, the linear air-fuel ratio sensor SW4 Deterioration diagnosis can be executed by detecting changes in transient characteristics (dead time L and first-order lag). For this reason, it becomes possible to analyze not only the degradation of the linear air-fuel ratio sensor SW4 but also the degradation mode in detail. Moreover, the rich side of the decision parameter (in this embodiment, the mean transit time T _LR) (in the present embodiment, the mean transit time T _RL) determined parameters of the lean side on the basis of the difference DF _T and the sum AD _T of, Since the deterioration of the linear air-fuel ratio sensor SW4 is determined, it is possible to determine not only the deterioration on both sides but also whether or not the linear air-fuel ratio sensor SW4 has deteriorated on one side. There is no deviation from the target air-fuel ratio. That is, in the case of one-side deterioration of the linear air-fuel ratio sensor SW4, the difference between the rich-side determination parameter and the lean-side determination parameter increases. Therefore, if this difference is large, it means that the center air-fuel ratio has shifted to the lean side or the rich side, and it can be estimated that the emission has deteriorated due to the shift. On the other hand, when both sides of the linear air-fuel ratio sensor SW4 deteriorate, the sum of both determination parameters becomes large. Therefore, if this sum is large, it means that the air-fuel ratio shift accompanying the divergence of the feedback control system 110 due to the extension of the cycle of the feedback control system 110 has occurred, and the emission deteriorates with the shift. Can be estimated.

In this embodiment also, to the difference DF _T and sum AD _T, each predetermined weighting factor alpha, obtained a correction step of obtaining a correction parameter by multiplying the beta, the sum calculated by the evaluation value ES of the correction parameters An evaluation value calculation step and a determination step of comparing the calculated evaluation value ES with a threshold value ThES to determine pass / fail are executed (steps S217 to S220). For this reason, in this embodiment, it is possible to easily execute a deterioration evaluation that comprehensively considers the center air-fuel ratio shift due to one-side deterioration of the linear air-fuel ratio sensor SW4 and the emission deterioration due to the divergence of feedback control accompanying both-side deterioration. become.

In the present embodiment, the linear air-fuel ratio sensor SW4 is disposed on the upstream side of the three-way catalyst 32, and the feedback control system 110 is disposed on the downstream side of the three-way catalyst 32. The oxygen concentration sensor SW5 that outputs the oxygen concentration in the exhaust gas purified by the catalyst 32 as the detection value SF and the output PF of the linear air-fuel ratio sensor SW4 are input, and the difference between the output PF and the target value DV And a detection value SF from the oxygen concentration sensor SW5 is input, and a sub-correction amount SbS is corrected based on the sub-correction amount SbS based on the detection value SF. Learning for correcting a steady deviation between the target air-fuel ratio and the actual air-fuel ratio, provided in the control element 115 and the main control element 114 and the sub-control element 115, respectively. The CPU 101 serving as the emission deterioration estimation means includes a main learning means 114a and a sub learning means 115a that execute the difference so that the emission deterioration standard decreases as the learning degree of the learning means 114a and 115a progresses. it is intended to change the weighting coefficient of DF _T α.

  The learning coefficients StP and StS of the main learning means 114a and the sub learning means 114b described above are gradually updated from the viewpoint of preventing erroneous learning. affect. For example, when the learning update rate is low, the center air-fuel ratio shift cannot be sufficiently corrected, and even if the linear air-fuel ratio sensor SW4 is normal, the center air-fuel ratio shift becomes large. If not taken into account, the emission deterioration will be estimated incorrectly.

  However, in the present embodiment, it is possible to set an appropriate weighting factor α in accordance with the degree of learning by the learning means 114a and 115a provided in the feedback control system 110, and to evaluate the emission based on the deterioration diagnosis.

In this embodiment also, CPU 101 serving as the emission deterioration estimating means, the gain K _P of the feedback control system 110, as K _S is small, changes the weighting coefficient β of the sum AD _T as a reference in the emission deterioration is reduced Is.

  Normally, the gain of the feedback control system 110 may be changed according to the operating state, but when the feedback gain is changed, the period of the feedback control system 110 is affected. For example, if the feedback gain is reduced, the feedback responsiveness is reduced, and the period of the feedback control system 110 is extended even if the linear air-fuel ratio sensor SW4 is normal. The emission deterioration will be estimated incorrectly.

However, in the present embodiment, it is possible to set an appropriate weighting factor β according to the difference between the feedback gains K _P and K _S and evaluate the emission based on the deterioration diagnosis.

As described above, in this embodiment, on the basis of the difference DF _T and the sum AD _T richer the determination parameter determination parameters (mean transit time T _LR) and lean (mean transit time T _RL), the linear air-fuel ratio Since the deterioration of the sensor SW4 is determined, it is possible to determine not only the deterioration on both sides but also whether or not the linear air-fuel ratio sensor SW4 has deteriorated on one side. As a result, the target air-fuel ratio of the average air-fuel ratio There is a remarkable effect that it is possible to prevent deviation from the above.

  The above-described embodiment is merely a preferred specific example of the present invention, and the present invention is not limited to the above-described embodiment. For example, when actualizing the deterioration diagnosis, the deterioration state may be determined by executing the dead time L or the time constant τ instead of the transient time T shown in FIG.

  Furthermore, the present invention can also be applied to a feedback control system having a learning function only in the main control element 114 or only in the sub-control element 115. In this case, the weighting factor α is changed according to the learning degree of the main control element 114 (or the learning degree of the sub control element 115).

  Further, in realizing steps S215 and S216 shown in FIG. 8, the transfer function of the entire feedback control system 110 is obtained and its gain is determined, and the weighting factor β is changed according to the overall gain. Good.

  It goes without saying that various modifications can be made within the scope of the claims of the present invention.

1 is a system diagram of an engine according to an embodiment of the present invention. It is a control circuit block diagram of the degradation determination apparatus which concerns on this embodiment. It is a block diagram of the deterioration determination apparatus implement | achieved by the control circuit of FIG. It is a flowchart of the deterioration diagnosis program in this embodiment. It is a flowchart of the deterioration diagnosis program in this embodiment. 6 is a timing chart of signals obtained by executing the flowcharts of FIGS. 4 and 5. FIG. It is a graph which shows the step response characteristic of a linear air fuel ratio sensor. It is a flowchart which shows the detail of a deterioration determination process. It is a graph which shows the relationship between the difference of an average transient time, and the deterioration degree of an emission. It is a graph which shows the relationship between the sum of average transient time, and the deterioration degree of an emission. It is a flowchart which shows an example of the convergence determination threshold value setting subroutine in FIG.

Explanation of symbols

1 Degradation determination device 10 Engine 32 Three-way catalyst 100 Control unit 102 Auxiliary storage device 103 Main storage device 110 Feedback control system 111 Reference input element 112 BIAS correction element 114 Main control element 114a Main learning means 115 Sub control element 115a Sub learning means 116 Disturbance generating means 220 Map 101 CPU
DF _T difference D _O2 differential value D _O2PK differential peak value dOP fluctuation range dThC threshold value DV target value ES evaluation value G _P transfer function G _S transfer function h step response IP reference input K _P , K _S gain Ne Engine speed OP Fluctuation range OV Manipulation amount PF Output of linear air-fuel ratio sensor corresponding to actual air-fuel ratio Qa Intake air amount LR, RL Disturbance SbS Sub correction amount SF Detection value SS Correction amount StP, StS Learning coefficient SW4 Linear air-fuel ratio sensor SW5 Oxygen concentration sensor ThC, THES threshold T _LR, T _RL mean transit time (an example of the evaluation parameter)
TVO Throttle opening α, β Weight coefficient τ Time constant

Claims (5)

A feedback control system that performs feedback control of the air-fuel ratio based on the oxygen concentration in the exhaust gas of the engine;
A linear air-fuel ratio sensor that is provided in the feedback control system and outputs a value proportional to the oxygen concentration in the exhaust gas;
Disturbance generating means for outputting an impulse-like disturbance alternately to the lean side and the rich side a predetermined number of times when a predetermined diagnosis condition is satisfied;
Differential means for outputting a differential value of the output of the linear air-fuel ratio sensor after disturbance output by the disturbance generating means;
Determination parameter calculation means for calculating each of the rich side and the lean side as at least one of the dead time and the time constant of the linear air-fuel ratio sensor based on the differential value output by the differentiating means as a determination parameter for deterioration determination;
Emission deterioration estimation means for estimating the emission deterioration associated with the deterioration of the linear air-fuel ratio sensor based on the difference and sum of the calculated rich determination parameter and lean determination parameter. Deterioration diagnosis device for linear air-fuel ratio sensor. The deterioration diagnosis device for a linear air-fuel ratio sensor according to claim 1,
The emission deterioration estimating means is a correction step of multiplying the difference and the sum by a predetermined weighting factor to obtain a correction parameter, and
An evaluation value calculating step of calculating the sum of the correction parameters to obtain an evaluation value;
A deterioration diagnosis apparatus for a linear air-fuel ratio sensor, comprising: performing a determination step of comparing the calculated evaluation value with a threshold value to determine pass / fail. The deterioration diagnosis apparatus for a linear air-fuel ratio sensor according to claim 2,
The degradation deterioration diagnosis apparatus for a linear air-fuel ratio sensor, wherein the emission deterioration estimation means changes the sum weight coefficient so that the emission deterioration criterion is lowered as the gain of the feedback control system is smaller. The linear air-fuel ratio sensor deterioration diagnosis apparatus according to claim 2 or 3,
The linear air-fuel ratio sensor is disposed upstream of a catalyst that purifies the exhaust gas of the engine,
The feedback control system is provided with a main control element that receives an output of the linear air-fuel ratio sensor and outputs an operation amount based on a difference between the output and a target value. Main learning means for performing learning for correcting a steady deviation between the fuel ratio and the actual air-fuel ratio,
The emission deterioration estimation means changes the weighting factor of the difference in the correction parameter so that the standard of emission deterioration decreases as the learning degree of the main learning means progresses. Deterioration diagnostic device. The deterioration diagnosis apparatus for a linear air-fuel ratio sensor according to any one of claims 2 to 3,
The linear air-fuel ratio sensor is disposed upstream of a catalyst that purifies the exhaust gas of the engine,
The feedback control system is disposed on the downstream side of the catalyst, and an oxygen concentration sensor that outputs an oxygen concentration in exhaust gas purified by the catalyst as a detection value, and a detection value from the oxygen concentration sensor are input. In addition, a sub-control element that corrects the target value based on the sub-correction amount based on the detected value, and learning for correcting a steady deviation between the target air-fuel ratio and the actual air-fuel ratio are provided in the sub-control element. Sub-learning means to perform,
The emission deterioration estimating means changes a weighting factor of the difference in the correction parameter so that the emission deterioration reference is lowered as the learning degree of the sub-learning means progresses. Deterioration diagnostic device.

Images