Control System for Internal Combustion Engine

Abstract

To provide a control system of an internal combustion engine having an air amount sensor for detecting an intake amount as a detected air amount, and a control device for controlling the internal combustion engine on the basis of an air amount correlation value correlated with the detected air amount, and acquiring an air amount correlation value almost equal to an actual air amount even when detection accuracy of the air amount sensor is deteriorated.SOLUTION: A control device acquires a value equal to a ratio of a learning time volumetric efficiency value with respect to a confirming time volumetric efficiency value as a correction coefficient, and acquires a corrected air amount obtained by multiplying the detected air amount by the correction coefficient as an air amount correlation value. The learning time volumetric efficiency value is a volumetric efficiency value correlated with a ratio of an actual air amount being an actual intake amount with respect to a rotation speed of an internal combustion engine and not correlated with a temperature and a pressure of intake air, which is obtained by applying the detected air amount instead of the actual air amount when a learning condition is established. The confirming time volumetric efficiency value is a volumetric efficiency value which is obtained by applying the detected air amount instead of the actual air amount when a confirming condition is established after the learning time volumetric efficiency value is acquired.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to a control system for an internal combustion engine. For example, it relates to a control system for an internal combustion engine provided with an air flow meter (air quantity sensor) that detects the amount of air flowing through an intake passage (intake air quantity).

Background Art

[0002] In this type of control system for an internal combustion engine, it is desirable to accurately acquire the intake air quantity. Therefore, one of the conventional control systems (air quantity calculation device) eliminates the influence of the harmonics included in the pulsating detection signal input from the air flow meter and then acquires the amplitude median value (average flow rate) between the maximum value and the minimum value as the intake air quantity (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By the way, the detection accuracy of the intake air quantity of the air flow meter may decrease over time. That is, due to the secular deterioration of the air flow meter, the magnitude of the difference between the actual intake air quantity (actual air quantity) and the intake air quantity detected by the air flow meter (detected air quantity) may gradually increase. Specifically, the detected air quantity represented by the detection signal output from the air flow meter for a certain actual air quantity may increase or decrease due to secular deterioration. If the detection accuracy of the intake air quantity decreases, there is a possibility that it will interfere with the control of the internal combustion engine.

[0005] The present invention has been devised in view of such points, and an object thereof is to provide a control system for an internal combustion engine that can obtain an air amount correlation value substantially equal to the actual air amount even when the detection accuracy of an air flow meter decreases.

Means for Solving the Problems

[0006] To solve the above problems, a vehicle control system according to a first invention of the present invention includes an air amount sensor that detects the amount of intake air flowing through an intake passage of an internal combustion engine as a detected air amount, and a control device that controls the internal combustion engine based on an air amount correlation value correlated with the detected air amount. The control device acquires a value equal to the ratio of a learning-time volumetric efficiency value to a confirmation-time volumetric efficiency value as a correction coefficient, and acquires a corrected air amount obtained by multiplying the detected air amount by the correction coefficient as the air amount correlation value.

[0007] The learning-time volumetric efficiency value is a volumetric efficiency value that is correlated with the ratio of the actual air amount, which is the amount of intake air actually flowing through the intake passage, to the rotational speed of the internal combustion engine and is not correlated with the temperature and pressure of the intake air flowing into the cylinders of the internal combustion engine, and is obtained by applying the detected air amount instead of the actual air amount when a predetermined learning condition is satisfied. The confirmation-time volumetric efficiency value is a value obtained by applying the detected air amount instead of the actual air amount to the volumetric efficiency value when a predetermined confirmation condition is satisfied after the learning-time volumetric efficiency value is obtained.

[0008] A second invention of the present invention is the vehicle control system according to the first invention, wherein the control device determines that the learning condition is satisfied when a predetermined steady condition is satisfied, including that the variation in the actual air amount is likely to be small when the difference between the actual air amount and the detected air amount is likely to be small, and determines that the confirmation condition is satisfied when the steady condition is satisfied after the learning-time volumetric efficiency value is obtained.

[0009] A third invention of the present invention is a vehicle control system according to the first invention, wherein the control device acquires the correction coefficient for each of a plurality of operating states related to the internal combustion engine, and multiplies the detected air amount by the correction coefficient selected according to the operating state from among the plurality of acquired correction coefficients to obtain the air amount correlation value.

Advantages of the Invention

[0010] In the vehicle control system according to the first invention, the volumetric efficiency value can be treated as a value that does not change even after a relatively long period of time has elapsed (i.e., even if the detection accuracy of the air amount sensor has deteriorated over time). If, as a result of the detection accuracy of the air amount sensor deteriorating, the detected air amount is larger than the actual air amount when compared to the time when the learning-time volumetric efficiency value was obtained at the time when the confirmation-time volumetric efficiency value was obtained, the confirmation-time volumetric efficiency value will be larger than the learning-time volumetric efficiency value.

[0011] In other words, the ratio of the confirmation-time volumetric efficiency value to the learning-time volumetric efficiency value (air amount ratio) represents the degree to which the detection accuracy of the air amount sensor has deteriorated. Therefore, the air amount correlation value obtained by multiplying the detected air amount by the correction coefficient (i.e., the reciprocal of the air amount ratio) is a value that has restored the degree to which the detection accuracy of the air amount sensor has deteriorated, and is likely to be a value approximately equal to the actual air amount. Therefore, according to the first invention, even when the detection accuracy of the air amount sensor (air flow meter) has deteriorated, it is possible to obtain an air amount correlation value approximately equal to the actual air amount and control the internal combustion engine based on the air amount correlation value.

[0012] In the second invention, the learning-time volumetric efficiency value is obtained before the detection accuracy of the air amount sensor deteriorates. As described above, the air amount ratio represents the degree to which the detection accuracy of the air amount sensor has deteriorated during the period from the time when the learning-time volumetric efficiency value was obtained to the time when the confirmation-time volumetric efficiency value was obtained. Therefore, according to the second invention, the difference between the actual air amount and the air amount correlation value is likely to become very small.

[0013] In the third invention, even if the difference between the actual air quantity and the detected air quantity (i.e., the detection error of the air quantity sensor) changes according to the operating state of the internal combustion engine, a correction coefficient corresponding to the detection error of the air quantity sensor is obtained for each operating state. Therefore, according to the third invention, regardless of the operating state of the internal combustion engine, the difference between the actual air quantity and the air quantity correlation value is likely to become very small.

Brief Description of the Drawings

[0014]

Figure 1

Figure 2

Figure 3

Figure 4

Embodiments for Carrying Out the Invention

[0015] Embodiments of the present invention will be described with reference to FIGS. 1 to 4. The same reference numerals in the description mean the same elements having the same functions without redundant description. The control system for the internal combustion engine according to the present embodiment is applied to the vehicle 1 shown in FIG. 1. The vehicle 1 includes an internal combustion engine 2 as a driving force source, a supercharger 3, an intake system 4, an exhaust system 5, an EGR system 6, and an ECU 7. Note that the illustration of the appearance of the vehicle 1 is omitted.

[0016] The internal combustion engine 2 is a multi-cylinder diesel engine. The internal combustion engine 2 includes a plurality of fuel injection valves 21. Each of the fuel injection valves 21 injects high-pressure fuel supplied from a pressure accumulation chamber of a common rail device (not shown) into the cylinder according to an instruction from the ECU 7.

[0017] The turbocharger 3 includes a turbine 31, a variable nozzle mechanism 32, a nozzle actuator 32a, and a compressor 33. The turbine 31 is operated (rotated) by the pressure of exhaust gas (combustion gas) discharged from each cylinder of the internal combustion engine 2. The variable nozzle mechanism 32 is equipped with a nozzle vane, which is a throttle mechanism for the exhaust gas flowing into the turbine 31.

[0018] Specifically, the opening degree of the exhaust flow passage in the turbine 31 (or the closing degree, i.e., the degree to which the flow passage that introduces exhaust gas into the turbine 31 is closed by the variable nozzle mechanism 32) changes according to the nozzle closing degree Vn, which is the state of the nozzle vane. The nozzle actuator 32a changes the nozzle closing degree Vn between a predetermined fully open position (fully open state) and a fully closed position (fully closed state) in accordance with instructions from the ECU 7. The compressor 33 operates (rotates) in conjunction with the turbine 31, and pressurizes the air (intake air) that is drawn into each cylinder of the internal combustion engine 2.

[0019] The intake system 4 includes intake pipes 41a-41b, which are intake passages, an intake manifold 42, an intercooler 43, a throttle valve 44, and a throttle actuator 44a. The intake pipe 41a introduces intake air (fresh air) drawn in from the outside (outside the vehicle) into the compressor 33. The intake pipe 41b introduces intake air discharged from the compressor 33 into the intake manifold 42. The intake manifold 42 introduces intake air into each cylinder of the internal combustion engine 2. The intake pipes 41a-41b and the intake manifold 42 form an intake path for the internal combustion engine 2.

[0020] The intercooler 43 is disposed in the intake pipe 41b. The intercooler 43 cools the intake air that has been pressurized by the compressor 33 and has a raised temperature. The throttle valve 44 is disposed in the intake pipe 41b at a position downstream of the intercooler 43. The throttle valve 44 adjusts the opening of the intake pipe 41b according to its rotational position. The throttle actuator 44a changes the rotational position (valve opening state) of the throttle valve 44 between a predetermined fully open position (fully open state) and a fully closed position (fully closed state) in accordance with a command from the ECU 7.

[0021] The exhaust system 5 includes an exhaust manifold 51, exhaust pipes 52a to 52b, and an exhaust gas purification device 53. The exhaust manifold 51 and the exhaust pipes 52a to 52b form an exhaust path of the internal combustion engine 2.

[0022] The exhaust manifold 51 introduces exhaust (combustion gas) discharged from each cylinder of the internal combustion engine 2 into the exhaust pipe 52a. The exhaust pipe 52a introduces the exhaust into the turbine 31. The exhaust pipe 52b discharges the exhaust discharged from the turbine 31 to the outside (outside the vehicle). The exhaust gas purification device 53 is interposed in the exhaust pipe 52b. The exhaust gas purification device 53 includes a well-known oxidation catalyst, DPF (Diesel Particulate Filter), SCR (Selective Catalytic Reduction), etc., and purifies the exhaust.

[0023] The EGR system 6 includes an EGR pipe 61, an EGR cooler 62, a bypass pipe 63, an EGR valve 64, and an EGR bypass valve 65. The EGR pipe 61 communicates the exhaust pipe 52a and the intake pipe 41b. The EGR cooler 62 is disposed in the EGR pipe 61. The EGR cooler 62 cools the high-temperature exhaust (i.e., EGR gas) flowing in from the exhaust pipe 52a. The bypass pipe 63 communicates a position upstream (branch position) and a position downstream (confluence position) of the EGR cooler 62 in the EGR pipe 61 to form a flow path of the EGR gas that does not pass through the EGR cooler 62.

[0024] The EGR valve 64 is interposed at a position on the intake pipe 41b side rather than at the confluence position with the bypass pipe 63 in the EGR pipe 61. The valve opening state of the EGR valve 64 changes between a predetermined fully open position (fully open state) and a fully closed position (fully closed state) according to an instruction from the ECU 7. As a result, the amount of the EGR gas refluxing from the exhaust pipe 52a to the intake pipe 41b is adjusted. The amount of the EGR gas is represented by an EGR rate Er, which is the amount of the EGR gas with respect to the amount of the gas flowing into the cylinder of the internal combustion engine 2 (i.e., the sum of the amount of intake air (fresh air) represented by the actual air amount Ga described later and the amount of the EGR gas).

[0025] The EGR bypass valve 65 is disposed at the branch position with the bypass pipe 63 in the EGR pipe 61. The valve opening state of the EGR bypass valve 65 changes between a predetermined fully open position (fully open state) and a fully closed position (fully closed state) according to an instruction from the ECU 7. As a result, the amount of EGR gas that does not pass through the EGR cooler 62 is adjusted.

[0026] The ECU 7 is an electronic control unit (Electronic Control Unit) including a CPU, a ROM, a RAM, and an EEPROM, and is a control device (control unit) of the control system of the internal combustion engine 2. The CPU reads data, performs numerical calculations, outputs calculation results, etc. by sequentially executing a predetermined program. The ROM stores programs and maps (look-up tables) etc. executed by the CPU. The RAM temporarily stores data referred to by the CPU. The EEPROM stores data referred to by the CPU, and further retains the stored data even when the ECU 7 stops operating.

[0027] The ECU 7 is connected to a crank angle sensor 71, a cam position sensor 72, an air flow meter 73, a nozzle closing degree sensor 74, pressure sensors 75a to 75d, temperature sensors 76a to 76b, and an accelerator opening sensor 77.

[0028] The crank angle sensor 71 outputs a pulse signal to the ECU 7 every time the crankshaft (not shown) of the internal combustion engine 2 rotates by a predetermined angle. The cam position sensor 72 outputs a signal corresponding to the rotational position of the camshaft (not shown) of the internal combustion engine 2 to the ECU 7. The ECU 7 acquires the engine rotational speed NE [rpm] of the internal combustion engine 2 based on the signal input from the crank angle sensor 71. The ECU 7 acquires the crank angle CA of a specific cylinder provided in the internal combustion engine 2 based on the signals input from the crank angle sensor 71 and the cam position sensor 72.

[0029] The air flow meter 73 detects the amount of intake air flowing through the intake pipe 41a as the detected air amount Gd [g / sec], and outputs a signal representing the detected air amount Gd to the ECU 7. The air flow meter 73 is also referred to as an "air amount sensor" for convenience. The nozzle closing degree sensor 74 detects the nozzle closing degree Vn and outputs a signal representing the nozzle closing degree Vn to the ECU 7.

[0030] The pressure sensor 75a detects the atmospheric pressure Po, which is the pressure of the air around the vehicle 1, and outputs a signal representing the atmospheric pressure Po to the ECU 7. The pressure sensor 75b is disposed at a position downstream of the air flow meter 73 in the intake pipe 41a. The pressure sensor 75b detects the intake pressure Pa, which is the pressure of the intake air flowing into the compressor 33, and outputs a signal representing the intake pressure Pa to the ECU 7.

[0031] The pressure sensor 75c is disposed at a position between the compressor 33 and the intercooler 43 in the intake pipe 41b. The pressure sensor 75c detects the difference between the pressure of the intake air flowing out of the compressor 33 and the atmospheric pressure Po as the supercharging pressure Pb, and outputs a signal representing the supercharging pressure Pb to the ECU 7. Therefore, the sum of the atmospheric pressure Po detected by the pressure sensor 75a and the supercharging pressure Pb detected by the pressure sensor 75c is equal to the pressure of the intake air flowing out of the compressor 33.

[0032] The pressure sensor 75d is disposed in the intake manifold 42. The pressure sensor 75d detects the difference between the pressure of the intake air flowing into the cylinder of the internal combustion engine 2 and the atmospheric pressure Po as the in-manifold pressure Pi, and outputs a signal representing the in-manifold pressure Pi to the ECU 7.

[0033] The temperature sensor 76a is disposed at a position near the air flow meter 73 in the intake pipe 41a. The temperature sensor 76a detects the intake temperature To [°C], which is the temperature of the intake air flowing through the intake pipe 41a, and outputs a signal representing the intake temperature To to the ECU 7. The temperature sensor 76b is disposed in the intake manifold 42. The temperature sensor 76b detects the in-manifold temperature Ti [°C], which is the temperature of the intake air flowing into the cylinder of the internal combustion engine 2, and outputs a signal representing the in-manifold temperature Ti to the ECU 7.

[0034] The accelerator opening sensor 77 detects the accelerator pedal opening Ap, which is the opening of an accelerator pedal (not shown) that a driver of the vehicle 1 operates to control the acceleration of the vehicle 1, and outputs a signal representing the accelerator pedal opening Ap to the ECU 7. When the driver accelerates the vehicle 1 (i.e., when the required load on the internal combustion engine 2 increases), the accelerator pedal opening Ap increases.

[0035] The ECU 7 controls the engine speed NE based on the accelerator pedal opening Ap. When increasing the engine speed NE, the ECU 7 increases the fuel injection amount Qi, which is the amount of fuel injected from the fuel injection valve 21. Further, the ECU 7 increases the supercharging pressure Pb to increase the intake air amount (the amount of inhaled air) flowing into the cylinders of the internal combustion engine 2.

[0036] When the fuel injection amount Qi is determined, the ECU 7 determines a fuel injection pattern according to the fuel injection amount Qi, the fuel injection pattern including a (plurality of) combination of the timing at which the fuel represented by the crank angle CA is injected and the fuel injection amount at that timing. The ECU 7 causes the fuel injection valve 21 of each cylinder to inject fuel according to the fuel injection pattern.

[0037] Furthermore, the ECU 7 controls the nozzle closing degree Vn, the EGR rate Er, etc. based on the corrected air amount Gm. The corrected air amount Gm is obtained by multiplying the detected air amount Gd by a correction coefficient K (described later) (i.e., Gm = K × Gd). The corrected air amount Gm is treated as a value substantially equal to the actual air amount Ga, which is the amount of intake air actually flowing through the intake pipe 41a. The corrected air amount Gm is also referred to as the "air amount correlation value" for convenience.

[0038] For example, the ECU 7 changes the nozzle closing degree Vn (i.e., controls the nozzle actuator 32a) by a well-known process based on the corrected air amount Gm, the intake pressure Pa, the supercharging pressure Pb, etc. In addition, the ECU 7 changes the EGR rate Er (i.e., controls the EGR valve 64) by a well-known method based on the corrected air amount Gm and the fuel injection amount Qi, etc.

[0039] (Correction Coefficient Update Process) Next, the "Correction Coefficient Acquisition Process" executed by the ECU 7 to acquire (update) the correction coefficient K will be described with reference to FIGS. 2 to 3. The air flow meter 73 may deteriorate over time, and the detection accuracy of the detected air quantity Gd may decrease (that is, the detection error of the air flow meter 73 may increase). Specifically, the detected air quantity Gd detected by the air flow meter 73 may become a value less than the actual air quantity Ga or a value greater than the actual air quantity Ga due to aging deterioration. Therefore, the ECU 7 acquires the corrected air quantity Gm based on the correction coefficient K as described above, and controls the internal combustion engine 2 based on the corrected air quantity Gm.

[0040] The ECU 7 acquires the correction coefficient K for each operating state Se(x, y) of the internal combustion engine 2 specified by the combination of the engine speed NE and the fuel injection quantity Qi. Hereinafter, when it is explicitly stated that the correction coefficient K is acquired for a specific operating state Se(x, y), it is also referred to as the correction coefficient K(x, y). On the other hand, the operating state Se(x, y) is also simply referred to as the operating state Se.

[0041] Specifically, the operating state Se is divided for each combination of the range of the engine speed NE (that is, the range of the values of the engine speed NE specified by the lower limit value and the upper limit value) and the range of the fuel injection quantity Qi. An example of the correction coefficient K(x, y) acquired for each operating state Se is shown in FIG. 2. The operating state Se surrounded by the thick solid line in FIG. 2 will be described later.

[0042] For example, when the engine speed NE is included in the range from the rotational speed n2 to the rotational speed n3 and the fuel injection quantity Qi is included in the range from the injection quantity q1 to the injection quantity q2, the internal combustion engine 2 is in the operating state Se(3, 2). In this case, the ECU 7 acquires the corrected air quantity Gm by multiplying the coefficient k32, which is the value of the correction coefficient K(3, 2) for the operating state Se(3, 2), by the detected air quantity Gd (that is, Gm = k32 × Gd).

[0043] The initial value of each of the correction coefficients K(x, y) is "1". When the difference between the actual air quantity Ga and the detected air quantity Gd becomes relatively large, the ECU 7 updates (changes) the value of the correction coefficient K to a value other than "1" by the process described later.

[0044] The ECU 7 updates the correction coefficient K based on the volumetric efficiency value Ve [g / m 3 . The volumetric efficiency value Ve is obtained (calculated) by substituting the actual air quantity Ga, the engine rotational speed NE, the atmospheric pressure Po, the intake manifold pressure Pi, the intake air temperature To, the intake manifold temperature Ti, and the total displacement Vc [cc] of the internal combustion engine 2 into the following formula (1). Hereinafter, the engine rotational speed NE, the atmospheric pressure Po, the intake manifold pressure Pi, the intake air temperature To, and the intake manifold temperature Ti are also collectively referred to as "volumetric efficiency parameters". As understood from formula (1), the volumetric efficiency value Ve correlates (is proportional) to the ratio of the actual air quantity Ga to the engine rotational speed NE (i.e., Ga / NE).

Equation

[0045] The difference between the intake manifold pressure Pi and the atmospheric pressure Po, and the difference between the intake manifold temperature Ti and the intake air temperature To are mainly due to the pressurization by the compressor 33 and the cooling by the intercooler 43. The intake air quantity changes according to the changes in the intake manifold pressure Pi and the intake manifold temperature Ti respectively.

[0046] On the other hand, the volumetric efficiency value Ve is the intake air quantity per cycle (and per total displacement Vc), and is a value obtained by canceling out the effect of the increase in the intake air quantity due to the increase in the intake manifold pressure Pi and the effect of the decrease in the intake air quantity due to the increase in the intake manifold temperature Ti (i.e., volume expansion). That is, the volumetric efficiency value Ve is obtained as a value that does not correlate with the intake manifold pressure Pi and the intake manifold temperature Ti.

[0047] Therefore, the volumetric efficiency value Ve can be treated as a value that does not change even when the intake manifold pressure Pi, the intake manifold temperature Ti, etc. change. Furthermore, the volumetric efficiency value Ve obtained based on the actual air quantity Ga can be treated as a value that does not change even after a relatively long period of time has elapsed (i.e., even when the detection accuracy of the air flow meter 73 decreases due to aging deterioration). Therefore, the volumetric efficiency value Ve obtained at the timing before the accuracy of the detected air quantity Gd decreases and the volumetric efficiency value Ve obtained at the timing when the accuracy of the detected air quantity Gd has decreased (for example, the confirmation timing described later) can be treated as equal values to each other.

[0048] The update process of the correction coefficient K based on the volumetric efficiency value Ve will be described in more detail. The ECU 7 executes "learning processing" at the timing before the elapsed time Te (integrated time) since the internal combustion engine 2 first starts operating after the vehicle 1 is manufactured becomes longer than a predetermined time threshold Tth and when the "steady state condition" is first established.

[0049] The steady state condition is established when the EGR valve 64 is in the fully closed state (i.e., the EGR rate Er is "0") and the change amount per unit time of each of the engine rotational speed NE and the detected air quantity Gd continues to be smaller than a predetermined value (specifically, a minute value) for a predetermined time. In other words, the steady state condition is established when the fluctuation of the actual air quantity Ga is small and the EGR gas does not flow into the cylinders of the internal combustion engine 2, so the air quantity flowing into each cylinder (i.e., the intake air quantity) is equal to the actual air quantity Ga.

[0050] The learning processing is a process of storing the volumetric efficiency parameter and the detected air quantity Gd in the EEPROM for each operating state Se(x,y). Hereinafter, each of the volumetric efficiency parameters and the detected air quantity Gd stored by the learning processing is also referred to as the learning-time rotational speed NE1, the learning-time atmospheric pressure Po1, the learning-time intake manifold pressure Pi1, the learning-time intake air temperature To1, the learning-time intake manifold temperature Ti1, and the learning-time detected air quantity Gd1.

[0051] The volumetric efficiency value Ve obtained by applying these parameters to Equation (1) is also referred to as the in - learning volumetric efficiency value Ve1 for convenience. Therefore, the relationship between these parameters and the in - learning volumetric efficiency value Ve1 can be expressed by the following Equation (2). Note that the ECU 7 does not actually acquire the in - learning volumetric efficiency value Ve1, the volumetric efficiency value Ve2m at the time of correction confirmation (see Equation (3)), the volumetric efficiency value Ve1m at the time of correction learning (see Equation (4)), and the volumetric efficiency value Ve2a at the time of actual confirmation (see Equation (7)).

Number

[0052] The timing at which the learning process is executed for a certain operating state Se(x, y) is hereinafter also referred to as the "learning timing". The conditions that are satisfied when the learning timing arrives are also referred to as the "learning conditions" for convenience. The time threshold Tth is pre - adapted so that if the elapsed time Te is shorter than the time threshold Tth, the detection accuracy of the air flow meter 73 is likely not to have deteriorated.

[0053] Therefore, at the learning timing, the difference between the detected air quantity Gd and the actual air quantity Ga is likely to be minute. Accordingly, the difference between the volumetric efficiency value Ve obtained based on Equation (1) and the in - learning volumetric efficiency value Ve1 obtained based on Equation (2) is likely to be small.

[0054] After the learning timing arrives for a certain operating state Se(x, y), when the steady - state conditions are satisfied while the internal combustion engine 2 is in that operating state Se(x, y), the ECU 7 acquires the corrected air quantity Gd1m. The method for acquiring the corrected air quantity Gd1m will be described below.

[0055] The timing at which the corrected air quantity Gd1m is obtained for a certain operating state Se(x,y) is hereinafter also referred to as the "confirmation timing". The conditions that are satisfied when the confirmation timing arrives are, for convenience, also referred to as the "confirmation conditions". Each of the volumetric efficiency parameters at the confirmation timing is also referred to as the confirmation rotation speed NE2, the confirmation atmospheric pressure Po2, the confirmation intake manifold pressure Pi2, the confirmation intake air temperature To2, and the confirmation intake manifold temperature Ti2.

[0056] By applying these parameters to Equation (1), the corrected volumetric efficiency value Ve2m shown in the following Equation (3) can be obtained. Note that, instead of the actual air quantity Ga in Equation (1), the detected air quantity Gd2 at the confirmation timing, which is the detected air quantity Gd at the confirmation timing, is applied in Equation (3).

Equation

[0057] On the other hand, by applying the volumetric efficiency parameters and the corrected air quantity Gd1m at the learning timing, which are obtained for the operating state Se(x,y) at the confirmation timing, to Equation (1), the corrected volumetric efficiency value Ve1m shown in the following Equation (4) can be obtained.

Equation

[0058] By treating the corrected volumetric efficiency value Ve1m and the corrected volumetric efficiency value Ve2m at the confirmation timing as equal values to each other, the following Equation (5a) is obtained. Further, by solving Equation (5a) for the corrected air quantity Gd1m, the following Equation (5b) is obtained.

Equation

[0059] Note that, except when the interpolation process at the confirmation timing described later is executed, the ECU 7 acquires the corrected intake air amount Gd1m based on the following equation (6) instead of equation (5b). That is, the learning-time atmospheric pressure Po1 and the confirmation-time atmospheric pressure Po2 are regarded as equal to each other (that is, Po2 / Po1 = 1). In addition, the learning-time rotational speed NE1 and the confirmation-time rotational speed NE2 obtained for the same operating state Se(x, y) are regarded as equal to each other (that is, NE1 / NE2 = 1).

Equation

[0060] When the corrected intake air amount Gd1m is acquired, the ECU 7 acquires the ratio of the corrected intake air amount Gd1m to the learning-time detected intake air amount Gd1 as the air amount ratio Ra (that is, Ra = Gd1m / Gd1). The air amount ratio Ra can be treated as a value approximately equal to the ratio of the confirmation-time detected intake air amount Gd2 to the actual intake air amount Ga at the confirmation timing (hereinafter also referred to as the confirmation-time air amount Ga2) (that is, Ra = Gd2 / Ga2).

[0061] More specifically, as shown in the following equation (7), by applying the volumetric efficiency parameter at the confirmation timing and the confirmation-time air amount Ga2 to equation (1), the actual confirmation-time volumetric efficiency value Ve2a, which is the volumetric efficiency value Ve at the confirmation timing, can be acquired.

Equation

[0062] As understood from equations (3) and (7), the ratio of the corrected confirmation-time volumetric efficiency value Ve2m to the actual confirmation-time volumetric efficiency value Ve2a is equal to the ratio of the confirmation-time detected intake air amount Gd2 to the confirmation-time air amount Ga2 (that is, Ve2m / Ve2a = Gd2 / Ga2). On the other hand, since the corrected learning-time volumetric efficiency value Ve1m is treated as a value equal to the corrected confirmation-time volumetric efficiency value Ve2m, the ratio of the corrected intake air amount Gd1m to the learning-time detected intake air amount Gd1 is equal to the ratio of the confirmation-time detected intake air amount Gd2 to the confirmation-time air amount Ga2 (that is, Ra = Gd1m / Gd1 = Gd2 / Ga2).

[0063] In other words, when the difference between the confirmed air quantity Ga2 and the detected air quantity Gd2 at the time of confirmation (i.e., the detection error of the air flow meter 73 at the confirmation timing) increases, the difference between the corrected air quantity Gd1m and the detected air quantity Gd1 during learning increases. Therefore, the air quantity ratio Ra, which is the ratio of the corrected air quantity Gd1m to the detected air quantity Gd1 during learning, is a value representing the degree to which the detection error of the air flow meter 73 increases during the period from the learning timing to the confirmation timing. In other words, if the detection error of the air flow meter 73 has not increased, the air quantity ratio Ra becomes a value approximately equal to "1".

[0064] On the other hand, when the detection error of the air flow meter 73 increases, the air quantity ratio Ra becomes a value deviated from "1". If the air quantity ratio Ra is smaller than a predetermined lower threshold value Rthd smaller than "1", or if the air quantity ratio Ra is larger than a predetermined upper threshold value Rthu larger than "1" (i.e., Ra < Rthd < 1 or 1 < Rthu < Ra), the ECU 7 updates the correction coefficient K(x, y). Specifically, the ECU 7 sets the correction coefficient K(x, y) to a value equal to the reciprocal of the air quantity ratio Ra (i.e., K = 1 / Ra).

[0065] (Supplementary Explanation of the Correction Coefficient - Regarding the Correction Coefficient) The correction coefficient K obtained (updated) by the above-described process is equal to the ratio of the learning-time volumetric efficiency value Ve1 (obtained by applying the detected air quantity Gd1 during learning) to the corrected confirmation-time volumetric efficiency value Ve2m (obtained by applying the detected air quantity Gd2 at the time of confirmation). More specifically, as described above, since Ra = Gd1m / Gd1 and K = 1 / Ra, the correction coefficient K can be expressed by the following formula (8). [Number]

[0066] On the other hand, the ratio of the in - learning volumetric efficiency value Ve1 to the volumetric efficiency value Ve2m at the time of correction confirmation can be expressed by the following formula (9) based on formula (2) and formula (7). As can be understood from formula (8) and formula (9), the correction coefficient K is equal to the ratio of the in - learning volumetric efficiency value Ve1 to the volumetric efficiency value Ve2m at the time of correction confirmation.

Number

[0067] (Correction Coefficient Update Process - Interpolation Process at Confirmation Timing) As described above, when the steady - state condition is satisfied before the elapsed time Te reaches the time threshold value Tth, if the learning process corresponding to the driving state Se at that time has not been executed yet, the ECU7 executes the learning process. In other words, there is a possibility that the elapsed time Te reaches the time threshold value Tth before the learning process is executed for all driving states Se.

[0068] Therefore, after the elapsed time Te reaches the time threshold value Tth, when the steady - state condition is satisfied while in the driving state Se where the learning process has not been executed, the ECU7 executes the correction coefficient update process by referring to the volumetric efficiency parameter, etc. corresponding to the driving state Se in which the in - learning volumetric efficiency value Ve1 was obtained. Hereinafter, the driving state Se for which the learning process has been executed is also referred to as the learned driving state Sea. On the other hand, the driving state Se for which the learning process was not executed before the elapsed time Te reached the time threshold value Tth is also referred to as the unlearned driving state Seb.

[0069] When the steady - state condition is satisfied while in the unlearned driving state Seb (that is, when the confirmation timing arrives for the unlearned driving state Seb), the ECU7 selects one learned driving state Sea. In addition, the ECU7 obtains the corrected air amount Gd1m by applying the volumetric efficiency parameter obtained for the selected learned driving state Sea and the current volumetric efficiency parameter, etc. to formula (6). Further, the ECU7 obtains the air amount ratio Ra based on the obtained corrected air amount Gd1m and the in - learning detected air amount Gd1 obtained for the learned driving state Sea.

[0070] A method for selecting a learned operation state Sea will be described. When steady-state conditions are satisfied while the ECU 7 is in the unlearned operation state Seb, the ECU 7 selects the unlearned operation state Seb that is closest to the unlearned operation state Seb (for example, an adjacent unlearned operation state Seb). When there are multiple learned operation states Sea that are closest, the ECU 7 preferentially selects the learned operation state Sea in which the range of the engine rotational speed NE is the same.

[0071] Here, assume that the operation state Se(x, y) surrounded by the thick solid line in FIG. 2 is the unlearned operation state Seb, and the other operation states Se(x, y) are the learned operation states Sea. For example, when steady-state conditions are satisfied while in the unlearned operation state Seb(3, 1), the operation states Se(1, 1) and Se(3, 3) are the closest learned operation states Sea.

[0072] In this case, the ECU 7 selects the learned operation state Sea(3, 3) in which the unlearned operation state Seb(3, 1) and the range of the engine rotational speed NE (specifically, the range where the engine rotational speed NE is from the rotational speed n2 to the rotational speed n3) are common. In addition, the ECU 7 obtains the corrected air amount Gd1m by applying the volumetric efficiency parameter and the like obtained by the learning process for the operation state Se(3, 3) to Equation (6). Further, the ECU 7 obtains the air amount ratio Ra based on the obtained corrected air amount Gd1m and the detected air amount Gd1 during learning for the operation state Se(3, 3).

[0073] On the other hand, when steady-state conditions are satisfied while in the unlearned operation state Seb(2, 1), the ECU 7 selects the learned operation state Sea(1, 1). Thus, if the selected learned operation state Sea and the unlearned operation state Seb (i.e., the current operation state Se) are different in the range of the engine rotational speed NE, the ECU 7 obtains the corrected air amount Gd1m based on the following Equation (10). That is, in this case, the rotational speed NE1 during learning and the rotational speed NE2 during confirmation are treated as different values from each other.

Equation

[0074] (Corrected air quantity acquisition process) As described above, the ECU 7 updates the correction coefficient K corresponding to each driving state Se according to the detection error of the air flow meter 73. In other words, for a certain driving state Se, the confirmation timing arrives and the correction coefficient K is updated, while for other driving states Se, the confirmation timing does not arrive (that is, the steady state condition is not satisfied), and the correction coefficient K may not be updated.

[0075] That is, there may be a coexistence of a driving state Se in which the correction coefficient K is updated to a value different from "1" as a result of a relatively large detection error of the air flow meter 73, and a driving state Se in which the correction coefficient K has not been changed from "1". In such a case, when the internal combustion engine 2 is in the driving state Se where the correction coefficient K is equal to "1", the ECU 7 obtains the corrected air quantity Gm by referring to a correction coefficient K different from "1" obtained for other driving states Se.

[0076] More specifically, when the internal combustion engine 2 is in the driving state Se where the correction coefficient K is equal to "1", the ECU 7 selects the driving state Se that is different from "1" and is the closest. In addition, the ECU 7 obtains the corrected air quantity Gm based on the correction coefficient K (that is, a value different from "1") obtained for the selected driving state Se. When there are a plurality of driving states Se that are different from "1" and are the closest, the ECU 7 preferentially selects the driving state Se in which the range of the engine rotational speed NE is the same.

[0077] (Specific operation) Next, the specific operation of the ECU 7 will be described. The CPU of the ECU 7 (hereinafter also referred to as the unit "CPU") executes each of the "correction coefficient update processing routine" and the "corrected air quantity acquisition processing routine" represented by flowcharts in FIGS. 3 and 4 every time a predetermined time period elapses.

[0078] When it is an appropriate timing, the CPU starts processing from step 300 in FIG. 3 and proceeds to step 305 to determine whether or not the internal combustion engine 2 is operating and steady conditions are satisfied. That is, the CPU determines whether or not a state where the EGR valve 64 is fully closed and the change amount per unit time of each of the engine rotational speed NE and the detected air amount Gd is smaller than a predetermined value continues for a predetermined time or more. If the steady conditions are not satisfied, the CPU determines "No" at step 305 and directly proceeds to step 395 to end the processing of this routine.

[0079] On the other hand, if the steady conditions are satisfied, the CPU determines "Yes" at step 305 and proceeds to step 310 to specify the operating state Se(x, y) of the internal combustion engine 2 at the current time based on the engine rotational speed NE and the fuel injection amount Qi.

[0080] Next, the CPU proceeds to step 315 to determine whether or not learning processing is being executed for the operating state Se(x, y) at the current time. That is, the CPU determines whether or not the volumetric efficiency parameter and the detected air amount Gd1 acquired at the learning timing of the operating state Se(x, y) are stored in the EEPROM.

[0081] If the learning processing is not being executed, the CPU determines "No" at step 315 and proceeds to step 320 to determine whether or not the elapsed time Te is shorter than the time threshold Tth. Note that when the internal combustion engine 2 mounted on the vehicle 1 first starts operating, the CPU executes a routine (not shown) to store the operation start date and time in the EEPROM. Further, when executing step 320, the CPU determines whether or not the difference between the current date and time and the operation start date and time (that is, the elapsed time Te) is larger than the time threshold Tth.

[0082] If the elapsed time Te is shorter than the time threshold Tth, the CPU determines "Yes" at step 320 and proceeds to step 325 to execute learning processing. That is, the CPU stores the volumetric efficiency parameter and the detected air amount Gd at the current time in the EEPROM. Next, the CPU proceeds to step 395.

[0083] On the other hand, if the elapsed time Te is equal to or greater than the time threshold Tth, the CPU determines "No" in step 320 and proceeds to step 330, where it selects one of the operating states Se(x, y) in which the learning process has been executed. That is, the CPU selects the operating state Se(x, y) that has executed the learning process and is closest to the current operating state Se(x, y). If there are multiple operating states Se(x, y) that have executed the learning process and are closest to the current operating state Se(x, y), the CPU preferentially selects the operating state Se(x, y) in which the range of the engine rotational speed NE is the same.

[0084] Next, the CPU proceeds to step 335 and obtains the corrected air quantity Gd1m based on Equation (6) or Equation (10). In this case, the CPU obtains the corrected air quantity Gd1m based on the volumetric efficiency parameter obtained for the selected operating state Se(x, y). Further, the CPU proceeds to step 340 and obtains the ratio of the corrected air quantity Gd1m to the detected air quantity Gd1 during learning as the air quantity ratio Ra.

[0085] Next, the CPU proceeds to step 345 and determines whether the air quantity ratio Ra is less than the lower threshold Rthd or greater than the upper threshold Rthu. That is, the CPU determines whether the detection error of the air flow meter 73 is relatively large. If the air quantity ratio Ra is less than the lower threshold Rthd or greater than the upper threshold Rthu, the CPU determines "Yes" in step 345 and proceeds to step 350, where it sets the correction coefficient K to a value equal to the reciprocal of the air quantity ratio Ra. Further, the CPU proceeds to step 395.

[0086] On the other hand, if the air quantity ratio Ra is included in the range from the lower threshold Rthd to the air quantity ratio Ra, the CPU determines "No" in step 345 and proceeds directly to step 395. That is, in this case, the value of the correction coefficient K is not updated and is maintained at "1".

[0087] If the determination condition in step 315 is satisfied (that is, if the learning process has already been executed), the CPU determines "Yes" in step 315 and directly proceeds to step 335. That is, in this case (different from the case where "No" is determined in step 320 described above), the corrected air amount Gd1m is obtained based on the volumetric efficiency parameter acquired for the current operating state Se(x,y).

[0088] Next, the corrected air amount acquisition processing routine in FIG. 4 will be described. At an appropriate timing, the CPU starts the processing from step 400 in FIG. 4 and proceeds to step 405, and specifies the current operating state Se(x,y) by the same processing as in step 310 in FIG. 3.

[0089] Next, the CPU proceeds to step 410 and determines whether the value of the correction coefficient K corresponding to the current operating state Se(x,y) has been updated to a value other than "1". If the value of the correction coefficient K has been updated to a value other than "1", the CPU determines "Yes" in step 410 and proceeds to step 415, and obtains the corrected air amount Gm based on the correction coefficient K corresponding to the current operating state Se(x,y). That is, the CPU obtains the corrected air amount Gm by multiplying the detected air amount Gd by the correction coefficient K (in this case, a value other than "1").

[0090] Next, the CPU proceeds to step 495 and ends the processing of this routine. The CPU controls the internal combustion engine 2 based on the corrected air amount Gm by executing a routine (not shown). Specifically, the CPU controls the nozzle closing degree Vn, the EGR rate Er, etc. based on the corrected air amount Gm

[0091] On the other hand, if the value of the correction coefficient K corresponding to the current operating state Se(x, y) is "1" (i.e., if it has not been updated to a value other than "1"), the CPU determines "No" in step 410 and proceeds to step 420 to determine whether there exists an operating state Se(x, y) in which the correction coefficient K has been updated to a value other than "1". That is, the CPU determines whether, while the detection error of the air flow meter 73 is relatively large, the steady-state conditions are not satisfied for the current operating state Se(x, y) and the confirmation timing has not arrived.

[0092] If there exists an operating state Se(x, y) in which the correction coefficient K has been updated to a value other than "1", the CPU determines "Yes" in step 420 and proceeds to step 425 to select one of the operating states Se(x, y) in which the correction coefficient K has been updated (changed). That is, the CPU selects the operating state Se(x, y) in which the correction coefficient K has been updated and is closest to the current operating state Se(x, y). If there are multiple operating states Se(x, y) in which the correction coefficient K has been updated and is closest to the current operating state Se(x, y), the CPU preferentially selects the operating state Se(x, y) in which the range of the engine rotational speed NE is the same.

[0093] Next, the CPU proceeds to step 430 and obtains the corrected air amount Gm based on the correction coefficient K corresponding to the selected operating state Se(x, y). That is, the CPU obtains the corrected air amount Gm by multiplying the detected air amount Gd by the correction coefficient K (in this case, a value other than "1"). Further, the CPU proceeds to step 495.

[0094] On the other hand, if there does not exist an operating state Se(x, y) in which the correction coefficient K has been updated to a value other than "1", the CPU determines "No" in step 420 and proceeds to step 415. In this case, since the correction coefficient K is "1", the corrected air amount Gm is set to a value equal to the detected air amount Gd.

[0095] As described above, according to the control system of the internal combustion engine according to the present embodiment, even if the difference between the actual air quantity Ga and the detected air quantity Gd becomes relatively large due to the secular deterioration of the air flow meter 73, it is possible to control the internal combustion engine 2 based on the corrected air quantity Gm that is highly likely to be substantially equal to the actual air quantity Ga. If the EGR valve 64 is controlled based on the detected air quantity Gd (instead of the corrected air quantity Gm) when the detection accuracy of the air flow meter 73 decreases, the actual EGR rate Er may deviate from the target value, and the amount of nitrogen oxides (NOx) or soot contained in the exhaust gas of the internal combustion engine 2 may increase. On the other hand, when the EGR valve 64 is controlled based on the corrected air quantity Gm, the possibility of avoiding the occurrence of such events is increased.

[0096] In addition, since the ECU 7 acquires the correction coefficient K based on the volumetric efficiency parameter at the learning timing and the volumetric efficiency parameter at the confirmation timing, etc., no new sensor or the like is required to acquire the corrected air quantity Gm. In particular, since the learning process is executed before the detection accuracy of the air flow meter 73 decreases, it is possible to acquire the corrected air quantity Gm as a value with a small difference from the actual air quantity Ga.

[0097] Furthermore, since the actual air quantity Ga changes when the engine rotational speed NE and / or the fuel injection quantity Qi change, the ECU 7 has acquired the correction coefficient K for each operating state Se specified by the combination of the engine rotational speed NE and the fuel injection quantity Qi. Therefore, even if the magnitude of the difference between the actual air quantity Ga and the detected air quantity Gd changes according to the operating state Se (i.e., the actual air quantity Ga), it is possible to acquire the corrected air quantity Gm as a value with a small difference from the actual air quantity Ga regardless of the operating state Se (i.e., even if the actual air quantity Ga changes).

[0098] As described above, the embodiments of the present invention have been described with reference to the above structure, but it is obvious to those skilled in the art that many alternations, improvements, and changes are possible without departing from the object of the present invention. Therefore, the forms of the present invention may include all alternations, improvements, and changes that do not depart from the spirit and object of the appended claims. The forms of the present invention are not limited to the above special structure, and can be changed, for example, as follows.

[0099] Atmospheric pressure Po was applied when calculating the volumetric efficiency value Ve. Instead, intake pressure Pa may be applied when calculating the volumetric efficiency value Ve. That is, intake pressure Pa may be used as the volumetric efficiency parameter instead of atmospheric pressure Po. Similarly, when calculating the volumetric efficiency value Ve, supercharging pressure Pb may be used instead of intake manifold pressure Pi. In other words, the volumetric efficiency value Ve may be obtained based on the volumetric efficiency parameter or a similar parameter and the actual air quantity Ga, as long as it does not change with the aging deterioration of the air flow meter 73 and the changes in the temperature and pressure of the intake air flowing into the cylinders of the internal combustion engine 2.

[0100] The steady state conditions included that the EGR valve 64 was in a fully closed state. Instead, the steady state conditions may not include that the EGR valve 64 is in a fully closed state. In this case, the ECU 7 may obtain the amount of EGR gas flowing into the cylinders of the internal combustion engine 2 (i.e., the difference between the intake air quantity and the actual air quantity Ga) by a well-known method based on the volumetric efficiency parameter and the opening degree of the EGR valve 64, etc., and treat the value obtained by applying the sum of the actual air quantity Ga and the EGR gas quantity instead of the actual air quantity Ga to Equation (1) as the volumetric efficiency value Ve.

[0101] In addition, the steady state conditions included that the change amount per unit time of the engine rotational speed NE and the detected air quantity Gd continued for a predetermined time in a state smaller than a predetermined value. Instead, the steady state conditions may include that the change amount per unit time of the detected air quantity Gd continues for a predetermined time in a state smaller than a predetermined value. That is, the change in the engine rotational speed NE may not be considered for the establishment of the steady state conditions.

[0102] ECU7 executed learning processing at a timing before the elapsed time Te became longer than the time threshold value Tth, when steady conditions were established for a certain driving state Se. In other words, there may be a driving state Se in which the learning processing is not executed. Instead of this, ECU7 may establish a steady state for each of the driving states Se immediately after the manufacture of the vehicle 1 in sequence and execute the learning processing. That is, in this case, the learning processing is executed for all the driving states Se. This processing may be executed with the vehicle 1 placed on a predetermined test device.

[0103] ECU7 had acquired (a plurality of) correction coefficients K for each of the driving states Se. Instead of this, ECU7 may acquire one correction coefficient K. Or, ECU7 may acquire a correction coefficient K for each of the driving states classified only by the engine rotational speed NE (regardless of the fuel injection amount Qi). Further, ECU7 may acquire a correction coefficient K for each of the driving states classified by the detected air amount Gd.

[0104] ECU7 had updated the correction coefficient K when the air amount ratio Ra was smaller than the lower threshold value Rthd, or when the air amount ratio Ra was larger than the upper threshold value Rthu. Instead of this, ECU7 may update the correction coefficient K to a value equal to the reciprocal of the air amount ratio Ra each time the air amount ratio Ra is acquired. Or, when the air amount ratio Ra is acquired a predetermined number of times for a certain driving state Se (that is, when the confirmation timing arrives a predetermined number of times), ECU7 may update the correction coefficient K according to the average value of the plurality of acquired air amount ratios Ra.

[0105] The elapsed time Te had represented the time elapsed since the time when the internal combustion engine 2 mounted on the vehicle 1 first started operating. Instead of this, the elapsed time Te may be the total sum of the time during which the internal combustion engine 2 is operating.

[0106] ECU7 included an EEPROM as a non-volatile memory. Instead of this, ECU7 may include a non-volatile memory other than the EEPROM (for example, a flash memory). Further, the processing realized by the above-described ECU7 may be executed by a plurality of ECUs. In addition, the internal combustion engine 2 was a compression ignition type diesel engine. Instead of this, the internal combustion engine 2 may be a spark ignition type gasoline engine.

Explanation of Signs

[0107] 1…Vehicle 2…Internal combustion engine 3…Supercharger 4…Intake system 5…Exhaust system 6…EGR system 7…ECU 21…Fuel injection valve 31…Turbine 32…Variable nozzle mechanism 32a…Nozzle actuator 33…Compressor 41a, 41b…Intake pipes 42…Intake manifold 43…Intercooler 44…Throttle valve 44a…Throttle actuator 51…Exhaust manifold 52a, 52b…Exhaust pipes 53…Exhaust gas purification device 61…EGR pipe 62…EGR cooler 63…Bypass pipe 64…EGR valve 65…EGR bypass valve 71…Crank angle sensor 72…Cam position sensor 73…Air flow meter 74…Nozzle closing degree sensor 75a~75d…Pressure sensors 76a, 76b…Temperature sensors 77…Accelerator opening sensor

Claims

1. An air quantity sensor that detects the quantity of intake air flowing through the intake passage of an internal combustion engine as a detected air quantity, A control device that controls the internal combustion engine based on an air quantity correlation value correlated with the detected air quantity, A control system for an internal combustion engine comprising: The control device: Obtains a value equal to the ratio of the learning-time volumetric efficiency value to the confirmation-time volumetric efficiency value as a correction coefficient, Obtains a corrected air quantity obtained by multiplying the detected air quantity by the correction coefficient as the air quantity correlation value, The learning-time volumetric efficiency value: Is a volumetric efficiency value that is correlated with the ratio of the actual air quantity, which is the quantity of intake air actually flowing through the intake passage, to the rotational speed of the internal combustion engine and is not correlated with the temperature and pressure of the intake air flowing into the cylinders of the internal combustion engine, and is obtained by applying the detected air quantity in place of the actual air quantity when a predetermined learning condition is satisfied, The confirmation-time volumetric efficiency value: Is a value obtained by applying the detected air quantity in place of the actual air quantity to the volumetric efficiency value when a predetermined confirmation condition is satisfied after the learning-time volumetric efficiency value has been obtained, A control system for an internal combustion engine.

2. The control system for an internal combustion engine according to Claim 1, The control device: Determines that the learning condition is satisfied when a predetermined steady condition is satisfied, including that there is a high possibility that the variation in the actual air quantity is small when the difference between the actual air quantity and the detected air quantity is likely to be small, After the learning-time volumetric efficiency value has been obtained, determines that the confirmation condition is satisfied when the steady condition is satisfied, A control system for an internal combustion engine.

3. The control system for an internal combustion engine according to Claim 1, The control device: Obtains the correction coefficient for each of a plurality of operating states related to the internal combustion engine, Obtains the air quantity correlation value by multiplying the detected air quantity by the correction coefficient selected according to the operating state from among the plurality of obtained correction coefficients, A control system for an internal combustion engine.

Images