JP2022117558A - state estimator - Google Patents

Abstract

An object of the present invention is to more appropriately estimate related values relating to the amount of air passing through a throttle, the amount of air flowing into a cylinder, and the volume of a downstream portion. A storage device stores a first downstream pressure, which is a pressure downstream of an intake pipe throttle valve detected by a pressure sensor, and an intake pipe throttle valve estimated using a throttle passing air amount. An input variable including a second downstream pressure, which is the pressure on the downstream side, and a pressure difference between the first downstream pressure and the second downstream pressure, is input, and a first related value related to the amount of air passing through the throttle, cylinder A mapping is stored whose output is at least one of a second related value related to the inflow air amount and a third related value related to the downstream volume. The execution unit obtains input variables and applies the obtained input variables to the mapping to estimate at least one of a first association value, a second association value, and a third association value. [Selection drawing] Fig. 15

Description

The present invention relates to a state estimation device.

Conventionally, as this type of technology, during steady engine operation, the relationship between the throttle opening and the effective opening area is calculated from a volumetric efficiency correction coefficient map that is adapted to the valve timing during steady operation. In the period from the transient change until the exhaust manifold temperature converges, the learned relationship is used to calculate the amount of throttle air passing through the throttle valve and to model the response delay of the intake system. There has been proposed a method of calculating an in-cylinder air amount using a physical model (see Patent Document 1, for example).

JP 2014-84817 A

In the above technology, when calculating the amount of air flowing into the cylinder, manufacturing variations and aging of the parts downstream of the throttle valve of the engine intake pipe are not fully considered. There is room to do better. Taking this into consideration, it is required to more appropriately estimate related values related to the amount of air passing through the throttle, the amount of air flowing into the cylinder, and the volume of the downstream portion of the intake pipe, which is the volume of the portion downstream of the throttle valve. It is

A primary object of the state estimating device of the present invention is to more appropriately estimate related values related to the amount of air passing through the throttle, the amount of air flowing into the cylinder, and the volume of the downstream portion.

The state estimating device of the present invention employs the following means to achieve the above-mentioned main purpose.

The state estimation device of the present invention is
A first related value related to a throttle passing air amount passing through a throttle valve arranged in an intake pipe of an engine, a second related value related to a cylinder inflow air amount flowing into a combustion chamber of the engine, and the intake pipe A state estimating device that estimates at least one of a third related value related to a downstream volume, which is the volume of a portion downstream of the throttle valve,
having a storage device and an execution device,
The storage device stores a first downstream pressure, which is a pressure downstream of the throttle valve in the intake pipe detected by a pressure sensor, and the throttle valve in the intake pipe estimated using the amount of air passing through the throttle. An input variable including a second downstream pressure, which is a pressure on the downstream side, and a pressure difference between the first downstream pressure and the second downstream pressure, and the first related value, the second storing a mapping whose output is at least one of the associated value and the third associated value;
The executing device obtains the input variables and applies the obtained input variables to the mapping to estimate at least one of the first, second and third association values. ,
This is the gist of it.

The state estimation device of the present invention comprises a storage device and an execution device. The storage device stores the first downstream pressure, which is the pressure downstream of the throttle valve in the intake pipe detected by the pressure sensor, and the pressure downstream of the throttle valve in the intake pipe estimated using the amount of air passing through the throttle. and the pressure difference between the first downstream pressure and the second downstream pressure, and the first related value related to the throttle passing air amount, the cylinder inflow air amount A mapping is stored whose output is at least one of a second relevant value associated with , and a third relevant value associated with the downstream volume. The execution unit obtains input variables and applies the obtained input variables to the mapping to estimate at least one of a first association value, a second association value, and a third association value. The inventors have found that the relationship between the first downstream pressure, the second downstream pressure, and the pressure error for the same first related value, second related value, and third related value is greater than that of the throttle valve of the intake pipe. It was found that it differs due to variations in manufacturing of downstream parts and changes over time. Therefore, applying the input variables including the first downstream pressure, the second downstream pressure, and the pressure error to the mapping to estimate at least one of the first, second, and third associated values. At least one of the first, second and third relevant values can be estimated more appropriately. Here, the mapping may be determined by machine learning, or may be determined by human experimentation, analysis, or the like.

In the state estimation device of the present invention, the input variable may further include the amount of change per unit time of the first downstream pressure. The input variable is at least one of the number of revolutions of the engine, the degree of opening of the throttle valve, an advance amount of opening/closing timing of an intake valve of the engine, and a retard amount of opening/closing timing of an exhaust valve of the engine. It may further include one. Further, the input variables further include the opening degree of the exhaust gas recirculation valve in an exhaust gas recirculation device having an exhaust gas recirculation valve provided in a connecting pipe connecting the exhaust pipe and the intake pipe of the engine. good too. Additionally, the input variables may further include the boost pressure of a supercharger having a compressor located upstream of the throttle valve in the intake manifold. By doing so, at least one of the first related value, the second related value, and the third related value can be more sufficiently and appropriately estimated.

In the state estimation device of the present invention, the first related value may be a first correction value used for correcting the throttle passing air amount. In this case, the throttle passage air amount may be corrected using the first correction value when the operating state of the engine is in a transient state.

In the state estimation device of the present invention, the second related value may be a second correction value used for correcting the in-cylinder air amount. In this case, the cylinder inflow air amount may be corrected using the second correction value when the operating state of the engine is in a transient state.

In the state estimation device of the present invention, the third related value may be a third correction value used for setting (correcting) the volume of a portion of the intake pipe downstream of the throttle valve. In this case, when the operating state of the engine is in a steady state, the third correction value may be used to correct the volume of the portion of the intake pipe downstream of the throttle valve.

1 is a configuration diagram showing a schematic configuration of an engine system 10 including a state estimation device as one embodiment of the present invention; FIG. 4 is an explanatory diagram showing an example of signals input to and output from an electronic control unit 70 of the engine device 10; FIG. It is an explanatory view showing an example of an air model. It is an explanatory view showing an example of a flow coefficient estimation map. It is explanatory drawing which shows an example of the map for opening cross-sectional area estimation. FIG. 4 is an explanatory diagram showing a function Φ(Pm/Pa); FIG. 4 is an explanatory diagram of a throttle model M10; FIG. FIG. 4 is an explanatory diagram of an intake pipe model M20; FIG. 4 is an explanatory diagram of an intake valve model M30; FIG. 4 is an explanatory diagram of an in-cylinder air amount mci and an in-cylinder charged air amount mcf; 4 is a flowchart showing an example of correction processing S10. It is a flow chart which shows an example of amendment processing S20. It is a flow chart which shows an example of amendment processing S30. 9 is a flowchart showing an example of correction value setting processing; FIG. 5 is an explanatory diagram showing an example of correction value mapping; It is a block diagram which shows the outline of a structure of the engine apparatus 10B of a modification. 4 is an explanatory diagram showing an example of signals input to and output from an electronic control unit 70 of an engine device 10B; FIG.

Next, a mode for carrying out the present invention will be described using examples.

FIG. 1 is a block diagram showing an outline of the configuration of an engine system 10 equipped with a state estimation device as one embodiment of the present invention, and FIG. It is an explanatory view showing an example. The engine device 10 of the embodiment includes an engine 12 and an electronic control unit 70, as shown in FIGS. The engine device 10 is mounted on a general vehicle that runs using the power from the engine 12, various hybrid vehicles that include a motor in addition to the engine 12, and the like. The electronic control unit 70 corresponds to the state estimation device of the embodiment.

The engine 12 is configured as a four-cylinder internal combustion engine that uses fuel such as gasoline or light oil to output power through four strokes of intake, compression, expansion, and exhaust. The engine 12 sucks air cleaned by an air cleaner 22 into an intake pipe 23 and passes it through a throttle valve 26 and a surge tank 27 in that order. to mix air and fuel. Hereinafter, the portion of the intake pipe 23 on the upstream side of the throttle valve 26 will be referred to as "throttle upstream portion 23u", and the portion of the intake pipe 23 on the downstream side of the throttle valve 26 will be referred to as "throttle downstream portion 23d". Then, this air-fuel mixture is sucked into the combustion chamber 30 through the intake valve 29 and is explosively burned by an electric spark generated by the spark plug 31. The reciprocating motion of the piston 32 pushed down by the energy of the explosive combustion causes the rotary motion of the crankshaft 14. Convert to Exhaust gas discharged from the combustion chamber 30 to an exhaust pipe 35 through an exhaust valve 33 is discharged to the outside air through purifiers 37 and 38 and passes through an exhaust gas recirculation device (hereinafter referred to as an "EGR (Exhaust Gas Recirculation) device"). ) 40 to the surge tank 27 of the intake pipe 23 (circulated).

Purifiers 37 and 38 respectively have catalysts (three-way catalysts) 37a and 38a that purify harmful components such as carbon monoxide (CO), hydrocarbons (HC), and nitrogen oxides (NOx). The EGR device 40 has an EGR pipe 42 and an EGR valve 44 . The EGR pipe 42 communicates between the purification device 37 and the purification device 38 of the exhaust pipe 35 and the surge tank 27 of the intake pipe 23 . The EGR valve 44 is provided on the EGR pipe 42 and controlled by the electronic control unit 70 . In this EGR device 40 , the amount of exhaust gas recirculated through the exhaust pipe 35 is adjusted by adjusting the opening of the EGR valve 44 , and the exhaust gas is recirculated to the intake pipe 23 . Engine 12 can draw a mixture of air, exhaust, and fuel into combustion chamber 30 in this manner.

The engine 12 also includes a variable valve timing device 34 . The variable valve timing device 34 is configured such that the opening/closing timings of the intake valve 29 and the exhaust valve 33 can be changed continuously.

The electronic control unit 70 includes a microcomputer having a CPU 71, a ROM 72, a RAM 73, a flash memory 74, and an input/output port, as shown in FIG. Signals from various sensors are input to the electronic control unit 70 through input ports. Signals input to the electronic control unit 70 include, for example, the crank angle θcr from the crank position sensor 14a that detects the rotational position of the crankshaft 14 of the engine 12, and the water temperature sensor 15 that detects the temperature of the cooling water of the engine 12. The cooling water temperature Tw from the throttle valve 26 and the throttle opening θt from the throttle position sensor 26a that detects the position (opening) of the throttle valve 26 can be mentioned. The cam angles θci and θco from the cam position sensor 16 that detects the rotational position of the intake camshaft that opens and closes the intake valve 29 and the rotational position of the exhaust camshaft that opens and closes the exhaust valve 33 can also be used. An intake air amount Qa from an air flow meter 23a installed upstream of the throttle valve 26 of the intake pipe 23, and an intake air temperature Ta from an intake temperature sensor 23t installed upstream of the throttle valve 26 of the intake pipe 23. , an intake pressure Pa from an intake pressure sensor 23b mounted on the upstream side of the throttle valve 26 in the intake pipe 23. A first downstream pressure Ps, which is the pressure of the gas in the throttle downstream portion 23d of the intake pipe 23, from the pressure sensor 27a attached to the surge tank 27, and a temperature sensor 27t attached to the surge tank 27, from which the temperature sensor 27t of the intake pipe 23 is detected. A downstream portion temperature Ts, which is the temperature of the air in the throttle downstream portion 23d, can also be mentioned. A front air-fuel ratio AF1 from a front air-fuel ratio sensor 39a installed upstream of the purification device 37 of the exhaust pipe 35, and a rear air-fuel ratio sensor installed between the purification device 37 and the purification device 38 of the exhaust pipe 35. Rear air/fuel ratio AF2 from 39b can also be mentioned. The opening degree θegr of the EGR valve 44 from the opening sensor 45 that detects the opening degree of the EGR valve 44 can also be mentioned.

Various control signals are output from the electronic control unit 70 through the output port. The signals output from the electronic control unit 70 include, for example, a control signal to the throttle valve 26 of the engine 12, a control signal to the fuel injection valve 28, a control signal to the spark plug 31, and a control signal to the variable valve timing device 34. Control signals can be mentioned. A control signal to the EGR valve 44 may also be mentioned.

The electronic control unit 70 calculates the rotational speed Ne of the engine 12 based on the crank angle θcr of the crankshaft 14 from the crank position sensor 14a. In addition, the electronic control unit 70 determines the load factor (actually inhaled in one cycle with respect to the stroke volume per cycle of the engine 12) based on the intake air amount Qa from the air flow meter 23a and the rotation speed Ne of the engine 12. Air volume ratio) KL is calculated. Further, the electronic control unit 70 detects the cam angles θci and θco of the intake camshaft and the exhaust camshaft from the cam position sensor 16 with respect to the crank angle θcr of the crankshaft 14 from the crank position sensor 14a (θci−θcr), ( The opening/closing timings VTi and VTo of the intake valve 29 and the exhaust valve 33 are calculated based on .theta.co-.theta.cr).

In the engine apparatus 10 of the embodiment thus configured, the CPU 71 of the electronic control unit 70 controls the operation of the engine 12 (intake air amount control, fuel injection control, ignition control, variable valve timing control) and EGR control. In intake air amount control, the opening degree of the throttle valve 26 is controlled. In the fuel injection amount control, the fuel injection amount from the fuel injection valve 28 is controlled. In ignition control, the ignition timing of the spark plug 31 is controlled. In the variable valve timing control, the variable valve timing device 34 controls the opening/closing timings of the intake valve 29 and the exhaust valve 33 . The EGR control controls the opening of the EGR valve 44 .

Further, in the engine system 10 of the embodiment, the CPU 71 of the electronic control unit 70 uses the air model of FIG. , the downstream volume Vm, and the in-cylinder charged air amount mcf are calculated (estimated). At least some of the calculated throttle passing air amount mt, second downstream pressure Pm, downstream temperature Tm, cylinder inflow air amount mci, downstream volume Vm, and cylinder charged air amount mcf are used by the engine 12 It is used for intake air amount control and fuel injection control. Here, the throttle passing air amount mt is the flow rate of air passing through the throttle valve 26 per unit time. The second downstream pressure Pm is the gas pressure in the throttle downstream portion 23d of the intake pipe 23 corresponding to the current throttle opening θt. The downstream portion temperature Tm is the temperature of the air in the throttle downstream portion 23d. The cylinder inflow air amount mci is the flow rate of air flowing into the combustion chamber 30 per unit time (value obtained by averaging the flow rate of air flowing into the combustion chambers 30 of all cylinders from the downstream portion 23d of the throttle). The downstream portion volume Vm is the volume of the throttle downstream portion 23d. The in-cylinder charged air amount mcf is the amount of air charged in the combustion chamber 30 when the intake valve 29 is closed.

The air model of FIG. 3 has a throttle model M10, an intake pipe model M20, an intake valve model M30, correction processing S10, correction processing S20, and correction processing S30. The throttle opening θt, the intake air temperature Ta, the intake pressure Pa, and the second downstream pressure Pm are input to the throttle model M10. A value detected by the throttle position sensor 26a is input as the throttle opening θt. A value detected by the intake air temperature sensor 23t is input as the intake air temperature Ta. A value detected by the intake pressure sensor 23b is input as the intake pressure Pa. A value calculated by the intake pipe model M20 is input as the second downstream pressure Pm. The throttle model M10 substitutes the throttle opening θt, the intake air temperature Ta, the intake pressure Pa, and the second downstream pressure Pm into the model formula of the throttle model M10 to calculate the throttle passing air amount mt, and calculates the calculated throttle passing air amount. mt is output to the correction processing S10. In the correction process S10, the throttle passing air amount mt calculated by the throttle model M10 is corrected as necessary and output to the intake pipe model M20.

The intake air temperature Ta, the amount of air passing through the throttle mt, the amount of air flowing into the cylinder mci, and the downstream volume Vm are input to the intake pipe model M20. A value detected by the intake air temperature sensor 23t is input as the intake air temperature Ta. The throttle passing air amount mt is input as a value calculated by the throttle model M10 and corrected as necessary by the correction processing S10. The cylinder inflow air amount mci is input with a value that is calculated by the intake valve model M30 and corrected as necessary by the correction processing S20. As for the downstream volume Vm, the value set by the correction process S30 based on the downstream volume base value Vm0 is input. The downstream portion volume base value Vm0 is a value predetermined as a base value of the volume of the throttle downstream portion 23d (for example, a value at the time of completion of manufacture of the engine device 10). The intake pipe model M20 substitutes the intake air temperature Ta, the amount of air passing through the throttle mt, the amount of air flowing into the cylinder mci, and the downstream volume Vm into the model formula of the intake pipe model M20 to obtain the second downstream pressure Pm and the downstream temperature Tm. is calculated, and the calculated second downstream pressure Pm is output to the throttle model M10 and the intake valve model M30, and the calculated downstream temperature Tm is output to the intake valve model M30.

The intake valve model M30 receives the intake air temperature Ta, the rotation speed Ne, the opening/closing timings VTi and VTo of the intake valve 29 and the exhaust valve 33, the second downstream pressure Pm, and the downstream temperature Tm. A value detected by the intake air temperature sensor 23t is input as the intake air temperature Ta. A value calculated based on the crank angle θcr of the crankshaft 14 from the crank position sensor 14a is input as the rotational speed Ne. The opening/closing timings VTi and VTo of the intake valve 29 and the exhaust valve 33 are determined by the cam angles θci and θco of the intake camshaft and the exhaust camshaft from the cam position sensor 16 with respect to the crank angle θcr of the crankshaft 14 from the crank position sensor 14a. Values calculated based on (θci-θcr) and (θco-θcr) are input. The values calculated by the intake pipe model M20 are input as the second downstream pressure Pm and the downstream temperature Tm. The intake valve model M30 substitutes the intake air temperature Ta, the rotation speed Ne, the opening/closing timings VTi and VTo of the intake valve 29 and the exhaust valve 33, the second downstream pressure Pm, and the downstream temperature Tm into the model formula of the intake valve model M30. to calculate the in-cylinder air amount mci, and output the calculated in-cylinder air amount mci to the correction process S20. The correction processing S20 corrects the in-cylinder air amount mci calculated by the intake valve model M30 as necessary, converts it to the cylinder charged air amount mcf, and outputs it to the intake pipe model M20. Output.

Next, details of the throttle model M10, the intake pipe model M20, the intake valve model M30, and the correction processes S10, S20, and S30 will be described in order. First, details of the throttle model M10 will be described. The throttle model M10 calculates the throttle passing air amount mt using the throttle opening θt, the intake air temperature Ta, the intake pressure Pa, and the second downstream pressure Pm, as shown in equation (1).

Here, in equation (1), "μ(θt)" is the flow rate coefficient at the throttle valve 26 . In the embodiment, the flow coefficient μ(θt) is estimated by applying the throttle opening θt to the flow coefficient estimation map. Here, the flow coefficient estimating map is determined in advance by experiments and analyzes as the relationship between the throttle opening θt and the flow coefficient μ(θt), and is stored in the flash memory 74 . FIG. 4 is an explanatory diagram showing an example of a flow coefficient estimation map. As shown in the figure, the flow rate coefficient μ(θt) is set to decrease as the throttle opening θt increases.

In formula (1), “A(θt)” is the opening cross-sectional area of the throttle valve 26 . In the embodiment, the opening cross-sectional area A(θt) is estimated by applying the throttle opening θt to the opening cross-sectional area estimation map. Here, the map for estimating the opening cross-sectional area is determined in advance by experiments and analysis as the relationship between the throttle opening θt and the opening cross-sectional area A(θt), and is stored in the flash memory 74 . FIG. 5 is an explanatory diagram showing an example of an aperture cross-sectional area estimation map. As shown in the figure, when the throttle opening θt is less than the predetermined value θt1, the opening cross-sectional area A(θt) increases toward the predetermined value A1 as the throttle opening θt increases. In the region of θt1 or more, it is set to be constant at a predetermined value A1. Note that a predetermined relationship between the throttle opening θt and the value μ(θt)·A(θt) as the product of the flow coefficient μ(θt) and the opening cross-sectional area A(θt) depends on the throttle opening θt may be applied to estimate the value μ(θt)·A(θt).

In formula (1), "R" is a constant related to the gas constant. This constant R corresponds to a value obtained by dividing the gas constant by the mass of gas (air) per 1 mol. In formula (1), "Φ(Pm/Pa)" is a function obtained by formulas (2) and (3). In equations (2) and (3), "κ" is the ratio of specific heats. This specific heat ratio was set to a constant value. This function Φ(Pm/Pa) can be represented as a map as shown in FIG. Therefore, instead of formulas (2) and (3), the second downstream pressure Pm and the intake pressure Pa are applied to the map of FIG. 6 to obtain the value of the function Φ(Pm/Pa). good too.

FIG. 7 is an explanatory diagram of the throttle model M10. The above formulas (1) to (3) are obtained as follows. First, let the gas pressure in the throttle upstream portion 23u be the intake pressure Pa, the gas temperature in the throttle upstream portion 23u be the intake air temperature Ta, and the gas pressure in the throttle downstream portion 23d be the second downstream portion pressure Pm. Then, the law of conservation of mass, the law of conservation of energy, and the law of conservation of momentum are applied to the throttle model M10 of FIG. 7, and the equation of state of gas, the equation of the ratio of specific heats, and the Meyer relational expression are used. Equations (1) to (3) are thus obtained.

Next, details of the intake pipe model M20 will be described. The intake pipe model M20, as shown in equations (4) and (5), calculates the intake air temperature Ta, the throttle passing air amount mt, the cylinder inflow air amount mci, the downstream volume Vm, the constant R, and the specific heat ratio κ. are used to calculate the second downstream pressure Pm and the downstream temperature Tm.

FIG. 8 is an explanatory diagram of the intake pipe model M20. As can be seen from FIG. 8, when the total amount of gas (total air amount) in the downstream portion 23d of the throttle is M, the amount of change in the total amount of gas M over time is the flow rate of the gas flowing into the downstream portion 23d of the throttle, that is, the amount of air passing through the throttle. It is equal to the difference between mt and the flow rate of gas flowing out of the throttle downstream portion 23d, that is, the cylinder inflow air amount mci. Therefore, the formula (6) is obtained by the law of conservation of mass. Equation (4) is obtained from Equation (6) and the state equation (Pm·Vm=M·R·Tm) of the gas in the throttle downstream portion 23d.

Also, the amount of change over time of the energy M·Cv·Tm of the gas in the throttle downstream portion 23d is equal to the difference between the energy of the gas flowing into the throttle downstream portion 23d and the energy of the gas flowing out from the throttle downstream portion 23d. Therefore, if the temperature of the gas flowing into the throttle downstream portion 23d is taken as the intake air temperature Ta, and the temperature of the gas flowing out of the throttle downstream portion 23d is taken as the downstream portion temperature Tm, Equation (7) is obtained according to the law of conservation of energy. where, in equation (7), "Cp" is the specific heat at constant pressure of air, and "Cv" is the specific heat at constant volume of air. (5) is obtained.

Next, the details of the intake valve model M30 will be described. The intake valve model M30 calculates the cylinder inflow air amount mci using the intake air temperature Ta, the second downstream pressure Pm, and the downstream temperature Tm, as shown in equation (8). Here, "a" and "b" in equation (8) are determined based on the rotational speed Ne of the engine 12 and the opening/closing timings VTi and VTo of the intake valve 29 and the exhaust valve 33, respectively.

FIG. 9 is an explanatory diagram of the intake valve model M30. In general, the in-cylinder charged air amount mcf is determined when the intake valve 29 is closed, and is proportional to the pressure in the combustion chamber 30 at that time. Further, the pressure in the combustion chamber 30 when the intake valve 29 is closed can be regarded as equal to the pressure of the gas on the upstream side of the intake valve 29, specifically, the second downstream pressure Pm. Therefore, it can be approximated that the in-cylinder charged air amount mcf is proportional to the second downstream pressure Pm.

Here, the flow rate of air flowing into the combustion chambers 30 of all cylinders from the downstream portion 23d of the throttle per predetermined time (for example, 720° of the crank angle θcr) is divided by the predetermined time (averaged). Assuming that the in-cylinder air amount is mci, since the in-cylinder charged air amount mcf is proportional to the second downstream pressure Pm as described above, it can be considered that the in-cylinder air amount mci is also proportional to the second downstream pressure Pm. be done. From this, the above equation (8) is obtained based on theory and empirical rules. In equation (8), “a” is a proportional coefficient, and “b” is a fit value representing the burned gas remaining in the combustion chamber 30 . This adaptation value is obtained by dividing the amount of burned gas remaining in the combustion chamber 30 when the exhaust valve 33 is closed by the time ΔT180° required for the crankshaft 14 to rotate 180°. Here, 180° means an angle obtained by dividing the angle 720° at which the crankshaft 14 rotates in one cycle (four strokes of intake, compression, expansion, and exhaust) by the number of cylinders of the engine 12 (4). Further, in the actual operation of the engine 12, the downstream temperature Tm may change greatly. Therefore, in equation (8), "a·Pm−b" is multiplied by "Ta/Tm" derived based on theory and empirical rules as a correction that takes into account changes in the downstream temperature Tm.

FIG. 10 is an explanatory diagram of the cylinder inflow air amount mci and the cylinder charged air amount mcf. In FIG. 10, the horizontal axis is the crank angle θcr of the crankshaft 14, and the vertical axis is the flow rate of air actually flowing into the combustion chamber 30 of each cylinder from the downstream portion 23d of the throttle per unit time. In this embodiment, since the four-cylinder engine 12 is used, the intake valves 29 are opened in order of, for example, the first cylinder, the third cylinder, the fourth cylinder, and the second cylinder. Then, as shown in the figure, air flows into the combustion chamber 30 of each cylinder from the throttle downstream portion 23d according to the valve opening amount of the intake valve 29 corresponding to each cylinder. For example, the flow rate of air flowing into the combustion chamber 30 of each cylinder from the throttle downstream portion 23d is as indicated by the dashed line in FIG. Integrating this displacement, the flow rate of air flowing into the combustion chambers 30 of all cylinders from the downstream portion 23d of the throttle is as indicated by the solid line in FIG. Further, the in-cylinder charged air amount mcf of the first cylinder is as indicated by hatching in FIG.

On the other hand, the in-cylinder inflow air amount mci is obtained by averaging the flow rate of air flowing into the combustion chambers 30 of all cylinders from the downstream portion 23d of the throttle, and is indicated by the one-dot chain line in FIG. The in-cylinder charged air amount mcf is obtained by multiplying the in-cylinder air amount mci by the time ΔT180°. Therefore, by multiplying the value obtained by correcting the in-cylinder inflow air amount mci calculated by the intake valve model M30 as necessary in the correction process S20 by the time ΔT180°, the in-cylinder charged air amount mcf is calculated. can do. More specifically, considering that the cylinder charged air amount mcf is proportional to the pressure when the intake valve 29 is closed, the cylinder inflow air amount mci when the intake valve 29 is closed is multiplied by the time ΔT180°. The result is the in-cylinder charged air amount mcf.

Next, the correction processing S10 will be described. FIG. 11 is a flow chart showing an example of the correction processing S10. In the correction process S10, the CPU 71 of the electronic control unit 70 first inputs data such as the throttle passing air amount mt, the correction value mtad, and the steady state flag Fs (step S100). Here, a value calculated by the throttle model M10 is input as the throttle passing air amount mt. The correction value mtad is a value used for correction of the throttle passing air amount mt (correction for increasing when positive value, correction for decreasing when negative value), and is set by a correction value setting process to be described later. entered. A value set by a steady state flag setting routine (not shown) executed by the CPU 71 is input to the steady state flag Fs. In the steady state flag setting routine, the CPU 71 adjusts the engine speed based on at least a part of the rotation speed Ne of the engine 12, the load factor KL, the intake air amount Qa, the first downstream pressure Ps, the second downstream pressure Pm, and the like. 12 is determined to be in a steady state or a transient state, and when it is determined that the operating state of the engine 12 is in a steady state, the value 1 is set to the steady state flag Fs, and the operating state of the engine 12 is in a transient state, the value 0 is set to the steady state flag Fs.

When the data are input in this manner, the value of the input steady state flag Fs is checked (step S110). When the steady state flag Fs is 1, that is, when the operating state of the engine 12 is in a transient state, the value of the correction value mtad is checked (step S120). When the correction value mtad is 0, there is no need to correct the throttle passing air amount mt, so the correction process S10 is terminated.

When the correction value mtad is a positive value in step S120, a value obtained by adding the product of the correction value mtad and the coefficient kmt to the throttle passing air amount mt is added to the new throttle passing air amount mt as shown in equation (9). , the throttle passing air amount mt is corrected (step S140), and this routine ends. Here, the coefficient kmt gradually increases from a value of 0 to a value of 1 and is held at a value of 1. When the correction value mtad is a positive value (when the throttle passing air amount mt after correction becomes larger than the throttle passing air amount mt before correction), fuel is injected from the fuel injection valve 28 into the intake pipe 23. The fuel (deposit) adhering to the wall surface is removed by cleaning the intake pipe 23 at a dealer or the like, and the actual amount of air passing through the throttle is greater than the amount of air passing through the throttle mt calculated by the throttle model M10. be able to. In the embodiment, taking this into consideration, the throttle passing air amount mt is corrected by the processing of step S140. As a result, the throttle passing air amount mt can be set to a more appropriate value. Further, by gradually increasing the coefficient kmt from the value 0 to the value 1, the extent to which the correction value mtad is reflected in the throttle passing air amount mt is gradually increased. can be suppressed. Note that the coefficient kmt may gradually increase from the value 0 to the value kmt1 (0<kmt1<1) and be held at the value kmt1. Also, the coefficient kmt may be a constant value kmt2 (0<kmt2≦1).

mt←mt＋mtad・kmt (9)

When the correction value mtad is a negative value in step S120, it is determined whether or not the correction value mtad continues to be a negative value from the previous trip (step S130). When it is determined that the correction value mtad has continued to be a negative value since the previous trip, the throttle passing air amount mt is corrected according to the above equation (9) (step S140), and this routine ends. When it is determined that the correction value mtad has not continued to be a negative value since the previous trip (the value was 0 or more in the previous trip), the correction process S10 is terminated without correcting the throttle passing air amount mt. . The time when the correction value mtad is a negative value can include the time when the above-described deposit amount is gradually increasing. Since the amount of deposit accumulation does not increase rapidly but gradually increases, in order to suppress erroneous correction of the throttle passing air amount mt, in the embodiment, the correction value mtad continues to be a negative value from the previous trip. In some cases, the correction value mtad is used to correct the throttle passing air amount mt.

When the steady state flag Fs is 0 in step S110, that is, when the operating state of the engine 12 is in a steady state, the correction process S10 ends without correcting the throttle passing air amount mt. If the throttle passing air amount mt is corrected while the engine 12 is in a steady state, the throttle passing air amount mt changes even though the engine 12 is in a steady state. Therefore, in the embodiment, when the engine 12 is in the steady state, the throttle passing air amount mt is not corrected.

Next, the correction processing S20 will be described. FIG. 12 is a flow chart showing an example of the correction processing S20. In the correction process S20, 71 of the electronic control unit 70 first inputs data such as the cylinder inflow air amount mci, the correction value mciad, and the steady state flag Fs (step S200). Here, a value calculated by the intake valve model M30 is input as the cylinder inflow air amount mci. The correction value mciad is a value used for correcting the in-cylinder air amount mci (correction to increase when positive value, correction to decrease when negative value), and is determined by the correction value setting process described later. The set value is entered. The steady state flag Fs is input in the same manner as in the process of step S100 of the correction process S10.

When the data are input in this way, the value of the input steady state flag Fs is checked (step S210). When the steady state flag Fs is 1, that is, when the operating state of the engine 12 is in a transient state, the value of the correction value mciad is checked (step S220). When the correction value mciad is 0, there is no need to correct the in-cylinder air amount mci, so the correction process S20 ends.

When the correction value mciad is a positive value in step S220, the value obtained by adding the product of the correction value mciad and the coefficient kmci to the cylinder inflow air amount mci is added to the new cylinder inflow air amount mci as shown in equation (10). By setting the amount mci, the cylinder inflow air amount mci is corrected (step S240), and the routine ends. Here, the coefficient kmci gradually increases from a value of 0 to a value of 1 and is held at a value of 1. When the correction value mciad is a positive value (when the cylinder inflow air amount mci after correction is greater than the cylinder inflow air amount mci before correction), the above-mentioned deposits are stored in the intake pipe 23 at a dealer or the like. is removed by cleaning the intake valve model M30, etc., and the actual amount of air flowing into the cylinder is greater than the amount of air flowing into the cylinder mci calculated by the intake valve model M30. In the embodiment, in consideration of this, the cylinder inflow air amount mci is corrected by the processing of step S240. As a result, the cylinder inflow air amount mci can be set to a more appropriate value. Further, by gradually increasing the coefficient kmci from 0 to 1, the extent to which the correction value mciad is reflected in the cylinder inflow air amount mci is gradually increased. abrupt change in Note that the coefficient kmci may gradually increase from the value 0 to the value kmci1 (0<kmci1<1) and be held at the value kmci1. Also, the coefficient kmci may be a constant value kmci2 (0<kmci2≦1).

mci←mci＋mciad・kmci (10)

When the correction value mciad is a negative value in step S220, it is determined whether or not the correction value mciad continues to be a negative value from the previous trip (step S230). When it is determined that the correction value mciad continues to be a negative value from the previous trip, the cylinder inflow air amount mci is corrected according to the above equation (10) (step S240), and this routine ends. When it is determined that the correction value mciad has not continued to be a negative value since the previous trip (the value was 0 or more in the previous trip), the correction process S20 is terminated without correcting the cylinder inflow air amount mci. do. A case where the correction value mciad is a negative value can be a case where the amount of deposits is gradually increasing. Since the accumulated amount of deposits does not increase rapidly but gradually increases, in order to suppress erroneous correction of the in-cylinder air amount mci, in the embodiment, the correction value mciad continues to be a negative value from the previous trip. In this case, the correction value mciad is used to correct the in-cylinder air amount mci.

When the steady state flag Fs is 0 in step S210, that is, when the operating state of the engine 12 is in the steady state, the correction process S10 is terminated without correcting the throttle passing air amount mt. If the in-cylinder air amount mci is corrected while the engine 12 is in a steady state, the in-cylinder air amount mci changes even though the engine 12 is in a steady state. Therefore, in the embodiment, when the engine 12 is in the steady state, the cylinder inflow air amount mci is not corrected.

Next, the correction processing S30 will be described. FIG. 13 is a flow chart showing an example of the correction processing S30. In the correction process S30, the CPU 71 of the electronic control unit 70 first inputs data such as the correction value Vmad and the steady state flag Fs (step S300). Here, the correction value Vmad is a value used for correcting the downstream portion volume base value Vm0, and a value set by a correction value setting process described later is input. The steady state flag Fs is input in the same manner as in the process of step S100 of the correction process S10.

When the data is input in this way, the value of the input steady state flag Fs is checked (step S310). When the steady state flag Fs is 1, that is, when the operating state of the engine 12 is steady operation, the product of the correction value Vmad and the coefficient kvm is obtained from the downstream volume base value Vm0 as shown in equation (11). is set as the downstream volume Vm (step S320), and the routine ends. Here, the coefficient kvm is a constant value kvm1 (0<kvm1<1). It is assumed that the coefficient kvm depends on the amount of deposits accumulated as described above. In the embodiment, considering this, the downstream volume Vm is set by the process of step S320. Thereby, the downstream volume Vm can be set more appropriately. Further, by using the coefficient kvm smaller than the value 1, it is possible to suppress the sudden change of the downstream portion volume Vm with respect to the downstream portion volume base value Vm0. Note that the coefficient kvm may be set to a value of one. Also, the coefficient kvm may gradually increase from the value 0 to the value kvm1 and be held at the value kvm. This makes it possible to further suppress the sudden change of the downstream volume Vm with respect to the downstream volume base value Vm0.

Vm←Vm0－Vmad・kvm (11)

When the steady state flag Fs is 0 in step S310, that is, when the operating state of the engine 12 is in a transient state, the downstream volume base value Vm0 is set to the downstream volume Vm (step S330), and this routine exit. It is assumed that the downstream volume Vm can affect intake air amount control, fuel injection control, and the like when the operating state of the engine 12 is in a transient state. Therefore, when the operating state of the engine 12 is in a transient state, the downstream volume base value Vm0 is set to the downstream volume Vm.

Next, processing for setting the correction value mtad used in the first correction process in FIG. 11, the correction value mciad used in the second correction process in FIG. 12, and the correction value Vmad used in the third correction process in FIG. will be described using the correction value setting process of . This processing is repeatedly executed by the CPU 71 of the electronic control unit 70 .

In the correction value setting process of FIG. Data such as the degree of opening θegr, the first downstream pressure Ps, the second downstream pressure Pm, the pressure error ΔP, and the downstream pressure change rate dPs are input (step S400).

Here, as the rotation speed Ne, a value calculated based on the crank angle θcr of the crankshaft 14 from the crank position sensor 14a is input. A value detected by the throttle position sensor 26a is input as the throttle opening θt. The opening/closing timings VTi and VTo of the intake valve 29 and the exhaust valve 33 are determined by the cam angles θci and θco of the intake camshaft and the exhaust camshaft from the cam position sensor 16 with respect to the crank angle θcr of the crankshaft 14 from the crank position sensor 14a. Values calculated based on (θci-θcr) and (θco-θcr) are input. A value detected by the opening sensor 45 is input as the opening θegr of the EGR valve 44 . A value detected by the pressure sensor 27a is input as the first downstream pressure Ps. A value obtained from the air model (see FIG. 3) is input as the second downstream pressure Pm. The first downstream pressure Ps is a detected gas pressure in the throttle downstream portion 23d, and the second downstream pressure Pm is an estimated gas pressure in the throttle downstream portion 23d. A value obtained by subtracting the second downstream pressure Pm from the first downstream pressure Ps is input as the pressure error ΔP. A value calculated as the amount of change per unit time of the first downstream pressure Ps is input as the downstream pressure change rate dPs.

When the data are input in this way, the input rotation speed Ne, throttle opening θt, opening/closing timings VTi and VTo of the intake valve 29 and the exhaust valve 33, opening θegr of the EGR valve 44, first downstream pressure Ps, second downstream The pressure Pm, the pressure error ΔP, and the downstream pressure change rate dPs are applied to the correction value mapping to set the correction values mtad, mciad, and Vmad (step S410), and the correction value setting process ends.

Here, the correction value mapping includes the rotational speed Ne, the throttle opening θt, the opening/closing timings VTi and VTo of the intake valve 29 and the exhaust valve 33, the opening θegr of the EGR valve 44, the first downstream pressure Ps, the second downstream pressure It is a mapping that takes the pressure Pm, the pressure error ΔP, and the downstream pressure change rate dPs as inputs and outputs the correction values mtad, mciad, and Vmad, and is stored in the flash memory 74 . FIG. 15 is an explanatory diagram showing an example of correction value mapping. In the example of FIG. 15, the correction value mapping is constructed by a neural network having intermediate layers (hidden layers) 1 and 2. In FIG. The input layer has nodes A[1] to A[9], hidden layer 1 has nodes B[1] to B[3], hidden layer 2 has nodes C[1] to C[ 3], and the output layer has nodes D[1] to D[3]. The variables x[1] to x[9] of the nodes A[1] to A[9] in the input layer are the rotational speed Ne, the throttle opening θt, the opening/closing timings VTi and VTo of the intake valve 29 and the exhaust valve 33, respectively. , the degree of opening θegr of the EGR valve 44, the first downstream pressure Ps, the second downstream pressure Pm, the pressure error ΔP, and the downstream pressure change rate dPs. Outputs y[1] to y[3] of nodes D[1] to D[3] in the output layer are correction values mtad, mciad and Vmad, respectively. As the activation functions h1 and h2 of the intermediate layers 1 and 2, a sigmoid function, a hyperbolic tangent function, or the like is used. A ReLU (ramp function), an identity function, or the like is used as the activation function f of the output layer. The output z1[j1] of the node B[j1] (j1=1 to 3) of the hidden layer 1 is the variable x[i] of the input layer node A[i] (i=1 to 9) and the input layer Using the coefficient w[j1, i] for defining the input value from the node A[i] to the node B[j1] of the hidden layer 1 and the activation function h1 of the hidden layer 1, can be expressed as Output z2[j2] of node C[j2] (j2=1 to 3) of hidden layer 2 is output z1[j1] of node B[j1] of hidden layer 1 and node B[j1] of hidden layer 1 (13) using the coefficient w[j2, j1] for defining the input value from to the node C[j2] of the intermediate layer 2 and the activation function h2 of the intermediate layer 2. can. The output y[j3] of the output layer node D[j3] (j3=1 to 3) is obtained from the output z2[j2] of the intermediate layer 2 node C[j2] and the intermediate layer 2 node C[j2] Using the coefficient w[j3, j2] for defining the input value to the node D[j3] of the output layer and the activation function f of the output layer, it can be expressed as Equation (14). Note that the number of intermediate layers is not limited to two. Also, the number of nodes in each intermediate layer is not limited to three.

The correction value mapping is generated, for example, by providing various sensors in a test engine device similar to the engine device 10 as follows. In this test engine device, the rotation speed Ne, the throttle opening θt, the opening and closing timings VTi and VTo of the intake valve 29 and the exhaust valve 33, the opening θegr of the EGR valve 44, the first downstream pressure Ps, the second downstream pressure The pressure Pm, the pressure error ΔP, and the downstream pressure change rate dPs are applied to the correction value mapping to calculate the training data of the correction values mtad, mciad, and Vmad, and the correction values are obtained from various sensors and various models such as the air model described above. Teacher data of mtad, mciad, and Vmad are obtained and associated to generate sample data. Then, using each sample data, the coefficients (the above-mentioned coefficients w[j1, i] and Coefficient w[j2, j1], coefficient w[j3, j2]) are updated to generate a correction value map (learned model).

For the same correction values mtad, mciad, and Vmad, the inventors found that the rotational speed Ne, the throttle opening θt, the opening/closing timing VTi of the intake valve 29, the opening/closing timing VTo of the exhaust valve 33, and the opening θegr of the EGR valve 44 , the first downstream pressure Ps, the second downstream pressure Pm, the pressure error ΔP, and the downstream pressure change rate dPs, depending on manufacturing variations and aging of the throttle downstream portion 23d (including the above-mentioned deposit accumulation amount) I found out that it is different. Therefore, the rotation speed Ne, the throttle opening θt, the opening/closing timings VTi and VTo of the intake valve 29 and the exhaust valve 33, the opening θegr of the EGR valve 44, the first downstream pressure Ps, the second downstream pressure Pm, and the pressure error ΔP , the downstream pressure change rate dPs is applied to the correction value mapping to set the correction values mtad, mciad, and Vmad. A more appropriate value can be set. By using the correction value mtad for correcting the throttle passing air amount mt, using the correction value mciad for correcting the cylinder inflow air amount mci, and using the correction value Vmad for setting the downstream volume Vm, It is possible to more appropriately estimate the throttle passing air amount mt, the cylinder inflow air amount mci, and the downstream portion volume Vm.

In the electronic control unit 70 as the state estimation device of the embodiment described above, the input variables including the first downstream pressure Ps, the second downstream pressure Pm, and the pressure error ΔP are applied to the correction value mapping configured by the neural network. Then, the correction values mtad, mciad, and Vmad are set (estimated). Thereby, the correction values mtad, mciad, and Vmad can be set more appropriately. By using the correction value mtad for correcting the throttle passing air amount mt, using the correction value mciad for correcting the cylinder inflow air amount mci, and using the correction value Vmad for setting the downstream volume Vm, It is possible to more appropriately estimate the throttle passing air amount mt, the cylinder inflow air amount mci, and the downstream portion volume Vm.

In the engine device 10 of the embodiment, the correction value mapping includes the rotational speed Ne, the throttle opening θt, the opening/closing timings VTi and VTo of the intake valve 29 and the exhaust valve 33, the opening θegr of the EGR valve 44, the first downstream pressure Ps , second downstream pressure Pm, pressure error .DELTA.P, and downstream pressure change rate dPs as inputs, and correction values mtad, mciad, and Vmad as outputs. However, the input of the correction value mapping may include the first downstream pressure Ps, the second downstream pressure Pm, and the pressure error ΔP. At least a part of the opening/closing timings VTi and VTo of the EGR valve 44, the opening degree θegr of the EGR valve 44, and the downstream pressure change rate dPs may not be included. Further, the correction value mapping may output a part of the correction values mtad, mciad, and Vmad.

In the engine device 10 of the embodiment, the correction value mapping is configured by a neural network. However, the correction value mapping may be configured by, for example, a random forest, a support vector machine, an LSTM (Long Short Term Memory), etc., other than the neural network. Also, the correction value mapping may be configured as a map or an arithmetic expression based on human experiments or analysis.

In the engine device 10 of the embodiment, in the correction process S10, when the operating state of the engine 12 is in a transient state and the correction value mtad is a negative value, the correction value mtad continues to be a negative value from the previous trip. , the throttle passing air amount mt is corrected using the correction value mtad. However, when the operating state of the engine 12 is in a transient state and the correction value mtad is a negative value, the correction value mtad may be used to correct the throttle passing air amount mt.

In the engine device 10 of the embodiment, in the correction process S10, when the operating state of the engine 12 is in a steady state, the throttle passing air amount mt is not corrected. It may be corrected.

In the engine device 10 of the embodiment, in the correction process S20, when the operating state of the engine 12 is in a transient state and the correction value mciad is a negative value, the correction value mciad continues to be a negative value from the previous trip. In this case, the correction value mciad is used to correct the in-cylinder air amount mci. However, when the operating state of the engine 12 is in a transient state and the correction value mciad is a negative value, regardless of whether the correction value mciad continues to be a negative value from the previous trip, the correction value mciad may be used to correct the in-cylinder air amount mci.

In the engine device 10 of the embodiment, in the correction process S20, when the operating state of the engine 12 is in a steady state, the cylinder inflow air amount mci is not corrected. It is also possible to correct mci.

In the engine device 10 of the embodiment, in the correction process S30, when the operating state of the engine 12 is in a transient state, the downstream volume base value Vm0 is set to the downstream volume Vm without using the correction value Vmad. , the downstream volume Vm may be set using the downstream volume base value Vm0 and the correction value Vmad.

In the engine device 10 of the embodiment, the engine 12 does not have a supercharger. However, the engine 12 may be provided with a supercharger. FIG. 16 is a configuration diagram showing an outline of the configuration of an engine device 10B of a modified example, and FIG. 17 is an explanatory diagram showing an example of signals input to and output from the electronic control unit 70 of the engine device 10B. The engine device 10B of FIG. 16 differs from the engine device 10 of FIG. 1 in that instead of the engine 12, an engine 12B having a supercharger 50 and an intercooler 25 is provided. The points of the engine device 10B that are different from the engine device 10 will be described below.

The supercharger 50 is configured as a turbocharger and includes a turbine 51 , a compressor 52 , a wastegate valve 54 and a blow-off valve 55 . The turbine 51 is arranged upstream of the purification device 37 of the exhaust pipe 35 . The compressor 52 is arranged upstream of the throttle valve 26 of the intake pipe 23 and is connected to the turbine 51 via a rotating shaft 53 . The wastegate valve 54 is provided in the bypass pipe 36 that connects the upstream side and the downstream side of the exhaust pipe 35 with respect to the turbine 51 and is controlled by the electronic control unit 70 . The blow-off valve 55 is provided in the bypass pipe 24 that connects the upstream side and the downstream side of the compressor 52 of the intake pipe 23 and is controlled by the electronic control unit 70 .

In the turbocharger 50, the opening of the wastegate valve 54 is adjusted to adjust the distribution ratio between the amount of exhaust gas flowing through the bypass pipe 36 and the amount of exhaust gas flowing through the turbine 51, thereby adjusting the rotational driving force of the turbine 51. , the amount of air compressed by the compressor 52 is adjusted, and the boost pressure (intake pressure) of the engine 12 is adjusted. When the wastegate valve 54 is fully open, the engine 12 can operate like a naturally aspirated engine without the supercharger 50 .

Further, in the turbocharger 50, when the pressure downstream of the compressor 52 in the intake pipe 23 is higher than the pressure upstream to some extent, the excess pressure downstream of the compressor 52 is released by opening the blow-off valve 55. can be released. Instead of the valve controlled by the electronic control unit 70, the blow-off valve 55 is a check valve that opens when the pressure on the downstream side of the compressor 52 in the intake pipe 23 becomes higher than the pressure on the upstream side to some extent. It may be

The intercooler 25 is arranged between the compressor 52 of the intake pipe 23 and the throttle valve 26 . The intercooler 25 exchanges heat between air compressed by the compressor 52 and cooling water of a cooling device (not shown).

As shown in FIG. 17, the signals input to the electronic control unit 70 include, in addition to the same signals as those of the engine device 10, the supercharging pressure applied between the compressor 52 of the intake pipe 23 and the intercooler 25. The supercharging pressure Pc from the sensor 23c can be mentioned. Signals output from the electronic control unit 70 include, in addition to signals similar to those of the engine device 10, a control signal to the wastegate valve 54 and a control signal to the blow-off valve 55.

In the modified engine device 10B configured in this manner, the CPU 71 of the electronic control unit 70 performs supercharging control in addition to the operation control and EGR control of the engine 12 based on the required load factor KL* of the engine 12 . In supercharging control, the opening of the waste gate valve 54 is controlled.

Further, in the engine device 10B of the modified example, the CPU 71 of the electronic control unit 70 replaces the input intake pressure Pa of the throttle model M10 with the supercharging pressure Pc in the air model of FIG. By replacing the intake air temperature Ta input to the intake valve model M30 with the temperature Tc between the intercooler 25 of the intake pipe 23 and the throttle valve 26 (hereinafter referred to as "pre-throttle temperature"), the throttle passing air amount mt and the second A downstream pressure Pm, a downstream temperature Tm, a cylinder inflow air amount mci, a downstream volume Vm, and a cylinder charged air amount mcf are calculated (estimated). A value detected by the boost pressure sensor 23c is input as the boost pressure Pc. In addition, the throttle front temperature Tc is the intake air temperature Ta detected by the intake air temperature sensor 23t, the intake air amount Qa detected by the air flow meter 23a, the intake pressure Pa detected by the intake pressure sensor 23b, the supercharging pressure Pc, and the pressure A value estimated based on at least part of the first downstream pressure Ps detected by the sensor 27a and the downstream temperature Ts detected by the temperature sensor 27t is input.

In the engine devices 10 and 10B of the embodiments and modifications, the engine 12 is configured as a 4-cylinder engine, but may be configured as a 6-cylinder engine, an 8-cylinder engine, or the like. Also, although the engine 12 is provided with the EGR device 40 , it may be provided without the EGR device 40 .

In the embodiment, the electronic control unit 70 included in the engine device 10 applies data such as the first downstream pressure Ps, the second downstream pressure Pm, and the pressure error ΔP to the correction value mapping to generate the correction values mtad, mciad, and Vmad. shall be set. However, an external processing device (for example, a server) arranged outside the engine device 10 may set the correction values mtad, mciad, and Vmad. In this case, data such as the first downstream pressure Ps, the second downstream pressure Pm, and the pressure error ΔP are transmitted from the electronic control unit 70 to the external processing device, and the external processing device converts these data into correction value mapping. The correction values mtad, mciad, and Vmad may be applied and transmitted to the electronic control unit 70 . By doing so, the processing load on the electronic control unit 70 can be reduced.

Note that the correspondence relationship between the main elements of the examples and the main elements of the invention described in the column of Means for Solving the Problems is the Since it is an example for specifically explaining the mode for solving the problem, it does not limit the elements of the invention described in the column of the means for solving the problem. That is, the interpretation of the invention described in the column of Means to Solve the Problem should be made based on the description in that column, and the Examples should be based on the description of the invention described in the column of Means to Solve the Problem. This is merely a specific example.

Although the embodiments for carrying out the present invention have been described above, the present invention is not limited to such embodiments at all, and can be modified in various forms without departing from the scope of the present invention. Of course, it can be implemented.

INDUSTRIAL APPLICABILITY The present invention is applicable to the manufacturing industry of state estimation devices.

10, 10B engine device, 12, 12B engine, 14 crankshaft, 14a crank position sensor, 15 water temperature sensor, 16 cam position sensor, 22 air cleaner, 23 intake pipe, 23a air flow meter, 23b intake pressure sensor, 23c boost pressure sensor , 23d throttle downstream portion, 23t intake air temperature sensor, 23u throttle upstream portion, 24 bypass pipe, 25 intercooler, 26 throttle valve, 26a throttle position sensor, 27 surge tank, 27a pressure sensor, 27t temperature sensor, 28 fuel injection valve, 29 intake valve, 30 combustion chamber, 31 spark plug, 32 piston, 33 exhaust valve, 34 variable valve timing device, 35 exhaust pipe, 36 bypass pipe, 37, 38 purification device, 39a front air-fuel ratio sensor, 39b rear air-fuel ratio sensor , 40 EGR device, 42 EGR pipe, 44 EGR valve, 45 opening sensor, 50 supercharger, 51 turbine, 52 compressor, 53 rotating shaft, 54 waste gate valve, 55 blow-off valve, 70 electronic control unit, 71 CPU, 72 ROM, 73 RAM, 74 flash memory.

Claims (1)

A first related value related to a throttle passing air amount passing through a throttle valve arranged in an intake pipe of an engine, a second related value related to a cylinder inflow air amount flowing into a combustion chamber of the engine, and the intake pipe A state estimating device that estimates at least one of a third related value related to a downstream volume, which is the volume of a portion downstream of the throttle valve,
having a storage device and an execution device,
The storage device stores a first downstream pressure, which is a pressure downstream of the throttle valve in the intake pipe detected by a pressure sensor, and the throttle valve in the intake pipe estimated using the amount of air passing through the throttle. An input variable including a second downstream pressure, which is a pressure on the downstream side, and a pressure difference between the first downstream pressure and the second downstream pressure, and the first related value, the second storing a mapping whose output is at least one of the associated value and the third associated value;
The executing device obtains the input variables and applies the obtained input variables to the mapping to estimate at least one of the first, second and third association values. ,
State estimator.

Images