Internal combustion engine control device

Abstract

The present invention addresses the problem of obtaining an internal combustion engine control device that estimates the HC concentration of exhaust gas by a sensorless model calculation, corrects the NH3 concentration and NOx concentration of the exhaust gas, and enables real-time, highly accurate monitoring. The present invention is an internal combustion engine control device 29 comprising an NH3 sensor (22) downstream of a catalyst of an internal combustion engine (1), and characterized by comprising: an HC concentration estimation unit (S1101) that estimates the HC concentration of exhaust gas by a model calculation based on operating conditions of the internal combustion engine; and an NH3 correction unit (S1102) that uses an HC concentration estimation value estimated by the HC concentration estimation unit to correct an NH3 sensor detection value detected by the NH3 sensor.

Description

Control device for internal combustion engine

[0001] The present invention relates to a control device for an internal combustion engine.

[0002] In recent years, automobile exhaust gas regulations have become stricter year by year, creating a demand for on-board monitoring (OBM), which monitors multiple regulated exhaust gas components (emissions) that fall under regulations on harmful substances emitted from internal combustion engines in real time within an ECU. OBM requires real-time and highly accurate measurement of emissions downstream of the catalyst and the presentation (indication) of the measured information to the vehicle user. The presentation (indication) of the measured information to the vehicle user is expected to be, for example, by turning on an abnormality lamp. In the first phase of the OBM regulations of Euro 7, the next European exhaust gas regulations, NH3 sensors and NOx sensors are expected to be installed on vehicles.

[0003] Japanese Patent Application Laid-Open No. 2018-112094

[0004] Patent Document 1 describes a technique for calibrating an NH3 sensor for diesel engines, but the technique aims to improve the slope of the sensor output characteristics offline and is not intended for real-time correction.

[0005] With regard to sensors mounted on automobiles, when the catalyst is below its activation temperature, problems arise in that the NH3 sensor experiences large measurement errors due to HC, and the NOx sensor experiences large measurement errors due to NH3. Here, real-time measurement of HC has traditionally required very expensive and large exhaust analyzers, such as those typically used for research and development purposes, and it is practically impossible to mount such expensive and large exhaust analyzers on automobiles used on general public roads.

[0006] As described above, OBM regulations require monitoring using an NH3 sensor and a NOx sensor, but there is a problem in that when the catalyst is below its activation temperature, the error in the NH3 sensor detection value due to HC becomes large, and the error in the NOx sensor detection value due to NH3 becomes large.

[0007] The present invention has been made in view of these problems, and aims to estimate the HC concentration in the exhaust gas by sensorless model calculation, correct the NH3 concentration and NOx concentration in the exhaust gas, and monitor them in real time with high accuracy.

[0008] The control device for an internal combustion engine of the present invention that solves the above-mentioned problems is a control device for an internal combustion engine equipped with an NH3 sensor downstream of a catalyst of the internal combustion engine, characterized by having: an HC concentration estimation unit that estimates the HC concentration of exhaust gas by model calculation based on the operating conditions of the internal combustion engine; and an NH3 correction unit that corrects the NH3 sensor detection value detected by the NH3 sensor using the HC concentration estimated value estimated by the HC concentration estimation unit.

[0009] According to the present invention, by using an NN model, highly accurate HC concentration estimation substantially equivalent to sensor detection is realized, thereby enabling highly accurate real-time correction of HC sensorless NH3 detection values ​​and complying with OBM exhaust gas regulations. Further features related to the present invention will become apparent from the description of this specification and the accompanying drawings. Furthermore, problems, configurations, and effects other than those described above will become apparent from the following description of the embodiments.

[0010] FIG. 1 is a schematic diagram of an overall configuration of an internal combustion engine control system according to an embodiment of the present invention. FIG. 2 is a longitudinal sectional view of a fuel injection device according to an embodiment of the present invention, and a diagram showing the configuration of a drive circuit and an ECU connected to the fuel injection device. FIG. 3 is an enlarged sectional view of a drive unit structure of a fuel injection device according to an embodiment of the present invention. FIG. 4 is a diagram for explaining a model configuration according to an embodiment of the present invention. FIG. 5 is a diagram for explaining an NN model according to an embodiment of the present invention. FIG. 6 is a diagram showing the correlation between a CFD simulation result of an average in-cylinder temperature at ignition timing and an estimation result using an NN model. FIG. 7 is a diagram showing the correlation between a CFD simulation result of piston adhesion amount at ignition timing and an estimation result using an NN model. FIG. 8 is a diagram showing the correlation between a CFD simulation result of bore adhesion amount at ignition timing and an estimation result using an NN model. FIG. 9 is a diagram showing the correlation between a CFD simulation result of floating fuel droplets at ignition timing and an estimation result using an NN model. FIG. 10 is a diagram showing the correlation between an experimental result of HC using an actual engine and an estimation result using an NN model. FIG. 11 is a block diagram showing a flowchart according to an embodiment of the present invention. FIG. 12 is a diagram showing the relationship between HC concentration and an NH3 sensor detection value. 1 is a diagram showing the relationship between the NH3 concentration and the NOx sensor detection value, and is a diagram for explaining estimation of the HC concentration after catalyst warm-up and correction of the NH3 and NOx sensors according to an embodiment of the present invention.

[0011] 1 is a schematic diagram of an overall configuration of an internal combustion engine control system. The internal combustion engine control system includes an internal combustion engine (hereinafter also referred to as an engine) 1 and an ECU (Electronic Control Unit) 29 attached to the internal combustion engine 1. The ECU 29 is electrically connected to various sensors, actuators, etc. that constitute the internal combustion engine 1, and is an example of an internal combustion engine control device that controls the internal combustion engine 1.

[0012] The internal combustion engine 1 includes a flow sensor 2 (air flow sensor), a turbocharger 3, an air bypass valve 4, an intercooler 5, a supercharging temperature sensor 6, a throttle valve 7, an intake manifold 8, a supercharging pressure sensor 9, a flow enhancement valve 10, an intake valve 11, a valve opening / closing phase sensor 12, an exhaust valve 13, a valve opening / closing phase sensor 14, a fuel injection valve 15, an ignition plug 16, a knock sensor 17, a crank angle sensor 18, a wastegate valve 19, a TWC (Three Way Catalyst) 20, a GPF (Gasoline Particulate Filter) 21, an NH3 sensor 22, a NOx sensor 23, an EGR (Exhausted Gas Recirculation) pipe 24, an EGR cooler 25, an EGR valve 26, an exhaust gas temperature sensor 27, and a differential pressure sensor 28. An intake air temperature sensor (not shown) is attached to a flow rate sensor 2 provided in an intake air flow path of the internal combustion engine 1. The intake air temperature sensor measures the intake air temperature.

[0013] The turbocharger 3 is composed of a compressor 3a having compressor blades facing the intake passage, and a turbine 3b connected to the compressor 3a so as to rotate integrally with the compressor 3a and having turbine blades facing the exhaust passage. The compressor 3a and turbine 3b are rotatably supported within the turbocharger 3. The turbine 3b converts the energy of the exhaust gas from the internal combustion engine 1 into rotational energy. The compressor 3a connected to the turbine 3b compresses the intake air that flows in from the intake passage using the rotational energy of the turbine 3b.

[0014] In the intake air flow path of the internal combustion engine 1, the intercooler 5 is provided downstream of the compressor 3a of the turbocharger 3, and cools the intake air whose temperature has been increased by adiabatic compression by the compressor 3a. The supercharger temperature sensor 6 is mounted downstream of the intercooler 5, and measures the temperature of the intake air cooled by the intercooler 5 (supercharger temperature).

[0015] The throttle valve 7 is provided downstream of the intercooler 5 and restricts the intake flow path to control the amount of intake air flowing into the cylinders of the internal combustion engine 1. The throttle valve 7 is configured as an electronically controlled butterfly valve whose valve opening can be controlled by the ECU 29. Downstream of the throttle valve 7, an intake manifold 8 to which a boost pressure sensor 9 is attached is connected.

[0016] The intake manifold 8 provided downstream of the throttle valve 7 may be integrated with the intercooler 5. In this case, the volume from downstream of the compressor 3a to the cylinder can be reduced, which improves acceleration / deceleration response and controllability.

[0017] The flow enhancement valve 10 is disposed downstream of the intake manifold 8 and enhances the turbulence of the flow inside the cylinder by causing a bias in the intake air drawn into the cylinder. When exhaust gas recirculation combustion, which will be described later, is performed, the ECU 29 can close the flow enhancement valve 10 to promote turbulent combustion and stabilize the combustion that occurs in the combustion chamber.

[0018] The intake valve 11 and the exhaust valve 13 each have a variable valve mechanism for continuously varying the valve opening / closing phase. The variable valve mechanisms of the intake valve 11 and the exhaust valve 13 are equipped with valve opening / closing phase sensors 12, 14 for detecting the valve opening / closing phase. Each cylinder of the internal combustion engine 1 is equipped with a direct injection fuel injection valve 15 that injects fuel pressurized by a fuel pump (not shown) into the combustion chamber of the cylinder. The fuel injection valve 15 may be a port injection valve that injects fuel into the intake passage. Alternatively, multiple direct injection and port injection fuel injection valves 15 may be used for each cylinder. The fuel injection valve 15 is connected to the ECU 29 via a drive circuit (not shown in FIG. 1 ).

[0019] Each cylinder of the internal combustion engine 1 is fitted with an ignition plug 16, which has an electrode exposed inside the cylinder and ignites a combustible mixture with a spark. A knock sensor 17 is provided in the cylinder block and detects the presence or absence of knock by detecting cylinder block vibrations caused by combustion pressure vibrations generated in the combustion chamber. A crank angle sensor 18 is attached to the crankshaft and outputs a signal corresponding to the rotation angle of the crankshaft to the ECU 29.

[0020] The TWC 20 is installed downstream of the turbine 3b of the turbocharger 3 and purifies harmful exhaust gas components such as carbon monoxide, nitrogen compounds, and unburned hydrocarbons in the exhaust gas through catalytic reactions. Furthermore, a GPF 21 is installed downstream of the TWC 20 and collects particulate matter in the exhaust gas. An NH3 sensor 22 that detects ammonia in the exhaust gas and a NOx sensor 23 that detects NOx in the exhaust gas are installed downstream of the GPF 21, and these detection values ​​are sent to the ECU 29.

[0021] The turbocharger 3 is equipped with an air bypass valve 4 and a wastegate valve 19. The air bypass valve 4 is disposed in an intake bypass passage that connects the upstream and downstream of the compressor 3a in the intake passage to prevent an excessive increase in pressure from downstream of the compressor 3a to the upstream of the throttle valve 7. For example, if the throttle valve 7 is suddenly closed in a supercharging state, the air bypass valve 4 is opened under the control of the ECU 29, causing compressed intake air downstream of the compressor 3a to flow back through the bypass passage to the upstream of the compressor 3a. As a result, the supercharging pressure is immediately reduced, preventing a phenomenon known as surging and appropriately preventing damage to the compressor 3a.

[0022] The wastegate valve 19 is disposed in an exhaust bypass passage that connects the upstream and downstream of the turbine 3b in the exhaust passage. The wastegate valve 19 is an electrically operated valve whose valve opening can be freely controlled in accordance with the boost pressure under the control of the ECU 29. When the ECU 29 adjusts the opening of the wastegate valve 19 based on the boost pressure detected by the boost pressure sensor 9, part of the exhaust gas passes through the bypass passage, thereby reducing the energy imparted by the exhaust gas to the turbine 3b. As a result, the wastegate valve 19 can adjust the boost pressure to the target pressure.

[0023] The EGR pipe 24 connects the exhaust flow path downstream of the GPF 21 with the intake flow path upstream of the compressor 3a, and diverts exhaust gas from downstream of the GPF 21 and recirculates it upstream of the compressor 3a. An EGR cooler 25 provided in the EGR pipe 24 cools the diverted exhaust gas. An EGR valve 26 is provided in the EGR pipe 24 between the EGR cooler 25 and the upstream of the compressor 3a and controls the flow rate of exhaust gas recirculated upstream of the compressor 3a. The EGR pipe 24 is also provided with an exhaust gas temperature sensor 27 that detects the exhaust gas temperature of the exhaust gas upstream of the EGR valve 26 and a differential pressure sensor 28 that detects the differential pressure between the upstream and downstream of the EGR valve 26.

[0024] The ECU 29 is a control device that has a CPU (Central Processing Unit), ROM (Read Only Memory), RAM (Random Access Memory), an A / D (Analog-to-Digital) converter, a driver circuit, etc., and controls each component of the internal combustion engine 1 and executes various data processing. The ECU 29 controls the operation of actuators such as the throttle valve 7, the fuel injection valve 15, the variable valve mechanisms of the intake valve 11 and the exhaust valve 13, and the EGR valve 26. The ECU 29 also detects the operating state of the internal combustion engine 1 based on signals input from various sensors, and ignites the spark plug 16 at a timing determined according to the operating state. The ECU 29 also has a function of monitoring the detected values ​​of the NH3 sensor 22 and the NOx sensor 23, and if the emissions deteriorate, to notify the vehicle user by turning on an abnormality lamp or the like.

[0025] Next, the configuration and basic operation of a fuel injection device and its control device will be described. Figure 2 is a diagram showing an example of the configuration of a fuel injection device 200. The fuel injection device 200 is driven by a drive circuit 103, which is controlled by an ECU 29. The ECU 29 receives signals indicating the state of the internal combustion engine 1 from various sensors, and calculates the width of an injection pulse and injection timing to control the injection amount injected from the fuel injection device 200 according to the operating conditions of the internal combustion engine 1.

[0026] The ECU 29 is also equipped with an A / D converter and an I / O port for receiving signals from various sensors. The injection pulses output from the ECU 29 are input to the drive circuit 103 of the fuel injection device 200 via a signal line 110. The drive circuit 103 controls the voltage applied to the solenoid 205 and supplies current. The ECU 29 communicates with the drive circuit 103 via a communication line 111, and is able to switch the drive current generated by the drive circuit 103 depending on the pressure of the fuel supplied to the fuel injection device 200 and the operating conditions, and to change the set values ​​of the current and time.

[0027] Next, the configuration and operation of fuel injection device 200 will be described using the vertical cross section of fuel injection device 200 in Fig. 2 and an enlarged cross section of the vicinity of movable element 202 and valve element 214 in Fig. 3. Fuel injection device 200 shown in Fig. 2 and Fig. 3 is a normally closed solenoid valve (electromagnetic fuel injection device), and when solenoid 205 is not energized, valve element 214 is urged in the valve closing direction by spring 210, which is a first spring, and valve element 214 is in close contact with valve seat 218, maintaining a closed valve state.

[0028] In the valve-closed state, a force acting in the valve-open direction by the return spring 212 of the second spring acts on the movable element 202. At this time, the force acting on the valve element 214 by the spring 210 is greater than the force by the return spring 212, so the end face 302 of the movable element 202 comes into contact with the valve element 214, and the movable element 202 remains stationary.

[0029] The valve element 214 and the movable element 202 are configured to be relatively displaceable and are contained within the nozzle holder 201. The nozzle holder 201 also has an end surface 303 that serves as a spring seat for the return spring 212. The force of the spring 212 is adjusted during assembly by the amount of pressure applied to a spring retainer 224 that is fixed to the inner diameter of the fixed core 207.

[0030] The fuel injection device 200 also has a magnetic circuit formed by a fixed core 207, a moving element 202, a nozzle holder 201, and a housing 203, with a gap between the moving element 202 and the fixed core 207. A magnetic restriction 211 is formed in the nozzle holder 201 at a portion corresponding to the gap between the moving element 202 and the fixed core 207. The solenoid 205 is attached to the outer periphery of the nozzle holder 201 while being wound around a bobbin 204.

[0031] A rod guide 215 is provided near the tip of the valve element 214 on the valve seat 218 side so as to be fixed to the nozzle holder 201. The movement of the valve element 214 in the valve axis direction is guided by two sliding points: the spring seat 207 of the valve element 214 and the rod guide 215. An orifice 216, in which a valve seat 218 and a fuel injection hole 219 are formed, is fixed to the tip of the nozzle holder 201, and the internal space (fuel passage) provided between the moving element 202 and the valve element 214 is sealed from the outside.

[0032] Fuel supplied to the fuel injection device 200 is supplied from a rail pipe (not shown) provided upstream of the fuel injection device 200, flows through a first fuel passage hole 231 to the tip of the valve element 214, and is sealed by a seat portion formed at the end of the valve element 214 on the valve seat 218 side and the valve seat 218. When the valve is closed, a pressure difference is generated between the upper and lower parts of the valve element 214 due to fuel pressure, and the valve element 114 is pushed in the valve closing direction by the pressure difference calculated by multiplying the fuel pressure by the pressure-receiving area of ​​the seat inner diameter at the valve seat position and the load of the spring 210.

[0033] When current is supplied to the solenoid 205 in the closed valve state, a magnetic field is generated in the magnetic circuit, and magnetic flux passes between the fixed core 207 and the movable member 202, causing a magnetic attractive force to act on the movable member 202. When the magnetic attractive force acting on the movable member 202 exceeds the differential pressure and the load of the set spring 210, the movable member 202 begins to displace in the direction of the fixed core 207.

[0034] After the valve element 214 starts its valve-opening operation, the movable element 202 moves to the position of the fixed core 207 and collides with the fixed core 207. After the movable element 202 collides with the fixed core 207, the movable element 202 receives a reaction force from the fixed core 207 and bounces back, but the magnetic attractive force acting on the movable element 202 causes the movable element 202 to be attracted to the fixed core 207 and eventually stops.

[0035] At this time, the return spring 212 acts on the mover 202 in the direction of the fixed core 207, so the time until the rebound converges can be shortened. Since the rebound action is small, the time during which the gap between the mover 202 and the fixed core 207 becomes large is shortened, and stable operation can be performed even for a smaller injection pulse width.

[0036] After completing the valve-opening operation in this manner, the moving element 202 and the valve element 214 remain stationary in the open state. In the open state, a gap is formed between the valve element 214 and the valve seat 218, and fuel is injected from the fuel injection holes 219. The fuel flows downstream through a central hole in the fixed core 207 and a lower fuel passage hole 305 in the moving element 202.

[0037] When the solenoid 205 is de-energized, the magnetic flux generated in the magnetic circuit disappears, and the magnetic attractive force also disappears. As the magnetic attractive force acting on the armature 202 disappears, the armature 202 and the valve element 214 are pushed back to the valve-closed position in which they contact the valve seat 218 by the load of the spring 210 and the differential pressure.

[0038] Furthermore, when the valve element 214 closes from an open state, after the valve element 214 comes into contact with the valve seat 218, the movable element 202 separates from the valve element 214 and moves in the valve closing direction, and after moving for a certain period of time, is returned to the initial position in the closed state by the return spring 212. By the movable element 202 separating from the valve element 214 at the moment the valve element 214 completes opening, the mass of the movable member at the moment the valve element 214 collides with the valve seat 218 can be reduced by the mass of the movable element 202, and therefore the collision energy when the valve element 214 collides with the valve seat 218 can be reduced and the bounding of the valve element 214 caused by the collision of the valve element 214 with the valve seat 218 can be suppressed.

[0039] In the fuel injection device 200 of this embodiment, the valve body 214 and the moving element 202 undergo relative displacement for a short period of time at the moment when the moving element 202 collides with the fixed core 207 when the valve is opened, and at the moment when the valve body 214 collides with the valve seat 218 when the valve is closed, thereby suppressing the moving element 202 from bouncing against the fixed core 207 and the valve body 214 from bouncing against the valve seat 218.

[0040] The model configuration for HC concentration estimation in this embodiment is shown in Figure 4. The fuel injection device 200 of this embodiment estimates the HC concentration of the exhaust gas by model calculation based on the operating conditions of the internal combustion engine 1, corrects the detected value of the NH3 sensor using the estimated HC concentration, and then corrects the detected value of the NOx sensor using the corrected detected value of the NH3 sensor.

[0041] First, the HC concentration of the exhaust gas is estimated by model calculation based on the operating conditions of the internal combustion engine. In the fuel injection device 200 of this embodiment, multi-stage injection control is performed during cold start of the internal combustion engine 1 to reduce adhesion of injected fuel to the cylinder wall surface (S401). Then, wall adhesion NN estimation is performed using a neural network (NN) model to estimate the amount of fuel adhesion to the cylinder wall surface using parameters from the multi-stage injection control (S402). Next, HC concentration NN estimation is performed using the estimated amount of fuel adhesion to the cylinder wall surface using the NN model (S403). In other words, the HC concentration is estimated using a two-stage NN model.

[0042] Next, the configuration of the NN model in one embodiment of the present invention will be described with reference to Figure 5. The NN model of wall surface adhesion NN estimation S402 has blocks that use engine type 501, rotation speed 502, charging efficiency 503, intake air temperature 504, coolant temperature 505, fuel pressure 506, number of injection stages 507, ignition timing 508, injection start timing 509-511, and injection ratios 512-514, which correspond to the operating conditions of the internal combustion engine 1, as input conditions, and calculates in-cylinder gas temperature 515 at the ignition timing, piston adhesion amount 516, bore adhesion amount 517, and floating droplet amount 518.

[0043] Engine type 501 is classified according to differences in the engine bore, stroke, fuel injection method, etc. Piston adhesion amount 516 is the amount of fuel adhering to the piston head, which constitutes part of the inner cylinder wall surface, and bore adhesion amount 517 is the amount of fuel adhering to the inner circumferential surface of the bore, which constitutes part of the inner cylinder wall surface. Suspended droplet amount 518 is the amount of fuel droplets floating in the combustion chamber. Number of injection stages 507 is the number of times fuel is injected before the explosion stroke, and in this embodiment, an example is shown in which fuel is injected in three stages, as shown by injection start times 509-511 and injection rates 512-514.

[0044] Each block of the NN model that calculates the in-cylinder gas temperature 515, piston adhesion amount 516, bore adhesion amount 517, and floating droplet amount 518 uses the results of machine learning using teacher data in which the operating conditions of engine type 501, rotational speed 502, charging efficiency 503, intake air temperature 504, cooling water temperature 505, fuel pressure 506, number of injection stages 507, ignition timing 508, injection start timing 509-511, and injection ratio 512-514 are used as inputs, and the analysis results of the in-cylinder gas temperature 515, piston adhesion amount 516, bore adhesion amount 517, and floating droplet amount 518 at the ignition timing obtained by in-cylinder simulation using CFD (Computational Fluid Dynamics) under the same conditions are used as outputs.

[0045] Next, in the NN model of HC concentration NN estimation S403, the engine type 501, rotation speed 502, ignition timing 508, and in-cylinder gas temperature 515, piston adhesion amount 516, bore adhesion amount 517, and floating droplet amount 518 at the ignition timing calculated by each block of wall adhesion NN estimation S402 are used as input conditions, and HC concentration estimation 519 is performed to estimate the HC concentration of the exhaust.

[0046] Here, the NN model of the HC concentration NN estimation S403 uses the results of machine learning using training data in which the engine type 501, rotation speed 502, ignition timing 508, in-cylinder gas temperature 515 calculated by CFD, piston adhesion amount 516, bore adhesion amount 517, and floating droplet amount 518 are input, and the HC concentration measured by an actual machine is output.

[0047] Regarding fuel, in addition to regular gasoline, it is also possible to consider the use of carbon-neutral fuels such as biofuels and synthetic fuels, and add fuel types to the operating conditions 501-514 that are input to each NN model. This makes it possible to perform evaluations according to fuel properties.

[0048] Below, an example of the calculation results for each block of the NN model is shown.

[0049] Figure 6 shows the results of in-cylinder gas temperature at ignition timing. The horizontal axis shows the CFD simulation results, and the vertical axis shows the calculation results using the NN model, and the data is plotted. A high correlation is observed between the CFD results and the NN model results, indicating that the NN model can produce reasonable results.

[0050] Similarly, the results for the amount of fuel adhering to the piston at the ignition timing are shown in Figure 7. The results of the CFD simulation and the calculation results using the NN model show a high correlation, and it can be said that evaluation is possible using the NN model.

[0051] Similarly, the results for the amount of fuel adhering to the bore at the ignition timing are shown in Figure 8. The simulation results using CFD and the calculation results using the NN model show a high correlation, indicating that the evaluation using the NN model is valid.

[0052] Similarly, Figure 9 shows the results for the amount of floating droplets at the ignition timing. Floating droplets are fuel droplets that are floating without adhering to the inner cylinder wall surface. Here too, a high correlation was obtained between the simulation results using CFD and the calculation results using the NN model, demonstrating the validity of the results using the NN model.

[0053] In addition, with regard to calculation time, so-called in-cylinder simulations using CFD typically require several hours for a single calculation, whereas with an NN model, once machine learning has been performed, it is possible to execute the calculations in near real time.

[0054] Finally, Figure 10 shows the correlation between the measurement results of the HC concentration in exhaust gas from an experiment using an internal combustion engine and the calculation results of the HC concentration using the NN model. The two results are almost the same, demonstrating that the HC concentration can be estimated using the NN model.

[0055] A flowchart of one embodiment of the present invention is shown in Figure 11. The ECU 29 has the following functions, which are realized by the CPU executing a software program in the memory: an HC concentration estimation means, an NH3 correction means, an NOx correction means, and a monitoring device output means.

[0056] The HC concentration estimation means estimates the amount of fuel adhering to the wall surface of the cylinder at the ignition timing based on the number of injection stages, injection phase, and fuel pressure, which are fuel injection control parameters of the internal combustion engine 1, the cooling water temperature or the cylinder inner wall temperature, and the filling efficiency, which is load information, and estimates the HC concentration based on the estimated amount of fuel adhering to the wall surface, the rotation speed of the internal combustion engine 1, and the ignition timing.

[0057] The NH3 correction means calculates an NH3 correction value, which is a corrected NH3 concentration, by correcting the NH3 sensor detection value using the HC concentration estimated by the HC concentration estimation means. The NOx correction means calculates an NOx correction value, which is a corrected NOx concentration, by correcting the NOx sensor detection value using the NH3 correction value calculated by the NH3 correction means. The monitoring device output means performs processing to output the NH3 correction value and the NOx correction value to the monitoring device as on-board monitoring results.

[0058] The ECU 29 inputs the operating conditions of the internal combustion engine 1 (number of injection stages, injection phase (injection timing), fuel pressure, cylinder inner wall temperature, etc.) and outputs an HC concentration estimation value using an NN model-based HC concentration estimation means (HC concentration estimation unit) (S1101).

[0059] Next, the NH3 sensor detection value and the HC concentration estimated value are input, and the NH3 correction means (NH3 correction unit) corrects the NH3 sensor detection value using the HC concentration estimated value and outputs the corrected NH3 concentration (NH3 correction value) (S1102). ＨＣ The NH3 sensor detection value is C ＮＨ３ d, the corrected NH3 concentration C ＮＨ３ For example, c is expressed by the formula (1). NH3 c= C NH3 d-k NH3_HC ×C HC　　　　　　　 ...Formula (1)

[0060] k ＮＨ３＿ＨＣ indicates the sensitivity coefficient of HC detected by the NH3 sensor, and is given as the slope of the HC concentration and the NH3 sensor detection value as shown in FIG.

[0061] Next, the NOx sensor detection value and the corrected NH3 concentration are input, and the NOx correction means (NOx correction unit) corrects the NOx sensor detection value using the corrected NH3 concentration and outputs the corrected NOx concentration (NOx correction value) (S1103). ＮＯｘ d, the corrected NOx concentration C ＮＯｘ For example, c is expressed by the formula (2). NOx c= C NOx d-k NOx_NH3 ×C NH3 c 　　　　　　　 ...Formula (2)

[0062] k ＮＯｘ＿ＮＨ３indicates the sensitivity coefficient of NH3 detected by the NOx sensor, and is given as the slope of the NH3 concentration and the NOx sensor detection value as shown in Figure 13. Finally, the corrected NH3 concentration and corrected NOx concentration are output to the monitoring device (OBM) (S1104). As a result, the monitoring device (OBM) can monitor the NH3 concentration and NOx concentration corrected with high accuracy in real time.

[0063] According to the internal combustion engine control device of this embodiment, by using an NN model, it is possible to realize HC concentration estimation with a high degree of accuracy substantially equivalent to that of sensor detection. Therefore, it is possible to perform high-precision real-time correction of HC sensorless detected NH3 and NOx values, and it is possible to comply with OBM exhaust gas regulations.

[0064] Second Embodiment Next, a second embodiment of the present invention will be described with reference to Fig. 14. A distinctive feature of this embodiment is that the HC concentration downstream of the catalyst is estimated using the NN model even after the catalyst has warmed up, and the detected values ​​of the NH3 and NOx sensors are corrected.

[0065] When the engine is started cold and the catalyst is below its activation temperature, the HC emitted from the engine will remain at almost the same concentration and cause an error in the NH3 sensor. However, after the catalyst has warmed up, the HC is purified, so the effect is reduced.

[0066] In this case, as shown in FIG. 14 , the exhaust gas temperature and the exhaust gas air-fuel ratio are acquired (S1401), the catalyst temperature and the oxygen storage rate are calculated based on the acquired exhaust gas temperature and exhaust gas air-fuel ratio (S1402), and the catalyst purification efficiency is calculated from the catalyst temperature and the oxygen storage rate (S1403).

[0067] Then, the HC concentration estimation result at engine-out (S1404), that is, the HC concentration estimation result obtained by model calculation using the above-mentioned NN model for HC exhaust from the internal combustion engine 1, is used to estimate an estimated HC concentration downstream of the catalyst after catalyst warm-up, which is the HC concentration downstream of the catalyst after catalyst warm-up, by taking into consideration the catalyst purification efficiency of S1403 (S1405).Then, the detected values ​​of the NH3 sensor and the NOx sensor are corrected based on the estimated HC concentration downstream of the catalyst after catalyst warm-up (S1406).

[0068] Therefore, it is possible to achieve highly accurate HC concentration estimation even after the catalyst has warmed up, enabling highly accurate real-time correction of the NH3 detection value and NOx detection value, and making it possible to comply with OBM exhaust gas regulations.

[0069] Although the embodiments of the present invention have been described in detail above, the present invention is not limited to the above-described embodiments, and various design modifications can be made without departing from the spirit of the present invention as defined in the claims. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to those including all of the described configurations. Furthermore, it is possible to replace part of the configuration of one embodiment with the configuration of another embodiment, or to add the configuration of another embodiment to the configuration of one embodiment. Furthermore, it is possible to add, delete, or replace part of the configuration of each embodiment with other configurations.

[0070] 1... Internal combustion engine 29... ECU (control device) S1101... HC concentration estimation means (HC concentration estimation section) S1102... NH3 correction means (NH3 correction section) S1103... NOx correction means (NOx correction section) S1104... Monitoring device output

Claims

1. A control device for an internal combustion engine equipped with an NH3 sensor downstream of a catalyst of the internal combustion engine, characterized by having: an HC concentration estimation unit that estimates the HC concentration of exhaust gas emitted from the internal combustion engine by model calculation based on the operating conditions of the internal combustion engine; and an NH3 correction unit that corrects the NH3 sensor detection value detected by the NH3 sensor using the HC concentration estimated value estimated by the HC concentration estimation unit.

2. The control device for an internal combustion engine according to claim 1, characterized in that the internal combustion engine is equipped with a NOx sensor downstream of the catalyst, and the control device has a NOx correction unit that corrects the NOx sensor detection value detected by the NOx sensor using the NH3 correction value corrected by the NH3 correction unit.

3. The control device for an internal combustion engine as described in claim 2, characterized in that the HC concentration estimation unit estimates the amount of fuel adhering to the wall surface in the cylinder at the ignition timing based on the number of injection stages, injection phase, and fuel pressure, which are fuel injection control parameters of the internal combustion engine, the cooling water temperature or the cylinder inner wall temperature, and the filling efficiency, which is load information, and estimates the HC concentration based on the estimated amount of fuel adhering to the wall surface, the rotational speed of the internal combustion engine, and the ignition timing.

4. The control device for an internal combustion engine according to claim 3, wherein the HC concentration estimation unit estimates the amount of fuel adhering to the wall surface using a neural network model.

5. The control device for an internal combustion engine according to claim 4, wherein the HC concentration estimation unit calculates the estimated HC concentration value using a neural network model.

6. The control device for an internal combustion engine according to claim 5, wherein the NH3 correction value and the NOx correction value corrected by the NOx correction unit are output to a monitoring device as on-board monitoring results.

7. A control device for an internal combustion engine as described in claim 6, characterized in that the control device acquires an exhaust gas temperature and an exhaust gas air-fuel ratio of the internal combustion engine, calculates a catalyst temperature and an oxygen storage rate of the catalyst using the exhaust gas temperature and the exhaust gas air-fuel ratio, calculates a catalyst purification efficiency of the catalyst based on the catalyst temperature and oxygen storage rate of the catalyst, calculates an estimated HC concentration value downstream of the catalyst after catalyst warm-up using the catalyst purification efficiency and the estimated HC concentration value estimated by the model calculation, the NH3 correction unit corrects the NH3 sensor detection value detected by the NH3 sensor using the estimated HC concentration value downstream of the catalyst after catalyst warm-up, and the NOx correction unit corrects the NOx sensor detection value detected by the NOx sensor using the NH3 correction value corrected by the NH3 correction unit.

Images