JP2013079613A - Calculation device for intake parameter for internal combustion engine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a calculation device for an intake parameter for an internal combustion engine that can precisely calculate the intake parameter when an intake throttle valve is installed.SOLUTION: The calculation device 1 for an intake parameter includes an ECU2. The ECU2 calculates the error KTHERRCOR by error model equation (8) (step 2), calculates the correction factor KTHCOR as the inverse of sum of the error KTHERRCOR and value 1 (step 3), and calculates the passing air amount GAIRTH by correcting the basic passing air amount GAIRTHN calculated in equation (11) with the correction factor KTHCOR (step 6). A model parameter A of error model equation (8) is obtained by on-board identification calculation evenly weighted with equations (15)-(19), using the ratio KTHERR of the model equation value KTHCAL and the map value KTH_F corrected by the plugging factor KCLS (steps 48-53).

Description

  The present invention relates to an intake air parameter calculation device for an internal combustion engine that calculates an intake air parameter representing an air state in an intake passage such as an intake pressure and an intake air amount of the internal combustion engine.

  2. Description of the Related Art Conventionally, a device described in Patent Document 1 is known as an intake parameter calculation device for an internal combustion engine. This intake parameter calculation device calculates a required intake pressure P as an intake parameter, and includes an air flow meter, an intake pressure sensor, a throttle sensor, and the like. In this intake parameter calculation device, the accelerator opening Acc, which is the amount of operation of the accelerator pedal, is based on the detection signal of the throttle sensor, the actual intake air amount Gact is based on the detection signal of the air flow meter, and the actual intake pressure Pact is the intake pressure sensor. Are calculated based on the detected signals.

  Further, the required shaft torque is calculated using the accelerator opening degree Acc and the engine speed Ne, and the required intake pressure P is calculated using the intake system model of Expression (3) that defines the relationship between the required shaft torque and the required intake pressure P. Is calculated. This intake system model is derived from a gas state equation and includes a learning value Kn as a multiplication coefficient. This learning value Kn is for correcting the modeling error, and is calculated by a smoothing operation (weighted average operation) between the previous value and the base value Knbase. The base value Knbase is calculated by dividing the ratio between the actual intake pressure Pact and the actual intake air amount Gact by the ratio between the required intake pressure P and the required intake air amount G calculated based on the intake system model.

JP 2002-309993A

  When an intake throttle valve that changes the opening area of the intake passage, such as a throttle valve, is provided in the intake passage as in the internal combustion engine of Patent Document 1, the intake pressure is lower than the intake throttle valve in the intake passage. Because of the pressure, it has a characteristic that it is easily affected by the change in the opening of the intake throttle valve. On the other hand, in the case of the intake parameter calculation device of Patent Document 1, the required intake pressure P is calculated using an intake system model derived from the gas state equation without considering the opening of the intake throttle valve. However, there is a problem that the calculation accuracy is low. In particular, under a condition in which the opening degree of the intake throttle valve is likely to change, such as in a transient operation state, the degree of decrease in calculation accuracy increases.

  The present invention has been made to solve the above-described problem, and an object of the present invention is to provide an intake parameter calculation device for an internal combustion engine that can accurately calculate an intake parameter when an intake throttle valve is provided. And

  In order to achieve the above object, the invention according to claim 1 is characterized in that the amount of air passing through the intake throttle valve is changed as the passing air amount by the intake throttle valve (throttle valve 7a) provided in the intake passage 6. In the internal combustion engine 3 in which the intake throttle valve opening (throttle valve opening TH) is feedback controlled so that the rotational speed NE during idling converges to the target rotational speed NOBJ, the state of the air in the intake passage 6 is changed. An intake parameter calculation device 1 for an internal combustion engine 3 that calculates an intake parameter (passage air amount GAIRTH, intake air amount GAIR, intake pressure PBA) to be expressed, and a basic intake parameter (basic passing air amount GAIRTHN, Basic intake parameter calculation means (ECU 2, step 6, calculating basic intake air amount GAIRN, basic intake pressure PBAN) 0, 110), the upstream pressure (atmospheric pressure PA), which is the pressure in the intake passage 6 on the upstream side of the intake throttle valve, derived by a predetermined modeling method, and the intake passage 6 on the downstream side of the intake throttle valve Model equation defining the relationship between the downstream pressure (intake pressure PBA), which is the internal pressure, the opening function value KTH determined by the opening of the intake throttle valve, and the passing air amount GAIRTH [Expressions (7), (14 ), (24)], a first opening degree function value calculating means (ECU2, step 14) for calculating a first opening degree function value (model equation value KTHCAL) as a first calculated opening degree function value. 103, 123) and a correlation model (FIG. 2) representing the correlation between the opening of the intake throttle valve (throttle valve opening TH) and the opening function value KTH, As the calculated value, the second opening function value (map value K Second opening degree function value calculating means (ECU2, steps 11, 100, 120) for calculating H), and feedback correction amount (integral term) for feedback control of the opening degree of the intake throttle valve (throttle valve opening degree TH). In accordance with the learned value IXREF), the clogging degree value calculating means (ECU2, steps 162 to 166) for calculating the clogging degree value (clogging coefficient KCLS) representing the clogging degree of the intake passage 6 and the calculated clogging degree value ( A corrected second opening function value (corrected map value KTH_F) is calculated by correcting the calculated second opening function value (map value KTH) using the clogging coefficient KCLS). An opening function value calculating means (ECU2, steps 11, 100, 120), one of the calculated first opening function value and one of the calculated second opening function value and the other; Correction value calculation means (ECU2, steps 3, 100 to 107, 120) that calculates a correction value (correction coefficient KTHCOR, correction coefficient KAFMERR, correction term PBAERRCOR) using a function value ratio (function value error KTHERR) that is the ratio of 131) and the intake air parameter calculating means (ECU2, step 6, calculating the intake air parameters (passage air amount GAIRTH, intake air amount GAIR, intake air pressure PBA)) by correcting the basic intake air parameters with the calculated correction values. 91, 111).

  According to the intake air parameter calculation device for an internal combustion engine, the first opening function value is determined by the upstream pressure, which is the pressure in the intake passage on the upstream side of the intake throttle valve, and in the intake passage on the downstream side of the intake throttle valve. The second opening function value is calculated using a model formula that defines the relationship between the downstream pressure, which is the pressure, the opening function value determined by the opening degree of the intake throttle valve, and the passing air amount. Is calculated using a correlation model that represents the correlation between the opening degree and the opening degree function value. Further, a clogging degree value indicating the clogging degree of the intake passage is calculated according to the feedback correction amount for feedback control of the opening degree of the intake throttle valve, and the calculated second capping value is calculated using the calculated clogging degree value. By correcting the opening function value, the corrected second opening function value is calculated. Further, the correction value is calculated using the calculated first opening function value and the function value ratio which is the ratio of one of the calculated second opening function value and the corrected second opening function value. By correcting the basic intake parameter with the corrected value, the intake parameter is calculated. In this case, the corrected second opening degree function value is calculated by correcting the second opening degree function value using the clogging degree value indicating the clogging degree of the intake passage. The value can be calculated as a value reflecting the degree of clogging of the intake passage. Further, both the first opening function value and the corrected second opening function value are function values determined by the opening of the intake throttle valve, and a correction value calculated using the ratio of the two values. Since the intake air parameter is calculated by correcting the basic intake air parameter, the intake air parameter can be calculated while reflecting the opening state of the intake throttle valve. In addition to this, the function value ratio that is the ratio of one of the first opening function value and the corrected second opening function value to the other is the difference between the two opening function values, that is, the correlation with the model equation. Since it represents an error with the relational model, the correction value can be calculated as one that can correct such an error. As described above, the intake parameter can be calculated with higher accuracy than before even under conditions where the opening of the intake throttle valve is likely to change, such as in a transient operation state, or under conditions where the intake passage is clogged. Productivity can be improved (in the present specification, “clogged intake passage” is not limited to clogged intake passage, but also includes clogged intake throttle valve).

BRIEF DESCRIPTION OF THE DRAWINGS It is a schematic block diagram of the intake parameter calculation apparatus which concerns on one Embodiment of this invention, and the internal combustion engine to which this is applied. It is a figure which shows an example of the map used for calculation of the map value KTH of an opening degree function value. It is a figure which shows the example of a calculation result at the time of using the quadratic equation which defined the relationship between the function value error KTHERR and the throttle valve opening TH as an error model formula. It is a figure which shows the example of a calculation result at the time of using the error model formula of embodiment. It is a flowchart which shows the calculation process of passage air amount GAIRTH. It is a figure which shows an example of the map used for calculation of the flow function value FPBAPA. It is a flowchart which shows a model parameter learning process. It is a flowchart which shows a learning condition determination process. It is a flowchart which shows the identification calculation process of the model parameter A. It is a figure which shows an example of the identification calculation process result by an inhalation parameter calculation apparatus. It is a figure which shows the comparative example of an identification calculation process result. It is a flowchart which shows an atmospheric pressure estimation process. It is a flowchart which shows the calculation process of estimated atmospheric pressure HPA. It is a figure which shows an example of the map used for calculation of the rotation correction coefficient KTHNE. It is a figure which shows an example of the map used for calculation of the correction | amendment term CORHPA. It is a flowchart which shows the calculation process of intake air amount GAIR. It is a figure which shows an example of the map used for calculation of the basic intake air amount GAIRN. It is a flowchart which shows the calculation process of the correction coefficient KAFMERR. It is a flowchart which shows the calculation process of intake pressure PBA. It is a figure which shows an example of the map used for calculation of correction | amendment term PBAERRRCOR. It is a figure which shows an example of the map used for calculation of the value R_PCOR for correction | amendment calculation of a pressure ratio. It is a flowchart which shows an idle FB control process. It is a flowchart which shows a clogging coefficient calculation process. It is a figure which shows an example of the map each used for calculation of the reference | standard upper limit value thmax and the reference | standard minimum value thmin of a throttle valve opening degree.

  Hereinafter, an intake parameter calculation apparatus for an internal combustion engine according to an embodiment of the present invention will be described with reference to the drawings. As shown in FIG. 1, the intake parameter calculation device 1 of the present embodiment includes an ECU 2. As will be described later, the ECU 2 uses a passing air amount GAIRTH, an intake air amount GAIR, and an intake pressure PBA as intake parameters. Is calculated.

  The engine 3 is of a gasoline engine type mounted on a vehicle (not shown), and a fuel injection valve 4 and a spark plug 5 are attached to the engine 3 for each cylinder. The fuel injection valve 4 is electrically connected to the ECU 2, and the valve opening time and valve opening timing are controlled by the ECU 2, whereby fuel injection control is executed.

  The spark plug 5 is also electrically connected to the ECU 2, and the discharge state is controlled by the ECU 2 so that the air-fuel mixture in the combustion chamber is combusted at a timing corresponding to the ignition timing. That is, ignition timing control is executed.

  An air flow sensor 20, a throttle valve mechanism 7, a throttle valve opening sensor 21, and an intake pressure sensor 22 are provided in the intake passage 6 of the engine 3 in order from the upstream side. The air flow sensor 20 detects a flow rate of air flowing in the vicinity of the air flow sensor 20 in the intake passage 6 (hereinafter referred to as “intake air amount”), and outputs a detection signal representing it to the ECU 2. The ECU 2 calculates the intake air amount GAIR based on the detection signal of the air flow sensor 20, as will be described later. This intake air amount GAIR is calculated as a mass flow rate.

  The throttle valve mechanism 7 includes a throttle valve 7a and a TH actuator 7b that opens and closes the throttle valve 7a. The throttle valve 7a is rotatably provided in the middle of the intake passage 6, and changes the amount of air passing through the throttle valve 7a by a change in opening degree associated with the rotation. The TH actuator 7b is a combination of a motor connected to the ECU 2 and a gear mechanism (both not shown), and is controlled by a control input signal from the ECU 2 to change the opening of the throttle valve 7a. .

  The throttle valve opening sensor 21 is composed of, for example, a potentiometer, and detects the opening TH (hereinafter referred to as “throttle valve opening”) TH of the throttle valve 7a, and sends a detection signal indicating it to the ECU 2. Output. The ECU 2 calculates the throttle valve opening TH based on the detection signal of the throttle valve opening sensor 21. The throttle valve opening TH is calculated as an angle (°). In the present embodiment, the throttle valve 7a corresponds to the intake throttle valve, and the throttle valve opening TH corresponds to the opening of the intake throttle valve.

  Further, the intake pressure sensor 22 is disposed in a surge tank portion downstream of the throttle valve 7a of the intake passage 6 and detects the pressure in the intake passage 6 (hereinafter referred to as “intake pressure”). Is output to the ECU 2. The ECU 2 calculates the intake pressure PBA based on the detection signal of the intake pressure sensor 22, as will be described later. This intake pressure PBA is calculated as an absolute pressure.

  On the other hand, an intake air temperature sensor 23, an atmospheric pressure sensor 24, a crank angle sensor 25, and four wheel speed sensors 26 (only one is shown) are electrically connected to the ECU 2, respectively. The intake air temperature sensor 23 and the atmospheric pressure sensor 24 detect and represent the temperature of air in the intake passage 6 (hereinafter referred to as “intake air temperature”) and the pressure of the atmosphere (hereinafter referred to as “atmospheric pressure”), respectively. A detection signal is output to the ECU 2.

  The ECU 2 calculates the intake air temperature TA and the atmospheric pressure PA based on the detection signals of the intake air temperature sensor 23 and the atmospheric pressure sensor 24, respectively. The intake air temperature TA is calculated as an absolute temperature, and the atmospheric pressure PA is calculated as an absolute pressure.

  The crank angle sensor 25 includes a magnet rotor and an MRE pickup, and outputs a CRK signal and a TDC signal, which are pulse signals, to the ECU 2 as the crankshaft (not shown) rotates. The CRK signal is output at one pulse every predetermined crank angle (for example, 2 °), and the ECU 2 calculates the engine speed NE (hereinafter referred to as “engine speed”) NE based on the CRK signal. The TDC signal is a signal indicating that the piston of each cylinder is at a predetermined crank angle position slightly ahead of the TDC position of the intake stroke, and one pulse is output for each predetermined crank angle.

  Further, each of the four wheel speed sensors 26 detects the rotational speed of the corresponding wheel, and outputs a detection signal representing it to the ECU 2. The ECU 2 calculates the vehicle speed VP, the total travel distance DIST after starting the engine, and the like based on the detection signals of these wheel speed sensors.

  On the other hand, the ECU 2 is composed of a microcomputer including a CPU, a RAM, a ROM, an I / O interface (all not shown), and the engine 2 based on the detection signals of the above various sensors 20 to 26. While calculating various operating state parameters representing the operating state of the engine 3 such as the rotational speed NE, various calculation processes such as a calculation process of the passing air amount GAIRTH, an idle FB control process, etc., as described below. Various control processes are executed.

  In the present embodiment, the ECU 2 includes the basic intake parameter calculation means, the first opening function value calculation means, the second opening function value calculation means, the clogging degree value calculation means, and the corrected second opening function value calculation means. These correspond to correction value calculation means and intake parameter calculation means.

Hereinafter, the principle of the calculation method of the passing air amount GAIRTH in the present embodiment will be described. In the case of the engine 3 of the present embodiment, since the throttle valve 7a is provided in the intake passage 6, air passing through the throttle valve 7a (hereinafter referred to as “passing air”) is regarded as a compressive fluid and an adiabatic flow. When the throttle valve 7a is modeled as a nozzle, a model equation shown in the following equation (1) is obtained.

In the above equation (1), u is the flow velocity of the passing air, P ₁ and P ₂ are the upstream and downstream pressures of the throttle valve 7a, ρ ₁ is the density of the intake air upstream of the throttle valve 7a, κ Represents the specific heat ratio of the intake air.

Next, let G be the flow rate of the passing air, Ath is the opening area of the throttle valve 7a, Cd is the flow coefficient of the throttle valve 7a, T is the temperature of the air, and R is the gas constant of the air. When applying and deforming, the following equation (2) is obtained.

Here, the term of the pressure ratio P ₂ / P _{1 in} the above equation (2) is defined as the following equation (3) as the flow function value Ψ, and when the above equation (2) is rewritten using this, the following equation is obtained: (4) is obtained.

In the above equation (4), the opening area Ath and the flow coefficient Cd are functions determined by the throttle valve opening TH, so the value Cd · Ath is defined as the opening function value KTH (= Cd · Ath), When the above equation (4) is rewritten using this, the following equation (5) is obtained.

In the above equation (5), the pressure loss on the upstream side of the throttle valve 7a is ignored, the upstream pressure P ₁ is replaced with the atmospheric pressure PA, the flow rate G is replaced with the passing air amount GAIRTH, and the temperature T is replaced with the intake air temperature TA. In addition, when the flow function value ψ is expressed as FPBAPA, the square root of the gas constant R as RGAS (= R ^1/2 ), and the square root of the intake air temperature TA as RTTA (= TA ^1/2 ), the following equation (6) is obtained. A model formula is obtained. In addition, KC of the following formula (6) is a conversion coefficient for converting the unit of the passing air amount GAIRTH into (g / sec).

In the above equation (6), the flow rate function value FPBAPA is a value determined by the pressure ratio R_P (= PBA / PA), which is the ratio of the intake pressure PBA and the atmospheric pressure PA. Accordingly, it is calculated by searching the map. Further, when the above equation (6) is arranged for KTH and the passing air amount GAIRTH is replaced with the intake air amount GAIR, the following equation (7) is obtained.

  Here, the opening function value KTH is a function value determined by the throttle valve opening TH, and when an actual relationship between the opening function value KTH and the throttle valve opening TH is measured and a map is created (that is, When identified off-line), the one shown in FIG. 2 is obtained. The opening function value calculated by the above equation (7) is set as the model formula value KTHCAL, the opening function value calculated by the map search in FIG. 2 is set as the map value KTH, and the ratio between the model formula value KTHCAL and the map value KTH Is KTHERR (= KTHCAL / KTH), and when KTHERR = 1, KTH = KTHCAL, and there is no modeling error in the model equation (6). On the other hand, when the value KTHERR deviates from the value 1, the larger the deviation degree, the larger the deviation degree of the model formula value KTHCAL from the map value KTH. The error will be large.

  As described above, the value KTHERR represents the modeling error of the model formula (6), and in the following description, the value KTHERR is referred to as “function value error KTHERR”. This function value error KTHERR is mainly caused by the deviation of the opening area of the throttle valve 7a with respect to the standard product (reference product). This deviation is a deviation of the actual product with respect to the map value KTH set based on the standard product, and is caused by, for example, variation in the accuracy of the throttle bore diameter or foreign matter such as carbon adhering to the throttle bore. is there. In the following description, this deviation is referred to as “first error”. In addition, since the function value error KTHERR is calculated using the detection signals of the air flow sensor 20 and the intake pressure sensor 22, the function value error KTHERR is detected by the air flow sensor 20 (hereinafter referred to as “second error”) or This also occurs due to a detection error of the intake pressure sensor 22 (hereinafter referred to as “third error”).

  In the case of the first error, the larger the throttle valve opening TH (that is, the larger the opening area of the throttle valve 7a), the smaller the influence on the function value error KTHERR. In the above region, the first error can be ignored. The predetermined opening THB can be obtained experimentally and can be calculated based on an allowable error in the opening area of the throttle valve 7a. Further, when the pressure ratio R_P is less than the critical pressure ratio R_CRIT, the flow rate function value FPBAPA is a constant value. Therefore, the function value error KTHERR is not affected by the detection error of the intake pressure sensor 22 in the region of R_P <R_CRIT. Therefore, if the throttle valve opening TH at which the pressure ratio R_P becomes the critical pressure ratio R_CRIT is TH_CRIT, the third error can be ignored in the region of TH <TH_CRIT (that is, the region of R_P <R_CRIT). As a result, in the region of THB ≦ TH <TH_CRIT (that is, the region in which THB ≦ TH and R_P <R_CRIT is satisfied), the first error and the third error can be ignored. Therefore, the second error is caused by the function value error KTHERR. It becomes possible to specify.

  For the above reason, in this embodiment, as will be described later, in the region of TH <THB, in order to compensate (correct) the first error, the function value error KTHERR is used to correct the basic passing air amount GAIRTHN. Correction coefficient KTHCOR is calculated, and in the region of THB ≦ TH and R_P <R_CRIT, the correction coefficient KAFMERR for correcting the basic intake air amount GAIRN is calculated using the function value error KTHERR to compensate for the second error. In the region of TH_CRIT ≦ TH (that is, the region of R_CRIT ≦ R_P), the compensation term PBAERRRCOR for correcting the basic intake pressure PBAN is calculated using the function value error KTHERR to compensate for the third error.

  Further, when the intake passage 6 and the throttle valve 14a are clogged (hereinafter referred to as “clogging of the intake passage 6”) as the use period of the engine 3 becomes longer, the throttle valve opening TH and the map value in FIG. The relationship with KTHmap causes an error (hereinafter referred to as “clogging error”) with respect to the actual relationship between the throttle valve opening TH and the opening function value. In order to compensate for this clogging error, in this embodiment, a clogging coefficient KCLS representing the degree of clogging of the intake passage 6 is calculated by a method described later, and the map value KTH is corrected using this to calculate the opening function. A corrected map value KTH_F of the value is calculated. The corrected map value KTH_F is used in place of the map value KTH, and the ratio (= KTHCAL / KTH_F) between the model formula value KTHCAL and the corrected map value KTH_F is used as the function value error KTHERR. In this embodiment, the map of FIG. 2 is the correlation model, the model formula value KTHCAL is the first opening function value, the map value KTH is the second opening function value, and the corrected map value KTH_F is corrected. The function value error KTHERR corresponds to the function value ratio for the second opening function value.

Hereinafter, a method of calculating the passing air amount GAIRTH while compensating for the first error described above will be described. First, a value obtained by subtracting the value 1 from the function value error KTHERR is set as an error KTHERCOR (= KTHERR-1). In this case, as described above, the degree to which the value KTHERR deviates from the value 1 represents the modeling error. Therefore, the error KTHERCOR represents the modeling error, and the error model equation is expressed by the following equation (8 ). The reason for using this error model equation (8) will be described later.

A in the error model formula (8) is a model parameter. When the least square method is used as a calculation method for on-board identification of the model parameter A, the following formula (9) is used as an identification calculation formula for the model parameter A: Is obtained.

  As will be described later, in the actual calculation of the model parameter A, the calculation accuracy in the identification result of the model parameter A is reduced even when the sampling region of the throttle valve opening TH is biased based on the above equation (9). An arithmetic technique that can be avoided is used.

Next, using the identified model parameter A, the error KTHERCOR is calculated by (8) described above, and then the error correction coefficient KTHCOR is calculated by the following equation (10).

  As shown in the above equation (10), the error correction coefficient KTHCOR is calculated as a value corresponding to the sum of the error KTHERCOR and the value 1, that is, the inverse of the function value error KTHERR. This is due to the following reason. That is, the function value error KTHERR is the ratio of the model formula value KTHCAL and the corrected map value KTH_F. For example, when KTHERR> 1, in other words, when the model formula value KTHCAL exceeds the corrected map value KTH_F. Therefore, in order to correct the excess, the passing air amount GAIRTH may be divided by the function value error KTHERR. On the other hand, if KTHERR <1, that is, if the model formula value KTHCAL is lower than the corrected map value KTH_F, the passing air amount GAIRTH is divided by the function value error KTHERR to correct the lower value. It will be good. On the other hand, since the error correction coefficient KTHCOR is used as a multiplication coefficient as will be described later, in order to correct the amount by which the model formula value KTHCAL is below or above the corrected map value KTH_F, It is calculated as a value corresponding to the inverse of the function value error KTHERR.

Finally, the passing air amount GAIRTH is calculated by the following equations (11) and (12) using the error correction coefficient KTHCOR.

  Here, GAIRTHN in the above equation (11) is the basic passing air amount, and HPA is an estimated atmospheric pressure calculated as described later. This equation (11) replaces the passing air amount GAIRTH of the above-described equation (7) with the basic passing air amount GAIRTHN, the opening function value KTH with the corrected map value KTH_F, and the atmospheric pressure PA with the estimated atmospheric pressure HPA. It corresponds to that. The reason why the estimated atmospheric pressure HPA is used in this equation (11) instead of the atmospheric pressure PA is to improve the calculation accuracy of the passing air amount GAIRTH while avoiding the fluctuation of the atmospheric pressure PA. As shown in the above equation (12), the passing air amount GAIRTH is calculated by correcting the basic passing air amount GAIRTHN with a correction coefficient KTHCOR. In the present embodiment, the basic passing air amount GAIRTHN corresponds to a basic intake parameter, and the correction coefficient KTHCOR corresponds to a correction value.

  Next, the reason for using the error model equation (8) described above will be described. First, as described above, the opening function value KTH is a product of the opening area Ath and the flow coefficient Cd, and has a high correlation with the square value of the radius of the throttle valve 7a. Using the quadratic equation in which is a dependent variable and the throttle valve opening TH is an independent variable, better calculation accuracy can be obtained. In this case, in the region of THB ≦ TH, the function value error KTHERR is not caused by the calculation error of the throttle valve opening TH, but is caused by the calculation error of the intake air amount GAIR or the intake pressure PBA as described above. In addition, under the condition where there is no calculation error of the intake air amount GAIR and the intake pressure PBA, as shown in FIG. 3, the function value error KTHERR = 1, so that it is not necessary to identify the model parameter A of the error model equation become.

In addition, as shown in FIG. 3, when a quadratic expression having a function value error KTHERR as a dependent variable Y and a throttle valve opening TH as an independent variable X is used as an error model expression, the error model expression Is in the form of Y = a · X ² + b · X + c (where a, b, and c are model parameters), it becomes necessary to identify the three model parameters a, b, and c. As a result, the identification calculation becomes complicated and the calculation load increases. In addition, the black circle of FIG. 3 has shown the calculation result for every control timing of the function value error KTHERR.

On the other hand, when the above equation (8) is used, as shown in FIG. 4, when the error KTHERCOR is the dependent variable Y and TH-THB is the independent variable X, the error model equation is Y = A · X ^2. In this form, only one model parameter A needs to be identified, whereby the identification calculation is facilitated and the calculation load can be reduced. For the above reason, in this embodiment, the equation (8) is used as the error model equation. The black circles in FIG. 4 indicate the calculation results for each control timing of the error KTHERCOR.

  Hereinafter, the calculation process of the passing air amount GAIRTH will be described with reference to FIG. This calculation process calculates the passing air amount GAIRTH by the above-described calculation method, and is executed by the ECU 2 every time the CRK signal is generated a predetermined number of times (that is, every predetermined crank angle). In the following description, it is assumed that the various calculated values are all stored in the RAM of the ECU 2, and the values read from the RAM at the current control timing, that is, the values calculated at the previous control timing are the previous values. That's it.

  As shown in the figure, first, in step 1 (abbreviated as “S1” in the figure, the same applies hereinafter), it is determined whether or not the throttle valve opening TH is less than a predetermined opening THB. When the determination result is YES, it is determined that the passing air amount GAIRTH needs to be corrected by the error correction coefficient KTHCOR, the process proceeds to step 2, and the error KTHERCOR is calculated by the above-described equation (8).

  Next, in step 3, the error correction coefficient KTHCOR is calculated by the above-described equation (10).

  On the other hand, when the determination result in step 1 is NO, it is determined that there is no need to correct the passing air amount GAIRTH by the error correction coefficient KTHCOR, and the process proceeds to step 4 where the error correction coefficient KTHCOR is set to a value 1.

  In step 5 following step 3 or 4 described above, the flow rate function value FPBAPA is calculated by searching the map shown in FIG. 6 according to the pressure ratio R_PH. The pressure ratio R_PH is a value corresponding to the ratio PBA / HPA between the intake pressure PBA and the estimated atmospheric pressure HPA described above. In this case, the reason for using the pressure ratio R_PH is the same as the reason for using the estimated atmospheric pressure HPA described above, and the estimated atmospheric pressure HPA is a value calculated by the atmospheric pressure estimation process stored in the RAM. Is used.

  Next, the process proceeds to step 6, and the passing air amount GAIRTH is calculated by the above-described equations (11) and (12). In this case, as the square root RTTA and the opening degree function value KTH of the intake air temperature TA, values calculated in a model parameter learning process, which will be described later, stored in the RAM are used. As described above, after calculating the passing air amount GAIRTH in step 6, the present process is terminated.

  The passing air amount GAIRTH calculated in the calculation process of FIG. 5 is used in various control processes executed by the ECU 2. For example, in the fuel injection control process and the ignition timing control process, when calculating the fuel injection amount and the ignition timing under the transient operation condition, it is used to correct the intake air amount GAIR calculated as described later.

  Hereinafter, the model parameter learning process will be described with reference to FIG. This process is to perform identification calculation of the model parameter A of the error model while reflecting the function value error KTHERR, in other words, to learn the identification value of the model parameter A, and the ECU 2 generates the CRK signal a predetermined number of times. It is executed every time.

  As shown in the figure, first, in step 10, it is determined whether or not a learning condition flag F_LEarn is “1”. The learning condition flag F_LEARN indicates whether or not the learning condition for the identification value of the model parameter A is satisfied, and the value is set by the method shown in FIG.

  As shown in FIG. 8, first, at step 20, it is determined whether or not the total travel distance DIST of the vehicle after starting the engine is less than a predetermined value DLEARN. When the determination result is YES, it is determined that the learning condition for the identification value of the model parameter A is satisfied, and the process proceeds to step 21 to set the learning condition flag F_LEarn to “1”. Thereafter, this process is terminated.

  On the other hand, when the determination result in step 20 is NO, it is determined that the learning condition for the identification value of the model parameter A is not satisfied, and the process proceeds to step 22 where the learning condition flag F_LEEARN is set to “0”. Set to. Thereafter, this process is terminated.

Returning to FIG. 7, when the determination result of step 10 is NO and the learning condition for the identification value of the model parameter A is not satisfied, the present process is ended as it is. On the other hand, if the determination result in step 10 is YES and the learning condition for the identification value of the model parameter A is satisfied, the process proceeds to step 11, and the corrected map value KTH_F ( The corrected second opening function value) is calculated.

In the above equation (13), KREF is a predetermined constant (for example, a value of 0.003), and THREF is an opening (for example, 15 degrees) that eliminates the influence of clogging. Further, the clogging coefficient KCLS (clogging degree value) represents the clogging degree of the intake passage 6 and is calculated by a method described later, and KCLS = 1 when the clogging degree of the intake passage 6 is the largest. Further, the map value KTH is calculated by searching the above-described map of FIG. 2 according to the throttle valve opening TH. As shown in the above equation (13), the corrected map value KTH_F is a value reflecting the degree of clogging of the intake passage 6 by correcting the map value KTH with the value KREF · (TH−THREF) ² · KCLS. As well as being calculated, it is calculated as an opening function value in a state where the degree of clogging is the largest when KCLS = 1.

  Next, the process proceeds to step 12, and an identification value FPBAPAini for the flow function value is calculated. This identification value FPBAPAini is searched according to the pressure ratio R_PHi by replacing the flow function value FPBAPA on the vertical axis in FIG. 6 with the identification value FPBAPAini and the pressure ratio R_PH on the horizontal axis with the pressure ratio R_PHi. It is calculated by doing. This pressure ratio R_PHi is calculated as a ratio PBA / HPAini between the intake pressure PBA and the initial estimated atmospheric pressure HPAini, and this initial estimated atmospheric pressure HPAini is a value calculated in the atmospheric pressure estimation process stored in the RAM. Is used. The reason for using this initial estimated atmospheric pressure HPAini will be described later.

Next, in step 13, a square root RTTA of the intake air temperature TA is calculated by searching a map (not shown) according to the intake air temperature TA. Thereafter, in step 14, a model equation value KTHCAL of the opening function value is calculated by the following equation (14).

  In this case, the above equation (14) is obtained by replacing the opening function value KTH with the model equation value KTHCAL, the atmospheric pressure PA with the initial value HPAini of the estimated atmospheric pressure, and the flow function value FPBAPA with the flow function. This corresponds to a value identification value FPBAPAini. The reason why these values HPAini and FPBAPAini are used will be described later.

  In step 15 following step 14, the function value error KTHERR is set to the ratio KTHCAL / KTH_F between the model formula value and the corrected map value.

  Next, the routine proceeds to step 16, where it is determined whether or not the throttle valve opening TH is less than the predetermined opening THB. When this determination result is NO, this process is terminated as it is.

  On the other hand, when the determination result in step 16 is YES, the throttle valve opening TH is in a region where a modeling error occurs due to the first error described above, and the identification calculation of the model parameter A should be performed. After the determination, the process proceeds to step 17, and the identification calculation process of the model parameter A is executed as will be described later. Thereafter, this process is terminated.

  As described above, in this model parameter learning process, the function value error KTHERR is calculated using the initial estimated atmospheric pressure HPAini and the flow function value identification value FPBAPAini when the learning condition flag F_LEarn = 1. The function value error KTHERR is calculated by such a method for the following reason. In other words, in the case of the present embodiment, the function value error KTHERR is calculated using the estimated value of the atmospheric pressure PA. Compensation accuracy (correction accuracy) of the third error is lowered, and as a result, calculated using three values HGAIRTH, GAIR, and PBA calculated while compensating for the first to third errors. The calculation accuracy (estimation accuracy) of the estimated atmospheric pressure HPA also decreases. In this case, in order to eliminate the influence of such an estimation error of the atmospheric pressure PA, the intake pressure PBA calculated from the detection signal of the intake pressure sensor 22 becomes equal to the true value of the atmospheric pressure PA, and the atmospheric pressure PA It is necessary to calculate the function value error KTHERR under the condition that the true value hardly fluctuates. Therefore, in order to achieve this, in the present embodiment, the learning condition flag F_LEARN is used to confirm the establishment of the learning condition, and the function is calculated using the initial estimated atmospheric pressure HPAini and the flow function value identification value FPBAPAini. The value error KTHERR is calculated.

  Next, the identification calculation process of the model parameter A will be described with reference to FIG. In this process, as described below, the throttle valve opening TH is divided into a first region where 0 ≦ TH <THN1, a second region where THN1 ≦ TH <THN2, and a third region where THN2 ≦ TH <THN3. And the fourth region of THN3 ≦ TH <THB, and applying the equal weighting process to these four regions, the model parameter A Is calculated. In this case, THN1 to THN3 are predetermined opening amounts of the throttle valve opening TH, and are set so that 0 <THN1 <THN2 <THN3 <THB.

  As shown in the figure, first, in step 40, the error KTHERCOR is set to the deviation KTHERR-1 between the function value error KTHERR and the value 1. Next, the routine proceeds to step 41, where it is determined whether or not the throttle valve opening TH is less than a predetermined value THN1. When the determination result is YES and the throttle valve opening TH is in the first region, the process proceeds to step 44 and the region value n is set to the value 1.

  On the other hand, when the determination result of step 41 is NO and THN1 ≦ TH, the routine proceeds to step 42, where it is determined whether or not the throttle valve opening TH is less than a predetermined value THN2. When the determination result is YES and the throttle valve opening TH is in the second region, the process proceeds to step 45 and the region value n is set to the value 2.

  On the other hand, when the determination result of step 42 is NO and THN2 ≦ TH, the process proceeds to step 43, where it is determined whether or not the throttle valve opening TH is less than a predetermined value THN3. When the determination result is YES and the throttle valve opening TH is in the third region, the process proceeds to step 46 and the region value n is set to the value 3.

  On the other hand, if the determination result in step 43 is NO and the throttle valve opening TH is in the fourth region, the process proceeds to step 47 and the region value n is set to the value 4. As described above, the region value n is calculated as a value representing the four regions of the throttle valve opening TH.

In step 48 following the above steps 44 to 47, the integral term XXXX [n] of the nth region is calculated by the following equation (15).

  In the above equation (15), the integral term XXXX [n] is a value corresponding to the denominator of the identification calculation equation (9) described above, and the value XXXX [n] z represents the previous value of the integral term. Further, the value n in [] of the integral term XXXX [n] is the above-described region value, and this point is the same in the following description. That is, in step 48, for example, when the region value n = 1, the integral term XXXX [1] of the first region is calculated, and when n = 2, the integral term XXXX [2] of the second region is calculated.

Next, in step 49, the integral term XXY [n] of the nth region is calculated by the following equation (16).

  In the above equation (16), the integral term XXY [n] is a value corresponding to the numerator of the identification calculation equation (9) described above, and the value XXY [n] z represents the previous value of the integral term.

  In step 50 following step 49, the sampling number SAMPL [n] of the nth region is set to the sum of the previous value and the value 1 (SAMPL [n] z + 1). This sampling number SAMPL [n] represents the sampling number of the integral term in the nth region, that is, the number of calculation results.

Next, the routine proceeds to step 51, where the weighted average value XXXTTL is calculated by the following equation (17).

As apparent from reference to the above equation (17), the weighted average value XXXTTL calculates the arithmetic average value of the value (TH-THB) ⁴ for each region, and is equal to the arithmetic average value. It is calculated by performing a weighted average calculation with weighting.

Next, in step 52, the weighted average value XXYTTL is calculated by the following equation (18).

As apparent from reference to the above equation (18), the weighted average value XXYTTL calculates the arithmetic mean value of the value KTHERCOR · (TH−THB) ² for each region, and for these arithmetic mean values, It is calculated by performing a weighted average operation with equal weighting.

In step 53 following step 52, the model parameter A is calculated by the following equation (19), and then this process is terminated.

As described above, in the identification calculation processing of the model parameter A, the four arithmetic average values KTHERCOR · (TH−THB) ² calculated for each region of the throttle valve opening TH are calculated with equal weighting. By performing a weighted average calculation, a weighted average value XXYTTL is calculated, and a weighted average calculation with equal weighting is performed on four arithmetic average values of values (TH-THB) ⁴ calculated for each region. The model parameter A is calculated by calculating the weighted average value XXXTTL and dividing the former by the latter. The advantage of this calculation method will be described with reference to FIGS.

  FIG. 10 shows an example of a calculation result when the error KTHERCOR is calculated using the model parameter A calculated by the above-described identification calculation method of the present embodiment. FIG. 11 shows the above-described formula for comparison. The example of a calculation result at the time of calculating error KTHERCOR using the model parameter A calculated by (9) is shown. In both figures, the data indicated by ■ is the data immediately after the start of calculation, the data indicated by ▽ is the data when the calculation time has passed than the data indicated by ■, and the data indicated by x is more than the data indicated by ▽. Each of them represents a time when the calculation time has further elapsed.

  As is clear from comparison between the two figures, in the case of FIG. 11, the degree to which the calculation result in the region where the sampling data of the throttle valve opening TH is large is reflected in the calculation result of the model parameter A as the calculation time elapses. As a result, the curve indicating the calculation result of the error KTHERCOR changes. That is, it can be seen that the identification accuracy of the model parameter A is lowered. On the other hand, in the case of the method of the present embodiment shown in FIG. 10, even when the calculation time has elapsed, even if the sampling data of the throttle valve opening TH is biased to a certain region, the above-described equal weighting is performed. It can be seen that the curve indicating the calculation result of the error KTHERCOR has not changed because the method is such that the calculation result in the region is reflected in the calculation result of the model parameter A in the same manner as the other regions. That is, it can be seen that high calculation accuracy can be ensured in the identification calculation result of the model parameter A.

  Next, the atmospheric pressure estimation process will be described with reference to FIG. This process calculates the estimated atmospheric pressure HPA and the initial estimated atmospheric pressure HPAini, which is the initial value, as described below, and is executed by the ECU 2 every time the CRK signal is generated a predetermined number of times after the start of cranking. Is done.

  As shown in the figure, first, at step 60, it is determined whether or not an initial pressure calculated flag F_FINHPAINI is “1”. When the determination result is NO, the process proceeds to step 61, and the initial estimated atmospheric pressure HPAini is calculated. In this step 61, the previous value HPAiniz of the initial estimated atmospheric pressure is compared with the intake pressure PBA, and the larger one of them is set as the initial estimated atmospheric pressure HPAini.

  Next, the routine proceeds to step 62, where it is determined whether or not the start mode flag F_STMOD is “1”. The start mode flag F_STMOD is set to “1” until the cranking of the engine 3 is finished, and is set to “0” when the cranking is finished. When the determination result of this step 62 is YES and cranking is being performed, this processing is ended as it is.

  On the other hand, when the determination result in step 62 is NO and the cranking is completed, it is determined that the calculation of the initial estimated atmospheric pressure HPAini should be ended, and the process proceeds to step 63, and the initial pressure calculation is performed to represent it. After the completion flag F_FINHPAINI is set to “1”, this process ends.

  Thus, when the initial pressure calculated flag F_FINHPAINI is set to “1” in step 63, the determination result in step 60 described above becomes YES, and in this case, the process proceeds to step 64 and the estimated atmospheric pressure HPA is set. The calculation process is executed as described below. Thereafter, this process is terminated.

  Next, the calculation process of the estimated atmospheric pressure HPA will be described with reference to FIG. As shown in the figure, first, at step 70, it is determined whether or not an initial setting flag F_FINHPAINIR is “1”. When the current loop is the first control timing of this process, the determination result in step 70 is NO, and in this case, the process proceeds to step 71 where both the estimated atmospheric pressure HPA and the delayed atmospheric pressure HPAD are initially estimated to be large. Set to atmospheric pressure HPAini.

  Next, the process proceeds to step 72, and the initial setting flag F_FINHPAINIR is set to “1”, and then this process is terminated.

  Thus, when the initial setting flag F_FINHPAINIR is set to “1” in step 72, the determination result in step 70 described above becomes YES, and in this case, the process proceeds to step 73 and step 11 in FIG. The post-correction map value KTH_F of the opening function value is calculated by the same method as.

  Next, the routine proceeds to step 74, where the flow function value FPBAPA is calculated. In this step 74, the pressure ratio R_PHD is calculated as the ratio PBA / HPDA of the intake pressure PBA and the delay estimated atmospheric pressure HPAD described above, and the horizontal axis pressure ratio R_PH in the map of FIG. 6 described above is calculated as the pressure ratio R_PHD. By using the map replaced by, and searching this map according to the pressure ratio R_PHD, the flow function value FPBAPA is calculated.

  Next, the process proceeds to step 75, and a square root RTTA of the intake air temperature TA is calculated by searching a map (not shown) in accordance with the intake air temperature TA in the same manner as in step 13 of FIG. Thereafter, in step 76, a map shown in FIG. 14 is searched according to the engine speed NE to calculate a rotation speed correction coefficient KTHNE. This rotation correction coefficient KTHNE is a value for correcting the pressure loss of an air cleaner (not shown) arranged on the upstream side of the air flow sensor 20.

In step 77 following step 76, the estimated passing air amount HGAIRTH is calculated by the following equation (20).

  This equation (20) is obtained by changing the right side passing air amount GAIRTH to the estimated passing air amount HGAIRTH, the left side atmospheric pressure PA and the opening degree function KTH as the delayed estimated atmospheric pressure HPAD and the corrected map value in the above-described equation (6). This corresponds to the replacement with KTH_F.

Next, the routine proceeds to step 78, where the air amount deviation DGAIR is calculated by the following equation (21).

  Next, in step 79, it is determined whether or not the vehicle speed VP is higher than a predetermined vehicle speed VPL. When the determination result is YES, it is determined that the vehicle is traveling, the process proceeds to step 80, and the correction term CORHPA is calculated by searching the map shown in FIG. 15 according to the air amount deviation DGAIR.

  On the other hand, when the determination result of step 79 is NO, it is determined that the vehicle is stopped, the process proceeds to step 81, and the correction term CORHPA is set to 0.

In step 82 following step 80 or 81 described above, the updated estimated atmospheric pressure HPACAL is calculated by the following equation (22).

Next, the routine proceeds to step 83, where the estimated atmospheric pressure HPA is calculated by the weighted average calculation (smoothing calculation) shown in the following equation (23).

  CA1 in the above equation (23) is a weighting coefficient, and is set to a predetermined value such that 0 <CA1 <1. HAPz is the previous value of the estimated atmospheric pressure HPA.

  Next, in step 84, the current value HPA of the estimated atmospheric pressure calculated as described above is set to the delayed estimated atmospheric pressure HPAD, and then the present process is terminated.

  As described above, in the calculation process of the estimated atmospheric pressure HPA, the updated estimated atmospheric pressure HPACAL is calculated by correcting the previous value HPAz of the estimated atmospheric pressure with the correction term CORHPA, and the correction term CORHPA is the air amount deviation. Since it is calculated in accordance with DGAIR, the updated estimated atmospheric pressure HPACAL is calculated such that the air amount deviation DGAIR is zero. In other words, the updated estimated atmospheric pressure HPACAL is calculated so that the estimated passing air amount HGAIRTH coincides with the intake air amount GAIR, and the weighted average calculation of the updated estimated atmospheric pressure HPACAL and the previous value HPaz of the estimated atmospheric pressure is performed. Since the estimated atmospheric pressure HPA is calculated, the estimated atmospheric pressure HPA can be calculated to accurately follow the actual atmospheric pressure PA.

  Hereinafter, the calculation process of the intake air amount GAIR will be described with reference to FIG. As will be described below, this processing calculates the intake air amount GAIR using the output voltage value VAFM of the detection signal of the airflow sensor 20, and is executed by the ECU 2 every time the CRK signal is generated a predetermined number of times. The

  First, in step 90, the basic intake air amount GAIRN is calculated by searching a map shown in FIG. 17 according to the output voltage value VAFM.

  Next, the routine proceeds to step 91, where a value GAIRN / KAFMERR obtained by dividing the basic intake air amount GAIRN by the correction coefficient KAFMERR is set as the intake air amount GAIR. A method for calculating the correction coefficient KAFMERR will be described later. As described above, after calculating the intake air amount GAIR in step 91, the present process is terminated. In the present embodiment, the basic intake air amount GAIRN corresponds to a basic intake parameter, and the correction coefficient KAFMERR corresponds to a correction value.

  The reason why the correction coefficient KAFMERR is used as a value for dividing the basic intake air amount GAIRN in step 91 is the same as the reason why the error correction coefficient KTHCOR described above is calculated as the reciprocal of the function value error KTHERR.

  Note that the intake air amount GAIR calculated in the calculation process of FIG. 16 is used in various control processes executed by the ECU 2. For example, it is used when calculating the fuel injection amount and the ignition timing in the fuel injection control process and the ignition timing control process.

  Next, the calculation process of the correction coefficient KAFMERR will be described with reference to FIG. This process is to calculate the correction coefficient KAFMERR by the above-described method using the function value error KTHERR, and is executed by the ECU 2 every time the CRK signal is generated a predetermined number of times.

  As shown in the figure, first, in step 100, the corrected map value KTH_F of the opening function value is calculated by the same method as in step 11 of FIG.

  Next, the routine proceeds to step 101, where a flow function value FPBAPA is calculated. In this case, the flow function value FPBAPA is obtained by using a map in which the pressure ratio R_PH on the horizontal axis in FIG. 6 described above is replaced with the pressure ratio R_P (= PBA / PA), and searching this map according to the pressure ratio R_P. Calculated.

Next, in step 102, the square root RTTA of the intake air temperature TA is calculated by the same method as in step 13 of FIG. Thereafter, in step 103, a model formula value KTHCAL of the opening function value is calculated by the following formula (24). This equation (24) corresponds to the above-described equation (7) in which KTH on the left side is replaced with KTHCAL.

  In step 104 subsequent to step 103, the function value error KTHERR is set to the ratio KTHCAL / KTH_H between the model formula value and the corrected map value.

  Next, the routine proceeds to step 105, where it is determined whether or not the throttle valve opening TH is less than the predetermined opening THB. When this determination result is YES, this processing is ended as it is.

  On the other hand, when the determination result of step 105 is NO, the process proceeds to step 106 to determine whether or not the pressure ratio R_P is less than the critical pressure ratio R_CRIT. When this determination result is NO, this process is terminated as it is.

On the other hand, when the determination result in step 106 is YES, that is, when THB ≦ TH and R_CRIT <R_P, it is in a region where a modeling error occurs due to the second error described above, and the correction coefficient KAFMERR should be updated. The process proceeds to step 107, and the correction coefficient KAFMERR is calculated by the following equation (25). Thereafter, this process is terminated.

  CA2 in the above equation (25) is a weighting coefficient, and is set to a predetermined value such that 0 <CA2 <1. The weighting factor CA2 may be calculated by a technique for searching a map according to the engine speed NE. In addition, KAFMERRz in Expression (25) is the previous value of the correction coefficient KAFMERR.

  As described above, in the calculation process of FIG. 18, the correction coefficient KAFMERR is calculated by the weighted average calculation of the previous value KAFMERRz and the function value error KTHERR. Therefore, the correction error KAFMERR is represented by the modeling error represented by the function value error KTHERR. Can be calculated as a value reflecting the above. For the same reason, even if the calculation error temporarily increases in the calculation result of the function value error KTHERR for some reason, the correction coefficient KAFMERR can be calculated with high accuracy while avoiding the influence.

  Next, the calculation process of the intake pressure PBA will be described with reference to FIG. As will be described below, this processing calculates the intake pressure PBA using the output voltage value VPBA of the detection signal of the intake pressure sensor 22, and is executed by the ECU 2 every time the CRK signal is generated a predetermined number of times. The

First, at step 110, based on the output voltage value VPBA, the basic intake pressure PBAN is calculated by the following equation (26). In the following expression (26), α, β, and γ are predetermined values.

  Next, the routine proceeds to step 111, where the intake pressure PBA is set to the sum of the basic intake pressure PBAN and the correction term PBAERRCOR (PBAN + PBAERRCOR). A method for calculating the correction term PBAERRRCOR will be described later. As described above, after calculating the intake pressure PBA in step 111, the present process is terminated. In the present embodiment, the basic intake pressure PBAN corresponds to a basic intake parameter, and the correction term PBAERRCOR corresponds to a correction value.

  In this case, the intake pressure PBA calculated in the calculation process of FIG. 19 is used in various control processes executed by the ECU 2. For example, it is used when calculating the fuel injection amount and the ignition timing in the fuel injection control process and the ignition timing control process.

  Next, the calculation process of the correction term PBAERRRCOR will be described with reference to FIG. This process is to calculate the correction term PBAERRCOR by the above-described method using the function value error KTHERR, and is executed by the ECU 2 every time the CRK signal is generated a predetermined number of times.

  As is clear from FIG. 18, steps 120 to 125 of this process are configured in the same manner as steps 100 to 105 in FIG. 18, and therefore the following description will focus on step 126 and subsequent steps. In the case of this process, if the determination result in step 125 is YES and THB ≦ TH, the process proceeds to step 126, where it is determined whether or not the pressure ratio R_P is greater than or equal to the critical pressure ratio R_CRIT. When this determination result is NO, this process is terminated as it is.

  On the other hand, if the determination result in step 126 is YES and R_CRIT ≦ R_P, it is determined that the modeling error is in the region due to the third error described above and the correction term PBAERRCOR should be updated. In step 127, the flow function value correction calculation value FPBAPACOR is set to the ratio FPBAPA / KTHERR between the flow function value and the function value error. The correction function value FPBAPACOR for the flow rate function value is a value representing an error of the flow rate function value FPBAPA caused by the modeling error.

  Next, the routine proceeds to step 128, where the pressure ratio correction calculation value R_PCOR is calculated by searching the map shown in FIG. 21 according to the flow rate function value correction calculation value FPBAPACOR. This map replaces the relationship between the horizontal axis and the vertical axis in the map of FIG. 6 described above, the vertical axis is the pressure ratio correction calculation value R_PCOR, and the horizontal axis is the flow rate function correction calculation value FPBAPACOR. Corresponds to the replacement. The pressure ratio correction calculation value R_PCOR calculated as described above is a value representing an error of the pressure ratio R_P caused by the modeling error.

  Next, in step 129, the correction value PBACOR for intake pressure correction is set to the product PA · R_PCOR of the atmospheric pressure PA and the correction value R_PCOR for pressure ratio. This intake pressure correction calculation value PBACOR is a value representing an error in the intake pressure PBA caused by a modeling error. Further, in step 130 following step 129, the intake pressure error PBAERR is set to a deviation PBACOR-PBAN between the intake pressure correction calculation value PBACOR and the basic intake pressure PBAN.

Next, the process proceeds to step 131, and the correction term PBAERRRCOR is calculated by the following equation (27), and then this process is terminated.

  CA3 in the above equation (27) is a weighting coefficient, and is set to a predetermined value such that 0 <CA3 <1. The weighting factor CA3 may be calculated by a technique for searching a map according to the engine speed NE. In addition, PBAERRRCORz in Expression (27) is the previous value of the correction term PBAERRRCOR.

  As described above, in the calculation process of FIG. 20, the correction term PBAERRRCOR is calculated by the weighted average calculation of the previous value PBAERRCORz and the intake pressure error PBAERR, so that the correction term PBAERRCOR is represented by the modeling error represented by the function value error KTHERR. Can be calculated as a value reflecting the above. For the same reason, even when the calculation error temporarily increases in the calculation result of the function value error KTHERR for some reason, the correction term PBAERRCOR can be calculated with high accuracy while avoiding the influence.

  Next, the idle FB control process executed by the ECU 2 will be described with reference to FIG. This idle FB control process is a feedback control of the throttle valve opening TH so that the engine speed NE converges to the target speed NOBJ during idle operation, and is executed at a predetermined control cycle (for example, 10 msec). The

  As shown in the figure, first, at step 140, it is determined whether or not an idle FB permission flag F_IDLEFB is “1”. The idle FB permission flag F_IDLEFB indicates whether or not an execution condition for the idle FB control process is satisfied. In the setting process (not shown), the operating state of the engine 3 (for example, the throttle valve opening TH, the engine rotation, Based on the number NE and the vehicle speed VP), it is set to “1” when the execution condition of the idle FB control process is satisfied, and “0” when the execution condition is not satisfied.

When the determination result of step 140 is NO, this process is ended as it is. On the other hand, when the determination result in step 140 is YES, it is determined that the idle FB control process should be executed, the process proceeds to step 141, and the previous value IAINz of the integral term is calculated by the following equation (28).

  In the above equation (28), ITW is a water temperature correction term, and is calculated by map search according to the coolant temperature of the engine 3. IXREF is a learning value of the integral term IAIN and is calculated by a method described later. Further, IUP is a rotational speed correction term, and is calculated by map search according to the engine rotational speed NE.

この式（２９）の目標回転数ＮＯＢＪは、図示しない設定処理において、エンジン３の冷却水温度などに応じて設定される。 Next, at step 142, the rotational speed deviation DNE is calculated by the following equation (29).

The target rotational speed NOBJ of the equation (29) is set according to the coolant temperature of the engine 3 or the like in a setting process (not shown).

この式（３０）のＫＩは積分項ゲインであり、エンジン３の運転状態（冷却水温度など）に応じてマップ検索により算出される。 Next, the process proceeds to step 143, and the integral term IAIN is calculated by the following equation (30).

KI in this equation (30) is an integral term gain, and is calculated by map search according to the operating state (cooling water temperature, etc.) of the engine 3.

この式（３１）のＫＰは比例項ゲインであり、エンジン３の運転状態（冷却水温度など）に応じてマップ検索により算出される。 In step 144 following step 143, the proportional term IP is calculated by the following equation (31).

KP in this equation (31) is a proportional term gain, and is calculated by map search according to the operating state of the engine 3 (cooling water temperature or the like).

この式（３２）において、ＫＰは比例項ゲインであり、エンジン３の運転状態（冷却水温度など）に応じてマップ検索により算出される。また、ＤＮＥｚは、回転数偏差ＤＮＥの前回値である。 Next, in step 145, the differential term ID is calculated by the following equation (32).

In this equation (32), KP is a proportional term gain, and is calculated by map search according to the operating state (cooling water temperature, etc.) of the engine 3. DNEz is the previous value of the rotational speed deviation DNE.

Next, the routine proceeds to step 146, where the feedback correction term IFB is calculated. Specifically, the feedback correction term IFB is finally calculated by calculating the feedback correction term IFB by the following equation (33) and applying a limit process to the calculated value.

  As described above, the feedback correction term IFB is calculated as a value for causing the engine speed NE to converge to the target speed NOBJ by a control algorithm to which the PID control algorithm is applied.

Next, at step 147, the control input ICMD is calculated by the following equation (34).

  In this equation (34), ILOAD is a load correction term and is calculated according to the auxiliary machine load and the like. KIPA and IPA are an atmospheric pressure correction coefficient and an atmospheric pressure correction term, respectively, and are calculated according to the atmospheric pressure PA.

  In step 148 following step 147, it is determined whether or not the learning value calculation permission flag F_IXREF is “1”. This learning value calculation permission flag F_IXREF indicates whether or not the calculation condition for the learning value IXREF of the integral term is satisfied. In the setting process (not shown), the operating state of the engine 3 (for example, the throttle valve opening TH The engine speed NE and the vehicle speed VP are set to “1” when the calculation condition for the learning value IXREF is satisfied, and “0” when the calculation condition is not satisfied.

When the determination result of step 148 is NO, this process is ended as it is. On the other hand, if the determination result in step 148 is YES, it is determined that the learning value IXREF should be calculated, the process proceeds to step 149, and the learning value IXREF (feedback correction amount) is calculated. Specifically, the learning value IXREF is calculated by the weighted average calculation represented by the following equation (35), and the learning value IXREF is finally calculated by performing limit processing on the calculated value.

  In this equation (35), CB1 is a weighting factor and is set to a predetermined value such that 0 <CB1 <1. IXREFz is the previous value of the learning value IXREF. As described above, after the learning value IXREF is calculated in step 149, the present process is terminated.

  In the idle FB control process of FIG. 22, when the control input ICMD is calculated as described above, a control input signal corresponding to the control input ICMD is supplied to the TH actuator 7b. Thereby, the throttle valve opening TH is feedback controlled so that the engine speed NE converges to the target speed NOBJ by the effect of the feedback correction term IFB included in the control input ICMD.

  Next, the clogging coefficient calculation process executed by the ECU 2 will be described with reference to FIGS. As will be described below, this calculation process calculates a clogging coefficient KCLS as a value representing the degree of clogging of the intake passage 6, and is executed at a predetermined control cycle (for example, 10 msec).

  As shown in FIG. 23, first, in step 160, it is determined whether or not the calculation permission flag F_KCLS is “1”. The calculation permission flag F_KCLS indicates whether or not the calculation condition for the clogging coefficient KCLS is satisfied. In the setting process (not shown), the calculation condition for the clogging coefficient KCLS is satisfied based on the presence or absence of blowby gas leak. If the calculation condition is not satisfied, the value is set to “1”.

  When the determination result of step 160 is NO, this process is ended as it is. On the other hand, if the determination result in step 160 is YES, it is determined that the clogging coefficient KCLS should be calculated, and the process proceeds to step 161, where the previous value KCLSz of the clogging coefficient is stored in the RAM. Set to.

  Next, in steps 162 and 163, the reference upper limit value thmax and the reference lower limit value thmin of the throttle valve opening are calculated by searching the map shown in FIG. 24 according to the learned value IXREF. In the drawing, a reference upper limit value thmax indicated by a broken line represents a value of the throttle valve opening with respect to the learned value IXREF when the intake passage 6 is in a clogged state where the throttle valve 7a can be controlled. On the other hand, the reference lower limit value thmin indicated by the solid line represents the value of the throttle valve opening with respect to the learned value IXREF when the intake passage 6 is not clogged. Therefore, as the degree of clogging of the intake passage 6 increases, the value of the throttle valve opening with respect to the learned value IXREF changes from the reference lower limit value thmin toward the reference upper limit value thmax.

  In FIG. 24, IXBASE is a predetermined threshold value of the learning value IXREF. This threshold value IXBASE indicates that the intake passage 6 is clogged when IXREF> IXBASE is satisfied. It is set to a value that can be estimated. THMAX and THMIN are predetermined values of the reference upper limit value thmax and the reference lower limit value thmin, respectively, when IXREF = IXBASE.

この式（３６）から明らかなように、スロットル弁開度の暫定値ｔｈｔｍｐは、詰まり係数の前回値ＫＣＬＳｚを重み係数とする、基準上限値ｔｈｍａｘおよび基準下限値ｔｈｍｉｎの加重平均演算によって算出される。 In step 164 following step 163, a temporary value thtmp of the throttle valve opening is calculated by the following equation (36).

As is apparent from this equation (36), the provisional value thtmp of the throttle valve opening is calculated by a weighted average calculation of the reference upper limit value thmax and the reference lower limit value thmin using the previous value KCLSz of the clogging coefficient as a weighting factor. .

この式（３７）から明らかなように、詰まり係数の暫定値ｋｃｌｓｔｍｐは、スロットル弁開度の暫定値ｔｈｔｍｐが基準下限値ｔｈｍｉｎから基準上限値ｔｈｍａｘに近づくほど（すなわち、吸気通路６の詰まり度合がより大きくなるほど）、値０から値１により近づくように算出される。 Next, the routine proceeds to step 165, where the provisional value kclstmp of the clogging coefficient is calculated by the following equation (37).

As is apparent from this equation (37), the clogging coefficient provisional value kclstmp increases as the throttle valve opening provisional value thtmp approaches the reference upper limit value thmax from the reference lower limit value thmin (that is, the degree of clogging of the intake passage 6 decreases). The larger the value), the closer it is to the value 1 from the value 0.

  Next, in step 166, the clogging coefficient KCLS is calculated by the following equations (38) to (42), and then the present process is terminated.

  In the above equation (38), KCLS_L represents a lower limit threshold value, and CREF represents a predetermined value. The lower limit threshold value KCLS_L is set to a value of 0 when the calculation result according to the equation (38) is less than the value of 0. Further, KCLS_H in the above equation (39) represents an upper threshold value, and this upper threshold value KCLS_H is set to a value 1 when the calculation result by the equation (39) exceeds a value 1. As is clear from the equations (40) to (42), the clogging coefficient KCLS is calculated by applying limit processing with the lower limit threshold value KCLS_L as the lower limit and the upper limit threshold value KCLS_H as the upper limit to the provisional value kclstmp. .

  As described above, in this embodiment, the clogging coefficient KCLS is calculated by the method shown in FIG. This is due to the following reason. That is, as described above, in the idle FB control process, the feedback correction term IFB included in the control input ICMD is calculated as a value for converging the engine speed NE to the target speed NOBJ. In this case, the integral term IAIN included in the feedback correction term IFB is calculated using the learning value IXREF, and the learning value IXREF is calculated by a weighted average calculation of the integral term IAIN and the previous value IXREFz of the learning value. Therefore, when the clogging of the intake passage 6 occurs, the learning value IXREF increases, and as the clogging degree increases, the learning value IXREF increases more. Therefore, by using the learned value IXREF highly correlated with the degree of clogging of the intake passage 6, the clogging coefficient KCLS can be calculated as a value that appropriately represents the degree of clogging of the intake passage 6. For this reason, in this embodiment, the clogging coefficient KCLS is calculated using the method shown in FIG.

As described above, according to the intake parameter calculating apparatus 1 of the present embodiment, the corrected map value KTH_F is obtained by correcting the map value KTH with the value KREF · (TH−THREF) ² · KCLS including the clogging coefficient KCLS. Since it is calculated, the corrected map value KTH_F can be calculated as a value reflecting the degree of clogging of the intake passage 6. Thereby, even when the above-described clogging error occurs due to clogging of the intake passage 6, the corrected map value KTH_F can be calculated with high accuracy while avoiding the influence of this clogging error.

  Further, when calculating the passing air amount GAIRTH, the function value error KTHERR is calculated as a ratio of the model formula value KTHCAL and the corrected map value KTH_F, and thus is calculated as representing the modeling error of the model formula (6). Is done. In the region of TH <THB, the model parameter A is identified on-board using such a function value error KTHERR, and the correction coefficient KTHCOR is calculated using the model parameter A identified on-board. Since the passing air amount GAIRTH is calculated by correcting the basic passing air amount GAIRTHN with the correction coefficient KTHCOR, the passing air amount GAIRTH is caused by the modeling error of the model equation (6) and the clogging of the intake passage 6. It is possible to calculate the corrected clogging error as a corrected value. Thereby, the passing air amount GAIRTH can be calculated with high accuracy.

  In addition, in the calculation formulas (11) and (12) for the passing air amount GAIRTH, the estimated atmospheric pressure HPA is used instead of the atmospheric pressure PA. The passing air amount GAIRTH can be calculated while avoiding the influence, and the calculation accuracy can be further improved.

  Further, since the model parameter A is identified on-board, the error model equation (8) is an actual value between the error KTHERCOR and the throttle valve opening TH due to the secular change in the throttle valve 7a and the variation between individuals. Even when there is a deviation from the above relationship, that is, when a modeling error occurs, such modeling error can be quickly compensated by using the on-board identified model parameter A, and the error model equation (8) The actual relationship between KTHERCOR and throttle valve opening TH can be quickly adapted. Thereby, the correction accuracy by the correction coefficient KTHCOR can be improved, and the calculation accuracy of the passing air amount GAIRTH can be further improved.

Further, in the identification calculation of the model parameter A, the weighted average calculation with equal weighting is performed on the four arithmetic average values of the value KTHERCOR · (TH−THB) ² calculated for each region of the throttle valve opening TH. Thus, the weighted average value XXYTTL is calculated, and the weighted average value XXXTTL is obtained by performing a weighted average operation with equal weighting on the four arithmetic average values of the values (TH-THB) ⁴ calculated for each region. And the model parameter A is calculated by dividing the former by the latter. Thus, even if the sampling data of the throttle valve opening TH is biased to a certain area during the identification calculation of the model parameter A, the degree to which the calculation result in that area is reflected in the calculation result of the model parameter A Can be made the same as other regions, and high calculation accuracy can be secured in the calculation result of the model parameter A.

  Further, when the total travel distance DIST of the vehicle after starting the engine is less than the predetermined value DLEARN, the identification calculation of the model parameter A is executed using the initial estimated atmospheric pressure HPAini instead of the atmospheric pressure PA, and DIST ≧ DLEARN In this case, the identification calculation of the model parameter A is prohibited. In this case, when the total mileage of the vehicle after the engine is started is small, the throttle valve opening TH changes in the low opening range, and the frequency of becoming TH <THB is high. The frequency can be increased and the calculation accuracy can be further improved.

  In addition to this, since the initial estimated atmospheric pressure HPAini is used in place of the atmospheric pressure PA in the calculation of the model value KTHCAL of the opening function value, the function value even under conditions where the atmospheric pressure PA is likely to change. The error KTHERR can be calculated with high accuracy, whereby the calculation accuracy of the model parameter A can be further improved. As described above, the calculation accuracy of the correction coefficient KTHCOR, that is, the calculation accuracy of the passing air amount GAIRTH can be further improved by ensuring high calculation accuracy in the calculation of the model parameter A.

  On the other hand, since the identification calculation of the model parameter A is stopped in the region where THB ≦ TH, unnecessary identification calculation in a region where no modeling error occurs can be avoided, and the calculation load can be reduced.

  Further, when calculating the intake air amount GAIR, the intake air amount GAIR is calculated by correcting the basic intake air amount GAIRN with the correction coefficient KAFMERR. When THB ≦ TH and R_P <R_CRIT, That is, when it is estimated that the modeling error of the model equation (6) occurs due to the second error described above, the function value error KTHERR representing the modeling error and the previous value KAFMERRz of the correction coefficient And a weighted average calculation. Accordingly, the correction coefficient KAFMERR can be calculated as a value reflecting the modeling error represented by the function value error KTHERR, and the intake air amount GAIR can be calculated with high accuracy.

  Further, when calculating the intake pressure PBA, the intake pressure PBA is calculated by correcting the basic intake pressure PBAN with the correction term PBAERRRCOR. This correction term PBAERRCOR is calculated when R_CRIT ≦ R_P, that is, the third error described above. Because of this, when it is estimated that the modeling error of the model equation (6) has occurred, it is calculated using the function value error KTHERR representing the modeling error. It can be calculated as a value reflecting the modeling error represented by the value error KTHERR, and the intake pressure PBA can be calculated with high accuracy.

  Further, since the estimated passing air amount HGAIRTH is calculated using the intake pressure PBA and the intake air amount GAIR corrected as described above and the correction coefficient KTHCOR described above, the calculation accuracy of the estimated passing air amount HGAIRTH is increased. Can be improved. Further, the correction term CORHPA is calculated using the air amount deviation DGAIR, which is a value obtained by subtracting the intake air amount GAIR from the estimated passing air amount HGAIRTH, and the estimated atmospheric pressure HPA is updated using the correction term CORHPA. Therefore, the calculation accuracy can be improved. In addition, since the passing air amount GAIRTH is calculated using such estimated atmospheric pressure HPA, the calculation accuracy of the passing air amount GAIRTH can be further improved.

  In the embodiment, as the identification calculation method of the model parameter A, two weighted average values XXXTTL and XXYTTL are calculated by the above-described equations (17) and (18) in steps 51 and 52 of FIG. Thus, although the method of calculating the model parameter A by the above-described equation (19) is used, the following identification calculation method may be used instead.

That is, after calculating two weighted average values XXXTTL and XXYTTL in steps 51 and 52 using the following equations (43) and (44) instead of the equations (17) and (18) described above, The model parameter A may be calculated by the equation (19) described above in FIG.

KG1 to KG4 in the above equations (43) and (44) are weighting factors, and are set so that KG1>KG2>KG3> KG4 is satisfied and KG1 + KG2 + KG3 + KG4 = 1 is satisfied. As is apparent from the above equation (43), the weighted average value XXXTTL is calculated by performing a weighted average operation on the arithmetic average value of the values (TH-THB) ^{4 in} the ^four regions. The weight coefficients KG1 to KG4 are set to larger values as the throttle valve opening TH is smaller. The weighted average value XXYTTL is also calculated by performing a weighted average operation on the arithmetic average value of the values KTHERCOR · (TH-THB) ² in the four regions, and the weighting coefficients KG1 to KG4 are: It is set to the same value as the weighted average value XXXTTL.

  In this case, as described above, the function value error KTHERR is caused by the deviation of the opening area of the throttle valve 7a with respect to the standard product (reference product), and the influence degree of the deviation is determined by the throttle valve opening degree. The smaller TH is, the larger it is. Therefore, by setting the four weighting factors KG1 to KG4 as described above, it is possible to identify the model parameter A while reflecting the degree of influence of such deviation, and thereby to calculate the correction coefficient KTHCOR. Can be improved. The weighting factors KG1 to KG4 may be set so that any one of them has the same value.

The embodiment is an example in which an error model equation of the form Y = A · X ² is used, but the error model equation of the present invention is not limited to this, and Y = a · X ² + b · X + c. Or Y = a · X + b may be used. Note that when these types of error model expressions are used, compared to the case where the error model expression (8) of the embodiment is used, from the viewpoint of achieving both improvement in calculation accuracy and reduction in calculation load, the embodiment The error model equation (8) is better.

  Furthermore, the embodiment is an example in which the identification calculation of the model parameter A is executed when the total travel distance DIST of the vehicle after starting the engine is less than the predetermined value DLEARN. The execution condition is not limited to this, and any condition may be used as long as the model parameter A can be accurately identified. For example, it may be determined that the operation time of the engine 3 after the start does not elapse for a predetermined time as an execution condition of the identification calculation of the model parameter A.

The embodiment is an example in which the clogging coefficient KCLS is used as the clogging degree value. However, the clogging degree value of the present invention is not limited to this and may be a value representing the clogging degree of the intake passage. For example, the reciprocal number (1 / KCLS) of the clogging coefficient KCLS may be used as the clogging degree value. In this case, the product of the map value KTH and the value KREF · (TH−THREF) ^{2 is} the reciprocal number (1 / KCLS). ) To calculate the corrected map value KTH_F.

  Furthermore, the embodiment is an example in which the learning value IXREF of the integral term is used as the feedback correction amount, but the feedback correction amount of the present invention is not limited to this, and is a value for feedback control of the opening degree of the intake throttle valve. If it is. For example, the integral term IAIN or the feedback correction term IFB may be used as the feedback correction amount. In this case, a map in which the learning value IXREF on the horizontal axis in FIG. 24 is replaced with the integral term IAIN or the feedback correction term IFB is used. Thus, the reference upper limit value thmax and the reference lower limit value thmin may be calculated. Even in such a configuration, when the degree of clogging of the intake passage 6 increases, the integral term IAIN or the feedback correction term IFB increases accordingly, so the same effect as in the embodiment can be obtained.

  On the other hand, the embodiment is an example using the map of FIG. 2 as the correlation model, but the correlation model of the present invention is not limited to this, and the correlation between the opening of the intake throttle valve and the opening function value. As long as it represents. For example, a mathematical expression defining the correlation between the opening of the intake throttle valve and the opening function value may be determined by offline identification, and such a mathematical expression may be used as the correlation model.

  The embodiment is an example in which the throttle valve 7a is used as the intake throttle valve. However, the intake throttle valve of the present invention is not limited to this, and the amount of air passing through the intake throttle valve can be changed without being limited thereto. Any valve can be used.

1 Intake parameter calculation device 2 ECU (basic intake parameter calculation means, first opening function value calculation means, second opening function value calculation means, clogging degree value calculation means, corrected second opening function value calculation means, correction (Value calculation means, intake parameter calculation means)
3 Internal combustion engine 6 Intake passage 7a Throttle valve (intake throttle valve)
NE speed
NOBJ target speed
PA atmospheric pressure (upstream pressure)
TH Opening of throttle valve (opening of intake throttle valve)
Model value of KTHCAL opening function value (first opening function value)
KTH Opening function value map value (second opening function value)
IXREF Learning value of integral term (feedback correction amount)
KCLS Clogging factor (clogging degree value)
KTH_F Corrected map value (corrected second opening function value)
KTHERR function value error (function value ratio)
GAIRTH Passing air volume (intake parameter)
GAIRTHN Basic passing air volume (Basic intake parameters)
KTHCOR correction coefficient (correction value)
A Model parameters
GAIR intake air volume (intake parameters)
GAIRN Basic intake air volume (Basic intake parameters)
KAFMERR correction coefficient (correction value)
PBA intake pressure (intake parameters, downstream pressure)
PBAN Basic intake pressure (Basic intake parameters)
PBAERRCOR correction term (correction value)

Claims (1)

The amount of air passing through the intake throttle valve is changed by the intake throttle valve provided in the intake passage as the passing air amount, and the intake throttle valve of the intake throttle valve is converged to the target rotational speed so as to converge to the target rotational speed. In an internal combustion engine whose opening degree is feedback-controlled, an intake air parameter calculation device for an internal combustion engine that calculates an intake air parameter representing a state of air in the intake passage,
Basic intake parameter calculation means for calculating a basic intake parameter as a basic value of the intake parameter;
An upstream pressure, which is a pressure in the intake passage on the upstream side of the intake throttle valve, and a downstream pressure, which is a pressure in the intake passage on the downstream side of the intake throttle valve, derived by a predetermined modeling method; A first opening function value as a first calculated value of the opening function value using a model formula that defines a relationship between the opening function value determined by the opening of the intake throttle valve and the passing air amount First opening function value calculating means for calculating
A second opening function value for calculating a second opening function value as a second calculated value of the opening function value using a correlation model representing a correlation between the opening degree of the intake throttle valve and the opening function value. Degree function value calculating means,
A clogging degree value calculating means for calculating a clogging degree value representing the clogging degree of the intake passage according to a feedback correction amount for feedback control of the opening degree of the intake throttle valve;
A corrected second opening degree function value calculating means for calculating a corrected second opening degree function value by correcting the calculated second opening degree function value using the calculated clogging degree value;
Correction value calculation means for calculating a correction value using a function value ratio that is a ratio between one of the calculated first opening function value and the calculated second opening function value after correction;
An intake parameter calculating means for calculating the intake parameter by correcting the basic intake parameter with the calculated correction value;
An intake parameter calculation device for an internal combustion engine, comprising:

Images