JP2010127110A - Catalyst abnormality diagnostic system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a catalyst abnormality diagnostic system capable of estimating the catalyst temperature with high accuracy, in the catalyst abnormality diagnostic system. <P>SOLUTION: A system identifying part 10 is composed of a data clustering part 12, a separate hyperplane estimating part 14 and a system parameter estimating part 16. The data clustering part 12 classifies the past data on the catalyst inlet gas temperature and the catalyst temperature into an active state cluster and an inactive state cluster. The separate hyperplane estimating part 14 specifies a boundary condition between an active state and an inactive state of a catalyst by these past data. The system parameter estimating part 16 specifies an ARX model realized between these data in the predetermined time and the catalyst temperature in the time before by one step more than the time. The future catalyst temperature is estimated by applying present data to the specified boundary condition and the ARX model. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a catalyst abnormality diagnosis device, and more particularly, to a catalyst abnormality diagnosis device that performs catalyst abnormality determination using a rule or the like corresponding to a catalyst state.

  Conventionally, for example, Patent Document 1 discloses a catalyst temperature estimation device that estimates the temperature of an exhaust purification catalyst disposed in an exhaust passage of an internal combustion engine. In this catalyst device, an unknown parameter of a relational expression established between the catalyst temperature and the exhaust flow rate is obtained by specifying based on actual measurement data during vehicle travel, and the relational expression is specified by the obtained parameter. .

JP 2004-052556 A JP 2004-263606 A

  By the way, it is conceivable that the catalyst temperature changes depending on whether the exhaust purification catalyst is in an active state or an inactive state. In other words, since the exhaust gas purification reaction occurs as theoretically in the active state, the catalyst temperature may function within a specific temperature range. Conversely, in the inactive state, the purification reaction is incomplete. It is conceivable that the catalyst temperature is not expected to increase due to a chemical reaction, and the catalyst temperature is affected by the exhaust gas flow rate, the exhaust gas temperature, and the like. In such a case, it is desirable that a relational expression corresponding to each catalyst state can be specified. However, the estimation model of Patent Document 1 is a single relational expression, and an exhaust flow rate calculated without distinction between active / inactive states is introduced into this single relational expression. For this reason, when the catalyst state is taken into account, there is a possibility that the accuracy of the estimated catalyst temperature is lowered.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a catalyst abnormality diagnosis device capable of estimating the catalyst temperature with high accuracy.

In order to achieve the above object, a first invention is a catalyst abnormality diagnosis device,
Based on these data, the unknown parameter of the first probability distribution corresponding to the active state of the catalyst and the unknown parameter of the second probability distribution corresponding to the inactive state of the catalyst according to the past data of the catalyst inlet gas temperature and the catalyst temperature are based on these data. Probability distribution specifying means for estimating the first probability distribution and the second probability distribution based on the estimated unknown parameter,
Based on the first probability distribution and the second probability distribution after specification, it is determined whether each of the past data is data in the active state or data in the inactive state, respectively. Data classification means for classifying into active state clusters and inactive state clusters;
Boundary condition specifying means for estimating a boundary condition established at a boundary between the active state and the inactive state based on the past data, and specifying the boundary condition by the estimated unknown parameter;
An unknown parameter of a relationship established between the catalyst inlet gas temperature and catalyst temperature data at a predetermined time and the catalyst temperature data at a time one step ahead of the predetermined time is used as the data of the active state cluster. First rule specifying means for estimating based on the estimated unknown parameter and specifying the relationship as a first rule corresponding to the catalyst active state;
An unknown parameter having a relationship established between the catalyst inlet gas temperature and the catalyst temperature data at a predetermined time and the catalyst temperature data at a time one step ahead of the predetermined time is defined as the inactive state cluster data. A second rule specifying means for specifying the relationship as a second rule corresponding to the catalyst inactive state based on the estimated unknown parameter;
The data including the current values of the catalyst inlet gas temperature and the catalyst temperature data is applied to the specified boundary condition, and the data including the current value is classified into the active state or the inactive state. A category estimation means for estimating whether or not
A catalyst temperature estimating means for estimating a future catalyst temperature by applying data including the current value to the first rule or the second rule;
And an abnormality determining means for determining an abnormality of the catalyst based on the estimation result by the classification estimating means and / or the future catalyst temperature.

The second invention is the first invention, wherein
An abnormality determining means for determining an abnormality of the catalyst based on the estimation result by the classification estimating means and / or the future catalyst temperature is provided.

The third invention is the second invention, wherein
Distance calculating means for calculating a distance indicating how far the data including the current value is away from a boundary between the active state and the inactive state;
Evaluation means for evaluating the validity of the determination result by the abnormality determination means depending on whether or not the distance is equal to or greater than a predetermined first determination value.

Moreover, 4th invention is 2nd or 3rd invention,
Based on the first probability distribution and the second probability distribution after identification, it is determined whether the data including the current value is the data in the active state or the data in the inactive state, Evaluation second means for evaluating the validity of the determination result by the abnormality determination means.

  According to the first aspect of the invention, the future catalyst temperature can be estimated from the two rules, the first rule corresponding to the catalyst active state and the second rule corresponding to the catalyst inactive state. These two rules are based on two probability distributions followed by past data of catalyst inlet gas temperature and catalyst temperature: a first probability distribution corresponding to the catalyst active state and a second probability distribution corresponding to the catalyst inactive state. Each past data is classified and specified based on an active state cluster and an inactive state cluster which are data groups after the classification. Therefore, according to the first invention, since the future catalyst temperature can be estimated by the two rules specified in this way, the future catalyst temperature can be more accurately compared with the case of estimating using a single relational expression. Can be estimated.

  Further, according to the first invention, the data including the current values of the catalyst inlet gas temperature and the catalyst temperature data are converted into the catalyst active state, the catalyst inactive state, based on the boundary condition established at the boundary between the catalyst active state and the catalyst inactive state. The active state can be estimated. Then, by combining the classification based on the boundary condition and the two rules described above, it is possible to switch the rule for estimating the future catalyst temperature depending on which state the data including the current value is in. . Therefore, according to the first invention, it is possible to estimate which state the data including the current value is, and to estimate the future catalyst temperature at that time with high accuracy.

  According to the second aspect of the present invention, it is possible to determine the abnormality of the catalyst based on the estimation result by the segment estimation means and / or the future catalyst temperature. As described above, according to the first invention, it is estimated which state the data including the current value is, and the catalyst temperature at that time is estimated. For this reason, catalyst abnormality determination can be performed using one or both of these two estimation results as a determination index.

  According to the third invention, it is possible to calculate a distance indicating how far the data including the current value is away from the boundary between the catalyst active state and the catalyst inactive state. Then, the credibility of the current value data used in performing the catalyst abnormality determination can be evaluated by using this distance as a determination index. By doing so, it is possible to evaluate the validity of the result of the catalyst abnormality determination.

  According to the fourth invention, based on the first probability distribution and the second probability distribution after specification, the data including the current value is data on the catalyst active state, or the catalyst inactivity. It is possible to determine whether the data is state data. Then, it is possible to evaluate the credibility of the current value data used when the catalyst abnormality determination is performed. Therefore, the validity of the result of the catalyst abnormality determination can be evaluated.

Embodiment 1. FIG.
The catalyst abnormality diagnosis device of Embodiment 1 is configured inside an ECU (Electronic Control Unit) mounted on a vehicle or the like. Therefore, first, the apparatus according to this embodiment will be described with reference to FIG. FIG. 1 is a functional block diagram in a case where the ECU functions as a diagnostic device that performs catalyst abnormality determination.

  As shown in FIG. 1, the ECU includes a system identification unit 10, a catalyst temperature estimation unit 18, and an abnormality determination unit 20. The system identification unit 10 includes a data clustering unit 12, a separated hyperplane estimation unit 14, and a system parameter estimation unit 16.

(System identification unit 10)
The system identification unit 10 obtains an identification expression that describes the characteristics from the temperature of exhaust gas (catalyst inlet gas temperature) flowing into the catalyst disposed in the exhaust passage of the internal combustion engine to the temperature of the catalyst, the active state of the catalyst and the catalyst. Identification is performed as a piecewise affine auto-regressive eXogeneous (PWARX) model in which the identification formula switches depending on the inactive state. The system identification unit 10 determines parameters θ ₁ , θ ₂ , a, and b in equations (1) and (2), which will be described later, based on time-series data regarding past input / output, and identifies the PWARX model. It is characterized by.

  The identification is performed by using the catalyst inlet gas temperature as past input data and the catalyst temperature as output data as past time-series data. The catalyst inlet gas temperature at time k is u (k), and the catalyst temperature is y (k). If time k is the current time, the past time series data is u (k-1), u (k-2), ..., y (k-1), y (k-2), ... It is represented by These past time-series data are assumed to be buffered in the ECU.

上式（１）が触媒活性状態に対応し、上式（２）が触媒不活性状態に対応するものとする。上式（１）及び（２）において、e(k)は式誤差である。また、右肩の添え字「T」は転置行列を意味する。また、x(k)は回帰ベクトルであり、過去の触媒入口ガス温度及び触媒温度を用いて、
x(k)=[y(k-1) ・・・ y(k-n_y) u(k-1) ・・・ u(k-n_u)]^T ・・・（３）
により与えられる。上式（３）において、kは、k=1,2,・・・,Nの離散時刻の値であり、Nはデータ数である。また、自然数n_y,n_uは、それぞれモデルを記述するために必要な過去の時系列データの個数を表す。また、θ₁及びθ₂は、これらのモデルを規定する未知のパラメータである。 Assuming that engine characteristics based on past time series data follow the PWARX model, the relationship between the catalyst inlet gas temperature and the catalyst temperature is expressed by the following equations (1) and (2).

The above equation (1) corresponds to the catalyst active state, and the above equation (2) corresponds to the catalyst inactive state. In the above equations (1) and (2), e (k) is an equation error. The subscript “T” on the right shoulder means a transposed matrix. X (k) is a regression vector, and the past catalyst inlet gas temperature and catalyst temperature are used.
x (k) = [y (k-1) ... y (kn _y ) u (k-1) ... u (kn _u )] ^T (3)
Given by. In the above equation (3), k is a discrete time value of k = 1, 2,..., N, and N is the number of data. The natural numbers n _y and n _u represent the number of past time series data necessary for describing the model. Θ ₁ and θ ₂ are unknown parameters that define these models.

Further, as shown in the above formulas (1) and (2), the catalytically active state and the catalytically inactive state are assumed to be divided by a separation hyperplane on the regression vector space. The separation hyperplane is a boundary between these two states and is expressed by the following equation (4).
a ^T x (k) + b = 0 (4)
In the above equation (4), a and b are unknown parameters that define the separation hyperplane.

(Data clustering unit 12)
The data clustering unit 12 in the system identification unit 10 is configured to classify past time-series data into the catalyst active cluster C ₁ corresponding to the catalyst active state or the catalyst inactive cluster C ₂ corresponding to the catalyst inactive state. Has been.

In the classification, first, an observation data vector z _k = (x ^T (k) y ^T (k)) ^T is defined. Then, it is assumed that N pieces of observation data z ₁ ,..., Z _{N of} this observation data vector z _k follow a mixed normal distribution composed of normal distributions corresponding to the catalyst active state and the catalyst inactive state.

上式（５）において、α₁、α₂は、混合正規分布の混合割合であり、α₁，α₂≧0，α₁+α₂=1の関係を満たす。また上式（５）〜（７）において、μ₁，μ₂は、触媒活性状態、触媒不活性状態のそれぞれに対応する正規分布の平均ベクトルを示す。同様に、Σ₁、Σ₂は、それぞれに対応する正規分布の共分散行列を示す。添え字「n_z」は、観測データベクトルz_kの次元数を示す。さらにdetΣは、Σの行列式を示し、行列の右肩の添え字「-1」は逆行列を意味する。 The mixed normal distribution is expressed by the following equation (5) using the parameter Φ = (α ₁ , α ₂ , μ ₁ , μ ₂ , Σ ₁ , Σ ₂ ).

In the above equation (5), α ₁ and α ₂ are the mixing ratio of the mixed normal distribution, and satisfy the relationship of α ₁ , α ₂ ≧ 0, α ₁ + α ₂ = 1. In the above formulas (5) to (7), μ ₁ and μ ₂ represent average vectors of normal distributions corresponding to the catalyst active state and the catalyst inactive state, respectively. Similarly, Σ ₁ and Σ ₂ indicate normal distribution covariance matrices corresponding to each. The subscript “n _z ” indicates the number of dimensions of the observation data vector z _k . Furthermore, detΣ indicates a determinant of Σ, and the subscript “−1” on the right shoulder of the matrix means an inverse matrix.

を最大化するパラメータΦは、ＥＭアルゴリズムを用いることにより推定することができる。ＥＭアルゴリズムでは、パラメータΦを反復的に計算することができる。 Subsequently, the parameter Φ of the mixed normal distribution is obtained by the maximum likelihood estimation method. Likelihood function

The parameter Φ that maximizes can be estimated by using the EM algorithm. In the EM algorithm, the parameter Φ can be calculated iteratively.

Subsequently, the observed data is classified using the obtained parameter Φ. Here, the assigned probability that the observed data belongs to the catalytically active cluster C ₁ is α ₁ p ₁ (z _k ; Φ), and the assigned probability that is attributed to the catalyst inactive cluster C ₂ is α ₂ p ₂ (z _k ; Φ). And The classification is performed into clusters corresponding to partial distributions having higher attribution probabilities. That is, data of α ₁ p ₁ (z _k ; Φ) ≧ α ₂ p ₂ (z _k ; Φ) is assigned to the catalytic activity cluster C ₁ and α ₁ p ₁ (z _k ; Φ) <α ₂ p ₂ (z _k; [Phi data) are classified into catalytically inactive cluster C _2. By doing so, the times k = 1, 2,..., N corresponding to the observation data can be classified into the respective clusters.

(Separated hyperplane estimation unit 14)
The separation hyperplane estimation unit 14 in the system identification unit 10 is configured to estimate the parameters of the separation hyperplane on the regression space that divides the catalyst active state and the catalyst inactive state. The parameters a and b of the separation hyperplane given by the above equation (4) can be estimated by using a soft margin support vector machine.

  A support vector machine is an algorithm that can determine which of two classes a given data belongs to, and that is true when there is a correct class for all sample data (linear separation is possible). This is a method of adjusting the parameters a and b of the plane. However, there is no single hyperplane that realizes such linear separation. Therefore, the distance from the hyperplane to each class is defined as a margin (1 / || a ||), and the hyperplane parameters a and b are determined so as to maximize this margin.

このように推定されたパラメータa,bは、ＥＣＵに記憶される。 Here, a correct class does not always exist for all sample data (not linearly separable). In such a case, a soft margin support vector machine in which a classification error term is introduced into the support vector machine is used. That is, when the data enters the opposite class beyond the hyperplane, the hyperplane parameters a and b are adjusted so as to minimize the sum of the distances v exceeding the hyperplane. In summary, if a soft margin support vector machine is used, the parameters a and b can be estimated by solving the quadratic optimization problem shown in the following equation (8).

The parameters a and b estimated in this way are stored in the ECU.

上式（９）及び（１０）において、N₁,N₂は、それぞれのクラスタC₁,C₂に含まれるデータ数を示す。このように推定されたパラメータθ₁^,θ₂^は、ＥＣＵに記憶される。 (System parameter estimation unit 16)
The system parameter estimation unit 16 in the system identification unit 10 is configured to estimate the parameters θ ₁ and θ ₂ using data corresponding to discrete times classified by the data clustering unit 12. The parameters θ ₁ and θ ₂ can be estimated by the following equations (9) and (10).

In the above formulas (9) and (10), N ₁ and N ₂ indicate the number of data included in the respective clusters C ₁ and C ₂ . The parameters θ ₁ ^ and θ ₂ ^ estimated in this way are stored in the ECU.

尚、上式（１１）が本発明の第１規則に、上式（１２）が本発明の第２規則にそれぞれ該当する。また、上式（１１）及び（１２）において、x(k+1)は回帰ベクトルであり、時刻kまでの触媒入口ガス温度及び触媒温度を用いて、
x(k+1)=[y(k) ・・・ y(k-n_y+1) u(k) ・・・ u(k-n_u+1)]^T ・・・（１３）
により与えられることになる。 As described above, the system identification unit 10 identifies the PWARX model in which the identification formula is switched between the catalyst active state and the catalyst inactive state. In summary, assuming that the current time is time k, the estimated catalyst temperature y ^ (k + 1) at time k + 1 is based on the catalyst inlet gas temperature and catalyst temperature data history up to time k-1. Using the estimated parameters θ ₁ ^, θ ₂ ^, the following equations (11) and (12) are used. The subscript “^” used here represents an estimated value.

The above equation (11) corresponds to the first rule of the present invention, and the above equation (12) corresponds to the second rule of the present invention. Further, in the above equations (11) and (12), x (k + 1) is a regression vector, and using the catalyst inlet gas temperature and the catalyst temperature up to time k,
x (k + 1) = [y (k) ... y (kn _y +1) u (k) ... u (kn _u +1)] ^T (13)
Will be given by.

  Thus, the estimated value y ^ (k + 1) of the catalyst temperature at time k + 1 sequentially assigns the catalyst inlet gas temperature and catalyst temperature data history up to time k-1 and the current catalyst temperature data. This can be done each time. Therefore, there is an effect that the degree of freedom of model identification is high.

(Catalyst temperature estimation unit 18)
The catalyst temperature estimation unit 18 uses the parameters a and b and the above equation (13) to estimate whether the current catalyst state is active or inactive, the estimation result, and the above equation (11). ) Or (12) is used to estimate y ^ (k + 1).

Specifically, first, parameters a and b estimated from the data history up to time k-1, and a regression vector x (k + 1) generated by adding the output data of the catalyst temperature at time k to this data history, From this, a ^T x (k + 1) + b is calculated. If a ^T x (k + 1) + b ≧ 0, the catalyst temperature estimation unit 18 estimates that the current catalyst state is active, and the catalyst state prediction flag i ^ (k + 1) = 1. Is buffered in the ECU. On the other hand, if a ^T x (k + 1) + b <0, the catalyst temperature estimation unit 18 estimates that the current catalyst state is inactive, and the catalyst state prediction flag i ^ (k + 1) = 0 is buffered in the ECU. Subsequently, an estimated value y ^ (k + 1) of the catalyst temperature at time k + 1 is estimated using the catalyst state prediction flag i ^ (k + 1) and the above expression (11) or (12). . The estimated y ^ (k + 1) is buffered in the ECU.

(Abnormality determination unit 20)
The abnormality determination unit 20 is configured to determine a catalyst abnormality based on y ^ (k + 1) estimated by the catalyst temperature estimation unit 18. The abnormality determination is first performed by calculating a deviation between this y ^ (k + 1) and the actual catalyst temperature y (k + 1) at time k + 1. Specifically, a difference Δy = y ^ (k + 1) −y (k + 1) between these values is calculated and compared with a predetermined set value (threshold value). This threshold value can be freely set according to the required degree of catalyst abnormality determination accuracy. If Δy is equal to or greater than the threshold value, a deviation is observed between the estimated catalyst temperature and the actual catalyst temperature. In such a case, the abnormality determination unit 20 issues an abnormality signal.

  The abnormality determination is made based on whether or not the abnormality signal is detected a set number of times within a predetermined determination time. Specifically, when the above-described abnormality signal is detected, the counter value of the abnormality counter separately incorporated in the abnormality determination unit 20 is increased. When the counter value reaches a set value or more within a predetermined determination time, it is determined that the catalyst is abnormal.

  Thus, according to the configuration of the diagnostic apparatus shown in FIG. 1, the future catalyst temperature can be sequentially estimated using the PWARX model identified based on the past time series data. Then, by comparing the estimated catalyst temperature with the actual catalyst temperature, it can be determined with high accuracy whether the catalyst is abnormal.

In the above-described embodiment, the data clustering unit 12 estimates the parameter Φ of the mixed normal distribution and specifies the above equation (5) so that the “probability distribution specifying means” in the first invention is the data By classifying the observation data in the clustering unit 12, the “data classification unit” in the first invention estimates the parameters a and b of the separation hyperplane in the separation hyperplane estimation unit 14, and the above formula (4 ) Is specified, the “boundary condition specifying means” in the first invention estimates the parameters θ ₁ ^ and θ ₂ ^ in the system parameter estimation unit 16 and specifies the above equations (11) and (12) "the first rule specifying part" and "second rule identification means" of the invention by the, in the catalyst temperature estimating unit 18 calculates a ^{a T x (k + 1)} + b, a catalyst like By setting the prediction flag, the “section estimation means” in the first aspect of the invention estimates the estimated value y ^ (k + 1) of the catalyst temperature at the time k + 1 in the catalyst temperature estimation unit 18. The “catalyst temperature estimating means” in the first invention is realized.

  In the embodiment described above, the “abnormality determination means” according to the second aspect of the present invention is realized by determining abnormality of the catalyst based on y ^ (k + 1) in the abnormality determination unit 20.

Embodiment 2. FIG.
The catalyst abnormality diagnosis device of the second embodiment is not limited to the estimated value y ^ (k + 1) of the catalyst temperature estimated by the catalyst temperature estimation unit 18 in the abnormality determination of the first embodiment, but also the catalyst state prediction flag i ^ (k It is characterized by taking into account the value of +1). For this reason, it demonstrates centering around difference with Embodiment 1 mentioned above, and the description is abbreviate | omitted or simplified about the same matter.

  FIG. 2 is a functional block diagram in a case where the ECU according to the second embodiment functions as a diagnostic device that performs catalyst abnormality determination. As shown in FIG. 2, the ECU includes a system identification unit 10, a catalyst temperature estimation unit 18, a catalyst state estimation unit 22, and an abnormality determination unit 24.

(Catalyst state estimation unit 22)
The catalyst state estimation unit 22 is configured to estimate whether the current catalyst state is active or inactive using the parameters a and b and the above equation (13).

Specifically, first, parameters a and b estimated from the data history up to time k-1, and a regression vector x (k + 1) generated by adding the output data of the catalyst temperature at time k to this data history, From this, a ^T x (k + 1) + b is calculated. If a ^T x (k + 1) + b ≧ 0, the catalyst state estimation unit 22 estimates that the current catalyst state is active, and the catalyst state prediction flag i ^ (k + 1) = 1. Is buffered in the ECU. On the other hand, if a ^T x (k + 1) + b <0, the catalyst state estimation unit 22 estimates that the current catalyst state is inactive, and the catalyst state prediction flag i ^ (k + 1) = 0 is buffered in the ECU.

(Abnormality determination unit 24)
The abnormality determination unit 24 performs catalyst abnormality determination based on y ^ (k + 1) estimated by the catalyst temperature estimation unit 18 and the catalyst state prediction flag i ^ (k + 1) estimated by the catalyst state estimation unit 22. Configured to do. Among these abnormality determinations, those based on the difference Δy between y ^ (k + 1) and the actual catalyst temperature y (k + 1) are the same as those already described in the first embodiment, and thus the description thereof is omitted. To do.

  The abnormality determination based on i ^ (k + 1) is based on the relationship between i ^ (k + 1) and i (k + 1) when the actual catalyst state at time k + 1 is i (k + 1). This is done by comparison. The actual catalyst state i (k + 1) is a parameter that can be acquired by the ECU, such as a change rate (purification rate) of a specific gas concentration in the upstream and downstream of the catalyst, and is a catalyst active / inactive state. There is no particular limitation as long as it can be used as a substitute value. This i (k + 1) is associated with the catalyst state prediction flag i ^ (k + 1), i (k + 1) = 1 in the active state, i (k + 1) in the inactive state. +1) = 0. When the value of i ^ (k + 1) is different from the value of i (k + 1), there is a discrepancy between the estimated catalyst state and the actual catalyst state. In such a case, the abnormality determination unit 24 issues an abnormality signal.

  The abnormality determination is made based on whether or not the abnormality signal is detected a set number of times within a predetermined determination time. Specifically, when the above abnormal signal based on Δy or i ^ (k + 1) is detected, the counter value of the abnormality counter separately built in the abnormality determination unit 24 is increased. When the counter value reaches a set value or more within a predetermined determination time, it is determined that the catalyst is abnormal. In the present embodiment, since the above abnormal signal may be detected based on the two determination indexes Δy and i ^ (k + 1), when one of the determination results is incorrect Thus, it is possible to prevent a misdiagnosis that the final determination is wrong.

  In this embodiment, not only the estimated value y ^ (k + 1) of the catalyst temperature estimated by the catalyst temperature estimation unit 18 but also the catalyst state prediction flag i ^ (k + 1) in the abnormality determination of the first embodiment. However, without considering this estimated catalyst temperature value y ^ (k + 1), an abnormality was determined only by the value of the catalyst state prediction flag i ^ (k + 1). Also good. Note that this modification can also be applied to Embodiments 3 and 4 described later.

  In the above-described embodiment, the abnormality determination unit 24 determines the abnormality of the catalyst based on y ^ (k + 1) and i ^ (k + 1), whereby the “abnormality determination” in the second invention is performed. Means "are realized.

Embodiment 3. FIG.
The catalyst abnormality diagnosis device of the third embodiment is characterized in that, in the abnormality determination of the first embodiment, the validity of the abnormality determination is evaluated using the distance from the separation hyperplane that is the boundary between the active state and the inactive state. And For this reason, it demonstrates centering around difference with Embodiment 1 mentioned above, and the description is abbreviate | omitted or simplified about the same matter.

  FIG. 3 is a functional block diagram in a case where the ECU according to the third embodiment functions as a diagnostic device that performs catalyst abnormality determination. As shown in FIG. 3, the ECU includes a system identification unit 10, a catalyst temperature estimation unit 18, a distance calculation unit 26, and an abnormality determination unit 28.

(Distance calculation unit 26)
The distance calculation unit 26 is configured to calculate d _k = | a ^T x (k + 1) + b | using the parameters a and b and the above equation (13). This d _k corresponds to the distance from the separation hyperplane _represented by the above equation (4). Specifically, the distance d _k = | a ^T x (k + 1) + b | is calculated. The calculated d _k is buffered in the ECU.

(Abnormality determination unit 28)
The abnormality determination unit 28 performs catalyst abnormality determination based on y ^ (k + 1) estimated by the catalyst temperature estimation unit 18 and also performs catalyst abnormality determination using the distance d _k calculated by the distance calculation unit 26. Configured to do. Among these abnormality determinations, those based on the difference Δy between y ^ (k + 1) and the actual catalyst temperature y (k + 1) are the same as those already described in the first embodiment, and thus the description thereof is omitted.

Distance abnormality determination based on the d _k is the distance d _k, is performed by comparing a predetermined set value (threshold). This threshold value can be freely set according to the required degree of determination accuracy of catalyst abnormality. If the distance d _k is less than the threshold value, it is recognized that there is a possibility that the reliability of the data used when calculating y ^ (k + 1) is low. The abnormality determination unit 28 is configured to generate an abnormality signal when the distance d _k is less than the threshold value.

The abnormality determination is made based on whether or not the above abnormality signal is detected a set number of times within a predetermined determination time. Specifically, when the above abnormal signal based on Δy or the distance d _k is detected, the counter value of the abnormal counter separately built in the abnormality determination unit 28 is increased. When the counter value reaches a set value or more within a predetermined determination time, it is determined that the catalyst is abnormal. In the present embodiment, the abnormal signal is detected based on Δy, and the credibility of data can be confirmed based on the distance d _k . Therefore, the reliability of the determination result based on Δy can be increased.

Note that this embodiment can be combined with the second embodiment. That is, the abnormality determination unit 28 can determine the abnormality of the catalyst based on y ^ (k + 1) and i ^ (k + 1), and can evaluate the reliability of data using the distance d _k. It is. Note that this modification can also be applied to Embodiment 4 described later.

In the above-described embodiment, the distance calculation unit 26 corresponds to the “distance calculation means” in the second invention. Further, in the present embodiment, the “evaluation means” in the third aspect of the present invention is realized by determining abnormality of the catalyst based on the distance d _k in the abnormality determination unit 28.

Embodiment 4 FIG.
The catalyst abnormality diagnosis device of the fourth embodiment classifies each of the observation data into the catalyst active cluster C ₁ corresponding to the catalyst active state or the catalyst inactive cluster C ₂ corresponding to the catalyst inactive state in the abnormality determination of the third embodiment. It is characterized by evaluating the validity of the abnormality determination using the belonging probabilities belonging to the two normal distributions. For this reason, it demonstrates centering around difference with Embodiment 3 mentioned above, and the description is abbreviate | omitted or simplified about the same matter.

  FIG. 4 is a functional block diagram in a case where the ECU according to the fourth embodiment functions as a diagnostic device that performs catalyst abnormality determination. As shown in FIG. 4, the ECU includes a system identification unit 10, a catalyst temperature estimation unit 18, a distance calculation unit 26, an attribution probability calculation unit 30, and an abnormality determination unit 32.

(Attribution probability calculation unit 30)
The attribution probability calculation unit 30 uses the parameter Φ and the above equation (6) or (7), and the attribution probability that the observation data vector z _{k at} time k belongs to the catalytically active cluster C ₁ and the catalytically inactive cluster C _2. It is comprised so that may be calculated. As described above, the data clustering unit 12 estimates the parameter Φ based on the time series data up to time k−1.

Membership probability, first, similar observation data vector z _k defined cluster analysis unit 12, the observation data vector ^{_{z k = (x T (k}} ) y T (k)) that defines the ^T. Then, α ₁ p ₁ (z _k ; Φ) and α ₂ p ₂ (z _k ; Φ) are calculated using the parameter Φ estimated by the data clustering unit 12. The calculated α ₁ p ₁ (z _k ; Φ) and α ₂ p ₂ (z _k ; Φ) are buffered in the ECU.

(Abnormality determination unit 32)
The abnormality determination unit 32 performs catalyst abnormality determination based on y ^ (k + 1) estimated by the catalyst temperature estimation unit 18, and calculates the distance d _k calculated by the distance calculation unit 26 and the attribution probability calculation unit 30. The abnormality determination of the catalyst is performed using the assigned probability. Among these abnormality determinations, those based on the difference Δy between y ^ (k + 1) and the actual catalyst temperature y (k + 1) and the distance d _k are the same as those already described in the third embodiment. Description is omitted.

The abnormality determination based on the probability of belonging is made by assigning the probability of belonging to the catalyst active cluster C ₁ α ₁ p ₁ (z _k ; Φ) and the probability of belonging to the catalyst inactive cluster C ₂ α ₂ p ₂ (z _k ; Φ). This is done by comparing. This comparison shows that when α ₁ p ₁ (z _k ; Φ) ≧ _^ α ₂ p ₂ (z _k ; Φ), there is a high probability that the catalyst is in an active state, but α ₁ p ₁ (z _k ; Φ ) < _^ Α ₂ p ₂ (z _k ; Φ) indicates that there is a high probability that the catalyst is in an inactive state. That is, if the catalyst state prediction flag is determined to be i ^ (k + 1) = 1, but the probability that the catalyst is inactive is high, y ^ (k + 1) is calculated. The possibility of low credibility of the data used at the time is recognized. The abnormality determination unit 32 is configured to generate an abnormality signal when α ₁ p ₁ (z _k ; Φ) < _^ α ₂ p ₂ (z _k ; Φ).

The abnormality determination is made based on whether or not the above abnormality signal is detected a set number of times within a predetermined determination time. Specifically, when the above abnormal signal based on Δy, distance d _k, or attribution probability is detected, the counter value of the abnormality counter separately incorporated in the abnormality determination unit 32 is increased. When the counter value reaches a set value or more within a predetermined determination time, it is determined that the catalyst is abnormal. In the present embodiment, the abnormal signal is detected based on Δy, and the credibility of the data can be confirmed based on the distance d _k and the belonging probability. Therefore, the reliability of the determination result based on Δy can be increased.

  In the embodiment described above, the “evaluation second means” in the fourth aspect of the present invention is realized by determining abnormality of the catalyst based on the probability of belonging in the abnormality determination unit 32.

It is a functional block diagram in case ECU of Embodiment 1 functions as a diagnostic apparatus which performs abnormality determination of a catalyst. It is a functional block diagram in case ECU of Embodiment 2 functions as a diagnostic apparatus which performs abnormality determination of a catalyst. It is a functional block diagram in case ECU of Embodiment 3 functions as a diagnostic apparatus which performs abnormality determination of a catalyst. It is a functional block diagram in case ECU of Embodiment 4 functions as a diagnostic apparatus which performs abnormality determination of a catalyst.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 System identification part 12 Data clustering part 14 Separation hyperplane estimation part 16 System parameter estimation part 18 Catalyst temperature estimation part 20, 24, 28, 32 Abnormality determination part 22 Catalyst state estimation part 26 Distance calculation part 30 Attribution probability calculation part

Claims (4)

Based on these data, the unknown parameter of the first probability distribution corresponding to the active state of the catalyst and the unknown parameter of the second probability distribution corresponding to the inactive state of the catalyst according to the past data of the catalyst inlet gas temperature and the catalyst temperature are based on these data. Probability distribution specifying means for estimating the first probability distribution and the second probability distribution based on the estimated unknown parameter,
Based on the first probability distribution and the second probability distribution after specification, it is determined whether each of the past data is data in the active state or data in the inactive state, respectively. Data classification means for classifying into active state clusters and inactive state clusters;
Boundary condition specifying means for estimating a boundary condition established at a boundary between the active state and the inactive state based on the past data, and specifying the boundary condition by the estimated unknown parameter;
An unknown parameter of a relationship established between the catalyst inlet gas temperature and catalyst temperature data at a predetermined time and the catalyst temperature data at a time one step ahead of the predetermined time is used as the data of the active state cluster. First rule specifying means for estimating based on the estimated unknown parameter and specifying the relationship as a first rule corresponding to the catalyst active state;
An unknown parameter having a relationship established between the catalyst inlet gas temperature and the catalyst temperature data at a predetermined time and the catalyst temperature data at a time one step ahead of the predetermined time is defined as the inactive state cluster data. A second rule specifying means for specifying the relationship as a second rule corresponding to the catalyst inactive state based on the estimated unknown parameter;
The data including the current values of the catalyst inlet gas temperature and the catalyst temperature data is applied to the specified boundary condition, and the data including the current value is classified into the active state or the inactive state. A category estimation means for estimating whether or not
A catalyst temperature estimating means for estimating a future catalyst temperature by applying data including the current value to the first rule or the second rule;
A catalyst abnormality diagnosis apparatus comprising:   The catalyst abnormality diagnosis device according to claim 1, further comprising an abnormality determination unit that determines an abnormality of the catalyst based on an estimation result obtained by the classification estimation unit and / or the future catalyst temperature. Distance calculating means for calculating a distance indicating how far the data including the current value is away from a boundary between the active state and the inactive state;
Evaluation means for evaluating the validity of the determination result by the abnormality determination means depending on whether the distance is equal to or greater than a predetermined first determination value;
The catalyst abnormality diagnosis device according to claim 2, further comprising: Based on the first probability distribution and the second probability distribution after identification, it is determined whether the data including the current value is the data in the active state or the data in the inactive state, An evaluation second means for evaluating the validity of the determination result by the abnormality determination means;
The catalyst abnormality diagnosis device according to claim 2, wherein the catalyst abnormality diagnosis device is provided.

Images