JP6036562B2 - Heat generation rate waveform creation device and combustion state diagnostic device for internal combustion engine - Google Patents

Description

  The present invention relates to an apparatus for creating a heat release rate waveform of an internal combustion engine such as a diesel engine, and an apparatus for diagnosing an actual combustion state using the created heat release rate waveform.

  As is well known in the art, in a diesel engine (hereinafter sometimes simply referred to as an engine) used as an automobile engine or the like, each control parameter such as a fuel injection amount is corrected according to the engine operating state. In this case, it is desirable to recognize the reaction state of the fuel in the cylinder (hereinafter sometimes referred to as a combustion state) and correct each control parameter accordingly so as to obtain a desired reaction state.

  Thus, as one of the means for correcting each control parameter according to the reaction state of the fuel in the cylinder, the heat generation rate waveform at the time of combustion is obtained, and each heat generation rate waveform is set to an ideal waveform. It is known to correct control parameters (Patent Document 1).

JP 2011-106334 A JP 2010-144573 A JP 2010-144527 A

  For example, in the case of diagnosing a combustion state by comparing an ideal heat generation rate waveform (hereinafter referred to as an ideal heat generation rate waveform) with an actual heat generation rate waveform, the ideal heat generation rate waveform is used. It is important to obtain the waveform components such as the reaction start timing and the reaction rate of the fuel with high accuracy and to optimize the ideal heat generation rate waveform.

  In view of this point, the inventors of the present invention have focused on the fact that the properties of the fuel affect the waveform component. That is, we focused on the fact that the ideal heat generation rate waveform can be optimized by defining the waveform components based on the properties of the fuel.

  Patent Document 2 discloses that when heavy fuel with poor volatility is used, combustion deteriorates and a target heat generation amount cannot be obtained. Further, in Patent Document 3, it is determined that the fuel is light when the peak value of the heat generation rate is larger than the threshold value, and it is determined that the fuel is heavy when the peak value is smaller than the threshold value. Is disclosed.

  However, no proposal has been made to use the properties of fuel to define the shape of the ideal heat generation rate waveform.

  The present invention has been made in view of such a point, and an object of the present invention is to provide a heat generation rate waveform generation device for an internal combustion engine capable of defining the combustion state of fuel in a cylinder with high accuracy and The object is to provide a combustion state diagnostic apparatus.

-Solution principle of the invention-
The solution principle of the present invention devised in order to achieve the above object is to optimize the shape of the ideal heat generation rate waveform by defining the waveform component based on the properties of the fuel used. Yes. Furthermore, the shape of the ideal heat generation rate waveform is optimized by changing the degree of influence of the fuel properties on the ideal heat generation rate waveform in accordance with the oxygen density and fuel density in the cylinder.

-Solution-
Specifically, the present invention is directed to an apparatus for creating a heat release rate waveform of the fuel reaction in an internal combustion engine that burns fuel injected into a cylinder from a fuel injection valve. When creating the ideal heat generation rate waveform of the reaction of the fuel injected from the fuel injection valve by defining the shape of the ideal heat generation rate waveform for the heat generation rate waveform creating device, Of the low-temperature oxidation reaction, thermal decomposition reaction, and high-temperature oxidation reaction that are the reactions of the fuel, the low-temperature oxidation reaction and the thermal decomposition reaction are performed according to only the oxygen density in the cylinder and the oxygen density in the cylinder. The ideal heat generation rate waveform of the minute and the ideal heat generation rate waveform of the heavy component, the lower the oxygen density in the cylinder, the larger the phase difference of these ideal heat generation rate waveforms, and the reaction amount of the heavy component The ratio of the reaction amount of the light component to the light is reduced, and the reaction gradient of these ideal heat generation rate waveforms is reduced to define the shape of the ideal heat generation rate waveform. On the other hand, for the high temperature oxidation reaction, the oxygen density in the cylinder And fuel The ideal heat generation rate waveform of light fuel and the ideal heat generation rate waveform of heavy fuel, depending on both, the lower the oxygen density in the cylinder, the more the phase difference between these ideal heat generation waveforms The ratio of the reaction amount of the light component to the reaction amount of the heavy component is decreased, the reaction gradient of these ideal heat generation rate waveforms is decreased, and the ideal heat generation rate waveform becomes lower as the fuel density is lower. The reaction gradient is reduced to define the shape of the ideal heat release rate waveform .

  The “ideal heat generation rate waveform” here refers to a state in which a fuel injection amount corresponding to the command injection amount, a fuel injection pressure corresponding to the command injection pressure, and a fuel injection period corresponding to the command injection period are secured. It is a heat generation rate waveform that should be theoretically obtained assuming that the combustion efficiency is sufficiently high.

  When the shape of the ideal heat generation rate waveform changes due to the influence of the fuel properties such as the light and heavy components of the fuel, the shape of the ideal heat generation rate waveform is defined according to the specific matter. It becomes possible to create an appropriate ideal heat generation rate waveform. For this reason, it becomes possible to obtain high reliability in the created ideal heat generation rate waveform.

  The “creation of the ideal heat generation rate waveform” in the present invention is not limited to the one that actually draws the ideal heat generation rate waveform. It is a concept that includes a state in which the heat generation amount for each unit rotation angle is defined.

  In addition, the ideal heat release rate waveform is created by using an ideal triangle consisting of a starting point of each reaction of the fuel, a reaction rate as a slope, a reaction amount as an area, and a reaction period as a base length. It is created by creating a heat release rate waveform model and smoothing the ideal heat release rate waveform model of each reaction by filtering.

  In this case, the ideal heat generation rate waveform model includes the light component ideal heat generation rate waveform model corresponding to the light reaction of fuel and the heavy ideal heat generation rate waveform model corresponding to the reaction of heavy fuel. The ideal heat generation rate waveform model is created by smoothing the light component ideal heat generation rate waveform model and the heavy component ideal heat generation rate waveform model by filtering.

  By creating a heat release rate waveform model that approximates a triangle in this way and creating an ideal heat release rate waveform using this heat release rate waveform model, the calculation process for the creation is simplified. It is possible to reduce the load on the calculation means such as the ECU.

  Specific examples of the apparatus for diagnosing the combustion state using the ideal heat generation rate waveform obtained by the above-described heat generation rate waveform generating apparatus for an internal combustion engine include the following configuration. That is, the ideal heat generation rate waveform is compared with the actual heat generation rate waveform when the fuel actually reacts in the cylinder, and the deviation of the actual heat generation rate waveform from the ideal heat generation rate waveform becomes a predetermined amount or more. In this case, it is configured to diagnose that an abnormality has occurred in the fuel reaction.

  The “reaction abnormality” herein is not limited to a reaction abnormality (such as equipment failure) that hinders the operation of the internal combustion engine, but can correct (or learn) the control parameters of the internal combustion engine (for example, This includes the case where there is a deviation in the heat generation rate waveform to such an extent that correction can be made to keep exhaust emission and combustion noise within the limits of regulation.

  By this specific matter, when the actual heat generation rate waveform deviates from the ideal heat generation rate waveform by a predetermined amount or more in the fuel reaction (reaction form), it is diagnosed that the reaction is abnormal. Become. For example, when diagnosis is performed for each of a plurality of reactions, the characteristics (reaction start temperature, reaction rate, etc.) of each fuel reaction are different from each other. By comparing the characteristics of the actual heat release rate waveform (measured), it is possible to identify the reaction in which an abnormality has occurred with high accuracy. For this reason, improvement of diagnostic accuracy can be aimed at. Then, by taking improvement measures (for example, correction of control parameters of the internal combustion engine) for the reaction form diagnosed as abnormal, a control parameter suitable for the reaction form diagnosed as abnormal is selected, The control parameter can be corrected. For this reason, the controllability of the internal combustion engine can be greatly improved.

  Specific operations when it is diagnosed that an abnormality has occurred in the reaction include the following. That is, when there is a reaction in which the deviation of the actual heat generation rate waveform with respect to the ideal heat generation rate waveform is equal to or greater than a predetermined abnormality determination deviation amount, and when it is diagnosed that the reaction is abnormal, When the deviation of the actual heat generation rate waveform from the ideal heat generation rate waveform is equal to or less than a predetermined correctable deviation amount, the control parameter of the internal combustion engine is corrected so that the deviation is less than the abnormality determination deviation amount. On the other hand, when the deviation of the actual heat generation rate waveform from the ideal heat generation rate waveform exceeds the correctable deviation amount, the internal combustion engine is diagnosed as having a failure.

  In this way, when it is diagnosed that an abnormality has occurred in the reaction, whether or not the abnormality can be resolved is determined based on the deviation amount of the actual heat generation rate waveform from the ideal heat generation rate waveform. I am doing so. For this reason, it is possible to accurately discriminate between a state in which a normal reaction state is obtained by correcting the control parameter and a state in which maintenance such as component replacement is required.

  In addition, as the control parameter when the control parameter of the internal combustion engine is corrected so that the deviation is less than the abnormality determination deviation amount, an oxygen amount and a fuel amount in the cylinder are exemplified. The amount of oxygen in the cylinder is determined by the oxygen density and can be adjusted by the EGR rate, the intake air supercharging rate, and the like. The amount of fuel in the cylinder is determined by the fuel density and can be adjusted by the fuel injection timing, the fuel injection pressure, and the fuel injection amount. On the other hand, an example of diagnosing that a failure has occurred in the internal combustion engine is a case where the deviation of the actual heat generation rate waveform exceeds the correctable deviation amount. In this case, the control parameter of the internal combustion engine Since the correction amount exceeds a predetermined guard value, it is possible to diagnose that a failure has occurred in the internal combustion engine. Specifically, when a lower limit value is set in advance for each of the in-cylinder temperature, oxygen density, and fuel density, and any of these in-cylinder temperature, oxygen density, and fuel density is lower than the lower limit value, If the correction amount of the control parameter of the internal combustion engine exceeds a predetermined guard value, it is diagnosed that a failure has occurred in the internal combustion engine.

  Specifically, the usage state of the combustion state diagnosis device for the internal combustion engine includes mounting on a vehicle or mounting on an experimental device.

  In the present invention, it is possible to obtain high reliability in the ideal heat generation rate waveform by defining the shape of the ideal heat generation rate waveform based on the properties of the fuel used. Further, when the abnormality diagnosis of the combustion state is performed using the ideal heat generation rate waveform, the diagnosis accuracy can be improved.

It is a figure which shows schematic structure of the diesel engine which concerns on embodiment, and its control system. It is sectional drawing which shows the combustion chamber of a diesel engine, and its peripheral part. It is a block diagram which shows the structure of control systems, such as ECU. It is a schematic diagram of the intake / exhaust system and the combustion chamber for explaining the outline of the combustion mode in the combustion chamber. It is sectional drawing which shows a combustion chamber at the time of main injection execution, and its peripheral part. It is a top view of a combustion chamber at the time of main injection execution. It is a flowchart figure which shows the procedure of a combustion state diagnosis and control parameter correction | amendment. It is a figure showing the presence or absence of the influence of the oxygen density and fuel density with respect to the high temperature oxidation reaction by low temperature oxidation reaction, thermal decomposition reaction, high temperature oxidation reaction by premixed combustion, and diffusion combustion, respectively. It is a figure which shows an example of a rotational speed correction coefficient map. It is a figure which shows an example of the correction | amendment retardation amount map which uses oxygen density as a parameter. It is a figure which shows an example of the gradient correction coefficient map which uses oxygen density as a parameter. It is a figure which shows an example of the gradient correction coefficient map which uses a fuel density as a parameter. It is a figure which shows an example of the phase difference map which uses oxygen density as a parameter. It is a figure which shows an example of the division | segmentation ratio map which uses oxygen density as a parameter. FIG. 15A shows a case where the ideal heat generation rate waveform model is an isosceles triangle, and FIG. 15B shows a case where the ideal heat generation rate waveform model is an unequal triangle. FIG. FIG. 16 (a) shows the relationship between the elapsed time and the amount of fuel supplied into the cylinder when fuel injection is performed from the injector, and FIG. 16 (b) shows the reaction of the fuel injected in each injection period. It is a figure which shows quantity. It is a figure for demonstrating the change of the light part ideal heat release rate waveform model at the time of the low oxygen density in each of a low temperature oxidation reaction and a high temperature oxidation reaction, and a heavy part ideal heat release rate waveform model. It is a figure which shows an example of the ideal heat release rate waveform model in each reaction form when one fuel injection is performed. It is a figure which shows the ideal heat release rate waveform obtained by filtering the ideal heat release rate waveform model of FIG. It is a figure which shows an example of the ideal heat release rate waveform (solid line) and the actual heat release rate waveform (a broken line and a dashed-dotted line) in case fuel injection is performed once.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the present embodiment, a case where the combustion state diagnosis device according to the present invention is mounted on a common rail in-cylinder direct injection multi-cylinder (for example, in-line 4-cylinder) diesel engine (compression self-ignition internal combustion engine) mounted on an automobile. explain.

-Engine configuration-
FIG. 1 is a schematic configuration diagram of a diesel engine 1 (hereinafter simply referred to as an engine) and its control system according to the present embodiment.

  As shown in FIG. 1, the engine 1 according to this embodiment is configured as a diesel engine system having a fuel supply system 2, a combustion chamber 3, an intake system 6, an exhaust system 7 and the like as main parts.

  The fuel supply system 2 includes a supply pump 21, a common rail 22, an injector (fuel injection valve) 23, an engine fuel passage 27, and the like.

  The supply pump 21 raises the fuel pumped from the fuel tank to a high pressure and then supplies the fuel to the common rail 22 via the engine fuel passage 27. The common rail 22 has a function as a pressure accumulation chamber that holds (accumulates) high pressure fuel at a predetermined pressure, and distributes the accumulated fuel to the injectors 23, 23,. The injector 23 is a piezo injector provided with a piezoelectric element (piezo element) inside.

  The intake system 6 includes an intake manifold 63 connected to an intake port 15 a formed in the cylinder head 15 (see FIG. 2), and an intake pipe 64 is connected to the intake manifold 63. The intake system 6 is provided with an air cleaner 65, an air flow meter 43, and an intake throttle valve (diesel throttle) 62 in order from the upstream side.

  The exhaust system 7 includes an exhaust manifold 72 connected to an exhaust port 71 formed in the cylinder head 15, and an exhaust pipe 73 is connected to the exhaust manifold 72. The exhaust system 7 is provided with an exhaust purification unit 77. The exhaust purification unit 77 is provided with an NSR (NOx Storage Reduction) catalyst 75 and a DPF (Diesel Particle Filter) 76 as NOx occlusion reduction type catalysts.

  As shown in FIG. 2, the cylinder block 11 is formed with a cylinder bore 12 for each cylinder (four cylinders), and a piston 13 is accommodated in each cylinder bore 12 so as to be slidable in the vertical direction. .

  A combustion chamber 3 is formed above the top surface 13 a of the piston 13. That is, the combustion chamber 3 is defined by the lower surface of the cylinder head 15 attached to the upper portion of the cylinder block 11, the inner wall surface of the cylinder bore 12, and the top surface 13 a of the piston 13. A cavity (concave portion) 13 b is formed in a substantially central portion of the top surface 13 a of the piston 13, and this cavity 13 b also constitutes a part of the combustion chamber 3.

  As the shape of the cavity 13b, the concave dimension is small in the central portion (on the cylinder center line P), and the concave dimension is increased toward the outer peripheral side.

  The piston 13 is connected to a crankshaft that is an engine output shaft by a connecting rod 18. Further, a glow plug 19 is disposed toward the combustion chamber 3.

  The cylinder head 15 is provided with an intake valve 16 for opening and closing an intake port 15a and an exhaust valve 17 for opening and closing an exhaust port 71.

  Further, as shown in FIG. 1, the engine 1 is provided with a supercharger (turbocharger) 5. The turbocharger 5 includes a turbine wheel 52 and a compressor wheel 53 that are connected via a turbine shaft 51. The turbocharger 5 in the present embodiment is a variable nozzle type turbocharger, and a variable nozzle vane mechanism (not shown) is provided on the turbine wheel 52 side.

  The intake pipe 64 is provided with an intercooler 61 for forcibly cooling the intake air whose temperature has been raised by supercharging in the turbocharger 5.

  Further, the engine 1 is provided with an exhaust gas recirculation passage (EGR passage) 8 for appropriately returning a part of the exhaust gas to the intake system 6. The EGR passage 8 is provided with an EGR valve 81 and an EGR cooler 82.

-ECU-
The ECU 100 includes a microcomputer (not shown) composed of a CPU, ROM, RAM, and the like and an input / output circuit. As shown in FIG. 3, the input circuit of the ECU 100 includes a crank position sensor 40, a rail pressure sensor 41, a throttle opening sensor 42, an air flow meter 43, A / F sensors 44a and 44b, exhaust temperature sensors 45a and 45b, a water temperature. A sensor 46, an accelerator opening sensor 47, an intake pressure sensor 48, an intake air temperature sensor 49, an in-cylinder pressure sensor 4A, an outside air temperature sensor 4B, an outside air pressure sensor 4C, and the like are connected. Since the function of each sensor is well-known, description here is abbreviate | omitted.

  On the other hand, the output circuit of the ECU 100 is connected to the supply pump 21, the injector 23, the intake throttle valve 62, the EGR valve 81, the variable nozzle vane mechanism 54 of the turbocharger 5, and the like.

  Then, the ECU 100 executes various controls of the engine 1 based on outputs from the various sensors described above, calculated values obtained by arithmetic expressions using the output values, or various maps stored in the ROM. .

  For example, the ECU 100 executes pilot injection (sub-injection) and main injection (main injection) as fuel injection control of the injector 23. Since the functions of the pilot injection and the main injection are well known, description thereof is omitted here.

  The fuel injection pressure when executing the fuel injection is determined by the internal pressure of the common rail 22. As the common rail internal pressure, generally, the target value of the fuel pressure supplied from the common rail 22 to the injector 23, that is, the target rail pressure, increases as the engine load (engine load) increases and the engine speed (engine speed) increases. It will be expensive.

  In addition to the pilot injection and main injection described above, after injection and post injection are performed as necessary. Since the function of these injections is also well-known, explanation here is omitted.

  Further, the ECU 100 controls the opening degree of the EGR valve 81 according to the operating state of the engine 1 to adjust the exhaust gas recirculation amount (EGR amount) toward the intake manifold 63.

-Outline of combustion mode-
Next, the outline of the combustion mode in the combustion chamber 3 in the engine 1 according to the present embodiment will be described.

  As shown in FIG. 4, the gas sucked into the cylinder includes fresh air sucked from the intake pipe 64 and EGR gas sucked from the EGR passage 8.

  The fresh air and EGR gas sucked into the cylinder in this way are sucked into the cylinder as the piston 13 (not shown in FIG. 4) is lowered through the intake valve 16 which is opened in the intake stroke. It becomes in-cylinder gas. This in-cylinder gas is sealed in the cylinder (inside the combustion chamber 3) by closing the intake valve 16 when the valve is closed according to the operating state of the engine 1 (in-cylinder gas confinement state). In the subsequent compression stroke, the piston 13 is compressed as it rises. When the piston 13 reaches the vicinity of the compression top dead center, the injector 23 is opened for a predetermined time by the injection amount control by the ECU 100 described above, so that fuel is directly injected into the combustion chamber 3 (pilot injection or main injection). Run).

  FIG. 5 is a cross-sectional view showing the combustion chamber 3 and its periphery when the main injection is performed, and FIG. 6 is a plan view of the combustion chamber 3 when the fuel is injected (a view showing the upper surface of the piston 13).

  The fuel sprays A, A,... Injected from the injection holes of the injector 23 diffuse in a substantially conical shape. Further, since the fuel injection from each nozzle hole (the main injection) is performed when the piston 13 reaches the vicinity of the compression top dead center, as shown in FIG. 5, the sprays A, A,. It diffuses in the cavity 13b.

  As described above, the fuel sprays A, A,... Injected from the injection holes formed in the injector 23 are mixed with the in-cylinder gas with the passage of time and become air-fuel mixtures in the cylinders. It diffuses in a conical shape and burns by self-ignition. That is, each of the fuel sprays A, A,... Forms a substantially conical combustion field together with the in-cylinder gas, and combustion is started in each of the combustion fields (eight combustion fields in this embodiment). It will be.

  The energy generated by this combustion is kinetic energy for pushing down the piston 13 toward the bottom dead center (energy serving as engine output), thermal energy for raising the temperature in the combustion chamber 3, cylinder block 11 and cylinder head 15 It becomes the heat energy radiated to the outside (for example, cooling water) through.

  The in-cylinder gas after combustion is discharged to the exhaust port 71 and the exhaust manifold 72 as the piston 13 rises through the exhaust valve 17 that opens in the exhaust stroke, and becomes exhaust gas.

−Calculation of in-cylinder environmental parameters−
When creating an ideal heat release rate waveform used for the combustion state diagnosis described later, it is necessary to define the fuel reaction start timing, reaction rate, and reaction amount. And in order to prescribe | regulate these waveform components (reaction start time, reaction rate, reaction amount), it is necessary to obtain | require the oxygen density in a cylinder, a fuel density, and cylinder temperature.

  Hereinafter, a method for calculating these in-cylinder environment parameters (oxygen density, fuel density, in-cylinder temperature) will be described.

(Oxygen density)
The oxygen density is an index representing the oxygen supply capacity (temporal oxygen supply capacity) for the fuel, whether or not EGR is performed, the EGR amount (including the residual gas amount in the cylinder (so-called internal EGR amount)), It fluctuates according to the altitude of the road that is running. And if this oxygen density changes, it will affect the reaction start time, reaction rate, and reaction amount in each reaction of fuel. That is, the lower the oxygen density, the more the reaction start timing shifts to the retarded side, the reaction rate becomes lower (the reaction becomes slower), and the reaction amount decreases. In particular, the effects of low-temperature oxidation reaction, thermal decomposition reaction, and high-temperature oxidation reaction among fuel reactions appear.

  Even if fuel multi-stage injection (for example, pilot injection and main injection) is performed, the amount of oxygen consumed by the combustion of fuel injected in fuel injection for preheating (pilot injection) is the oxygen in the entire cylinder. It is very small compared to the amount. For this reason, it is assumed here that the oxygen densities of the space in the cavity (hereinafter referred to as the cavity area) and the space outside the cavity (hereinafter referred to as the cavity area) constituting the cylinder space are substantially the same. (For example, even when combustion is performed only in the region outside the cavity, it is assumed that they are substantially the same.) A case will be described in which the oxygen density is obtained for the entire cylinder.

The oxygen density ρo ₂ for the entire cylinder is obtained by the following equation (1) or (2).

Oxygen density ρo ₂ = Oxygen amount in intake air / volume of gap (1)
Oxygen density ρo ₂ = Oxygen amount in intake / stroke volume at start of reaction (2)
Here, the oxygen amount (mass) in the intake air can be calculated from the intake air amount detected by the air flow meter 43, the outside air temperature detected by the outside air temperature sensor 4B, the outside air pressure detected by the outside air pressure sensor 4C, and the like. An example of the clearance volume is a stroke volume (compression end volume) when the piston 13 reaches the compression top dead center. According to this, since the gap volume can be handled as a fixed value, the calculation of the oxygen density ρo ₂ can be simplified, and the reliability thereof is increased. The gap volume used in Formula (1) is not limited to this.

When the oxygen density ρo ₂ is calculated by the above equation (2), in each fuel reaction (vaporization reaction, low temperature oxidation reaction, thermal decomposition reaction, high temperature oxidation reaction by premixed combustion, high temperature oxidation reaction by diffusion combustion) The reaction start time, reaction rate, and reaction amount vary depending on the oxygen density ρo ₂ at the time of the reaction. For this reason, in order to obtain the reaction start time, reaction rate, and reaction amount in each reaction, it is necessary to individually specify the oxygen density ρo ₂ at the time of the reaction. As will be described later, the phase difference between the light reaction and heavy reaction of the fuel and the split ratio of the reaction amount also change according to the oxygen density ρo ₂ during the fuel reaction.

In this embodiment, the calculation timing of the oxygen density ρo ₂ corresponding to each reaction of the fuel is set, and the oxygen density ρo ₂ corresponding to each reaction is specified individually using the stroke volume at this timing. I can do it.

Note that the oxygen density ρo ₂ at a predetermined timing set in advance may be obtained, and the oxygen density ρo ₂ at the start of each reaction may be individually specified by calculating backward from the oxygen density ρo ₂ .

In calculating the oxygen density ρo _2, which one of the formulas (1) and (2) is adopted is appropriately selected in consideration of simplification of arithmetic processing and high reliability of the oxygen density ρo _2. The

(Fuel density)
Next, a method for obtaining the fuel density in the cylinder will be described.

  Here, in order to facilitate understanding, description will be made on the assumption that the fuel densities in the cavity inner region and the outer cavity region are substantially the same.

  The fuel density ρfuel is calculated by the following formula (3) or formula (4).

Fuel density ρfuel = fuel injection amount / stroke volume at the start of reaction (3)
Fuel density ρfuel = fuel injection amount / stroke volume at the start of fuel injection (4)
Here, the fuel injection amount is a fuel amount injected from the injector 23 (for example, a fuel amount in main injection). This fuel injection amount can be calculated from the fuel injection pressure detected by the rail pressure sensor 41 and the valve opening period (command injection period) of the injector 23. The stroke volume at the start of the reaction is the in-cylinder volume (the sum of the volume in the cavity area and the volume in the outside cavity area) when the in-cylinder temperature reaches the reaction temperature described later. The relationship between the in-cylinder temperature and the in-cylinder volume is based on the outside air temperature detected by the outside air temperature sensor 4B, the compression ratio, the amount of preheating in the cylinder (the amount of preheating by pilot injection, etc.), etc. as parameters. It is prescribed. The reaction start timing, reaction rate, and reaction amount in each reaction of fuel vary according to the fuel density ρfuel at the time of the reaction. For this reason, in order to obtain the reaction start time, reaction rate, and reaction amount in each reaction, it is necessary to individually specify the fuel density ρfuel at the time of the reaction. Note that the phase difference between the reaction of the light component of the fuel and the reaction of the heavy component and the split ratio of the reaction amount may change depending on the fuel density ρfuel at the time of the fuel reaction.

  In the present embodiment, the calculation timing of the fuel density ρfuel corresponding to each reaction of fuel is set, and the fuel density ρfuel corresponding to each reaction can be individually specified using the stroke volume at this timing. I have to. Alternatively, the fuel density ρfuel at a predetermined timing set in advance may be obtained, and the fuel density ρfuel at the time of each reaction may be individually specified by calculating backward from the fuel density ρfuel. The stroke volume at the start of fuel injection is the in-cylinder volume at the time when fuel injection from the injector 23 is started (when the fuel injection command signal is transmitted from the ECU 100). Since the in-cylinder volume is determined according to the crank angle position, the in-cylinder volume can be obtained based on the crank angle position at the time when fuel injection from the injector 23 is started.

  Furthermore, if the time point when the piston 13 reaches the compression top dead center (TDC) is set as the calculation timing of the fuel density ρfuel, the volume in the cylinder is determined in advance, so that the stroke volume is easily specified. For this reason, the calculation of the fuel density ρfuel can be simplified, and the reliability thereof is increased.

  In calculating the fuel density ρfuel, which one of the formulas (3) and (4) is adopted is appropriately selected in consideration of the simplification of the arithmetic processing and the high reliability of the fuel density ρfuel.

(Cylinder temperature)
As a method for obtaining the temperature in the cylinder, parameters such as the intake air temperature, piston position (intake gas compression degree), preheating state by the pilot injection, etc. are used as parameters, and these parameters and the temperature in the cylinder are determined in advance through experiments and simulations. And the map is stored, and this map is stored in the ROM. That is, the temperature in the cylinder is obtained by fitting parameters such as the intake air temperature, piston position, preheat state, etc. to the map.

  Further, the in-cylinder temperature may be calculated from the thermal energy equation Q = mcT. Here, Q is the heat energy input into the cylinder, m is the mass of the gas in the cylinder, c is the specific heat of the gas, and T is the temperature in the cylinder.

−Creation of heat release rate waveform, combustion state diagnosis, and control parameter correction−
Next, according to the creation of the heat release rate waveform (creation of the ideal heat release rate waveform), the combustion state diagnosis (diagnosis of each reaction mode of the fuel in the cylinder), and the diagnosis result, which are the features of this embodiment The control parameter correction executed in this way will be described.

  In this heat generation rate waveform creation, combustion state diagnosis, and control parameter correction, as shown in FIG. 7, (1) creation of an ideal heat release rate waveform, and (2) creation of an actual heat release rate waveform (3) The combustion state diagnosis is performed by comparing the ideal heat generation rate waveform and the actual heat generation rate waveform. (4) The control parameters of the engine 1 are corrected according to the result of the combustion state diagnosis. All of the configurations for performing the operations (1) to (4) may be mounted (implemented) on the vehicle, or only the operation (1) is performed by a laboratory or the like, and the result (creation) The ideal heat generation rate waveform) is stored in the ROM, and a configuration for performing the operations (2) to (4) may be mounted on the vehicle.

  The feature of this embodiment is that, when creating an ideal heat generation rate waveform for the entire cylinder, the “property of fuel”, “oxygen density” and “fuel density” in the cylinder are used. Specifically, the degree of influence of the “fuel property” on the shape of the ideal heat release rate waveform is changed according to the “oxygen density” and “fuel density” in the cylinder.

  Specific methods for creating the ideal heat release rate waveform include (1-A) separation of fuel reaction forms, (1-B) creation of ideal heat release rate waveform models for each of the separated reaction forms, ( 1-C) Creation of an ideal heat generation rate waveform by filtering (filtering) of an ideal heat generation rate waveform model is performed in order.

  In the (3) combustion state diagnosis, the combustion state diagnosis is performed by comparing the ideal heat generation rate waveform and the actual heat generation rate waveform for the entire cylinder.

  Each operation will be specifically described below.

(1) Creation of ideal heat release rate waveform The creation of the ideal heat release rate waveform will be described. First, an outline of creating an ideal heat generation rate waveform will be described.

  The rate-limiting conditions for the reaction (chemical reaction, etc.) of the fuel injected into the cylinder from the injector 23 include the cylinder temperature, the cylinder oxygen amount (a value correlated with the cylinder oxygen density), and the cylinder fuel amount ( And a fuel distribution in a cylinder). Among these, the order of the low degree of control freedom is the order of the in-cylinder temperature, the in-cylinder oxygen amount, the in-cylinder fuel amount, and the in-cylinder fuel distribution.

  That is, the in-cylinder temperature is substantially determined by the intake air temperature and the compression ratio of the engine 1 before the fuel reacts, and the degree of freedom of control is the lowest. In addition, when the fuel injection is performed in advance (for example, when the fuel injection for preheating is performed), the in-cylinder temperature also varies depending on the amount of preheating due to the combustion of the fuel. Further, since the in-cylinder oxygen amount (oxygen density) can be adjusted by the opening degree of the intake throttle valve 62 and the opening degree of the EGR valve 81, the degree of freedom of control is higher than the in-cylinder temperature. Further, the in-cylinder oxygen amount also varies depending on the supercharging rate by the turbocharger 5. The in-cylinder fuel amount can be adjusted by controlling the fuel injection pressure (common rail pressure) by the supply pump 21 and by controlling the injection period of each of the multistage injections of fuel from the injector 23. The degree of freedom of control is high. In addition, since the in-cylinder fuel distribution can be adjusted by controlling the fuel injection pressure and the injection period of each of the multistage injections of the fuel, the degree of freedom in control is high.

  In the present embodiment, on the condition that the warm-up operation of the engine 1 is completed and the outside air temperature is equal to or higher than a predetermined temperature (for example, 0 ° C.), the reaction state of the fuel in the order of the low degree of freedom of control. The priority of the condition for determining the is set high. Here, the quantitative conditions of the in-cylinder temperature, the in-cylinder oxygen amount, and the in-cylinder fuel amount are given higher priority than the in-cylinder fuel distribution. That is, the start timing (reaction start timing) of each reaction of the fuel is determined using the in-cylinder temperature as an axis. That is, the reference temperature arrival angle (crank angle position at the reaction start timing of each reaction mode) is determined from the in-cylinder temperature (compressed gas temperature in the cylinder). In the present embodiment, when determining the start time of each reaction, the start time is corrected according to the oxygen density or the like.

  The oxygen density is an index representing the ability of supplying oxygen to the fuel. When oxygen supply is insufficient, the oxygen density is a rate-limiting condition for combustion. The fuel density is an index representing the reaction heat supply capacity for the unburned region, and is a rate-limiting condition for combustion when fuel supply is insufficient.

  Based on this reaction start time (reaction start time corrected by oxygen density or the like) as a base point, reference values for the reaction rate and reaction amount are obtained, respectively, and the reaction rate and reaction amount are determined based on the oxygen density and fuel density. A correction amount with respect to the reference value is obtained, and a correction by this correction amount is performed to obtain a reaction rate, a reaction amount, and a reaction period, and an ideal heat release rate waveform model is created for each reaction form.

  The feature of this embodiment is that when creating the ideal heat generation rate waveform model in which the heat generation rate waveform of each reaction of the fuel is approximated by an isosceles triangle, When creating multiple ideal heat release rate waveform models due to their properties, the phase difference, reaction gradient, and reaction split ratio of each ideal heat release rate waveform model are changed according to the oxygen density and fuel density (spray distribution). It is in the point which is trying to do.

  Specifically, for fuel, the reaction start time is relatively early (the reaction starts on the advance side) and the reaction rate is relatively high, and the reaction start time is later (retard) than this light component. The reaction starts on the side) and a heavy component having a low reaction rate. For this reason, for a single reaction, an ideal heat generation rate waveform model based on a light reaction and an ideal heat generation rate waveform model based on a heavy reaction can be created. Hereinafter, the ideal heat release rate waveform model corresponding to the light reaction is called the light ideal heat release waveform model, and the ideal heat release waveform model corresponding to the heavy reaction is the heavy ideal heat release waveform. It will be called a model.

  In the low-temperature oxidation reaction and thermal decomposition reaction among the fuel reactions, the light component ideal heat generation rate waveform model and the heavy component ideal heat generation rate waveform model are affected by the oxygen density. For this reason, when creating these ideal heat release rate waveform models, the phase difference, reaction gradient, and reaction split ratio of each ideal heat release rate waveform model are changed by the influence of the oxygen density (the fuel density is It does not affect).

  On the other hand, in the high temperature oxidation reaction by premixed combustion and the high temperature oxidation reaction by diffusion combustion, the light component ideal heat release rate waveform model and the heavy component ideal heat release rate waveform model are both oxygen density and fuel density. to be influenced. For this reason, when creating these ideal heat release rate waveform models, the phase difference, reaction gradient, and reaction split ratio of each ideal heat release rate waveform model are changed by the influence of the oxygen density and fuel density. Note that the influence of the fuel density is smaller than the influence of the oxygen density. For this reason, in the following description, it is assumed that there is almost no influence of the fuel density on the phase difference and the reaction splitting ratio, and this influence of the fuel density mainly affects the reaction gradient.

  FIG. 8 shows the presence or absence of the influence of oxygen density and fuel density on these low-temperature oxidation reaction, thermal decomposition reaction, high-temperature oxidation reaction by premixed combustion, and high-temperature oxidation reaction by diffusion combustion. In the figure, “◯” indicates a reaction that is affected when the density is low, and “X” indicates a reaction that is not affected even when the density is low.

  In the low-temperature oxidation reaction and thermal decomposition reaction, as the oxygen density decreases, the phase difference of each ideal heat generation rate waveform model (light component ideal heat generation rate waveform model and heavy component ideal heat generation rate waveform model) is As the reaction gradient increases and the reaction split ratio decreases, the reaction split ratio increases the amount of reaction in the heavy ideal heat generation rate waveform model. The ratio of the reaction amount in the light quantity and the reaction amount in the ideal heat generation rate waveform model for the light component becomes smaller. For example, the ratio of the reaction amount in the light component ideal heat release rate waveform model to the reaction amount in the heavy component ideal heat release rate waveform model approaches “3: 2” from the state of “5: 0”. To go. These values are not limited to this.

  Also, in the high-temperature oxidation reaction by premixed combustion and the high-temperature oxidation reaction by diffusion combustion, the phase difference of each ideal heat release rate waveform model increases as the oxygen density decreases, and the reaction gradient decreases together. The reaction split ratio increases the reaction amount in the heavy component ideal heat release rate waveform model, so that the reaction amount in the heavy component ideal heat release rate waveform model and the light component ideal heat release rate waveform model The ratio with the reaction amount becomes smaller. In these high temperature oxidation reactions, the reaction gradient of each ideal heat generation rate waveform model decreases as the fuel density decreases.

  In the high-temperature oxidation reaction by the thermal decomposition reaction and the premixed combustion, when an ignition delay due to the in-cylinder environment occurs, this ignition delay affects the reaction split ratio. For example, if an ignition delay occurs in the light component of the fuel, the amount of reaction in the light component ideal heat release rate waveform model increases, so that the light component ideal for the reaction amount in the heavy component ideal heat release rate waveform model The ratio of reaction amount in the heat release rate waveform model is increased. The cause of the ignition delay includes a condition in the temperature-controlled field (delay in rising of the in-cylinder temperature) and a fuel injection pressure (when the fuel injection pressure is low).

  In this way, the phase difference, reaction gradient, reaction split ratio, etc. of each ideal heat release rate waveform model in each of the plurality of reaction modes of fuel injected into the cylinder are calculated according to the oxygen density and fuel density in the cylinder. Thus, an ideal heat release rate waveform model for each reaction is created.

As an operation for creating the ideal heat release rate waveform model, specifically, the reference reaction rate efficiency [J / CA ² / mm ³ ] corresponding to the in-cylinder gas temperature (reference temperature), the fuel composition, etc. at the reaction start time, and The reference reaction amount efficiency [J / mm ³ ] is determined for each reaction mode, and the reference reaction rate efficiency and the reference reaction amount efficiency are corrected based on at least one of the oxygen density and the fuel density in the cylinder. The reaction rate and reaction amount are determined from the reaction rate efficiency and reaction amount efficiency. Further, the reaction speed is corrected according to the engine speed described later. The “reaction rate efficiency” is also referred to as “reaction rate gradient”, and the “reaction amount efficiency” is also referred to as “combustion efficiency”. Hereinafter, “reaction rate efficiency” will be described as “reaction rate gradient”.

  Then, the reaction period is determined from the ideal heat release rate waveform model (triangle model) created from the reaction start time, reaction rate, and reaction amount. This reaction period is determined by the following equation (5).

Reaction period = 2 × (reaction amount / reaction rate) ^1/2 (5)
Details of creation of the ideal heat release rate waveform model (triangle model) will be described later.

(1-A) Separation of Fuel Reaction Form Next, separation of fuel reaction form, which is the first procedure for creating the ideal heat release rate waveform, will be described.

  When fuel injection is performed from the injector 23, a vaporization reaction, a low temperature oxidation reaction, a thermal decomposition reaction, a high temperature oxidation reaction by premixed combustion, and a high temperature oxidation reaction by diffusion combustion are performed in the cylinder according to the in-cylinder environment. Is called. Hereinafter, each reaction mode will be described.

(A) Vaporization reaction In the vaporization reaction, the fuel injected from the injector 23 is vaporized by receiving heat in the cylinder. This reaction is generally a spray-controlled reaction that starts when the fuel spray spreads to some extent in a state where the fuel is exposed to an environment where the gas temperature in the cylinder is 500K or higher. .

  The boiling point of light oil used in the diesel engine 1 is generally 453K to 623K, and the practical range where fuel injection is performed in the cylinder (for example, the timing when the pilot injection is performed) is BTDC (before compression top dead center). ) 40 ° CA. Since the in-cylinder gas temperature at this timing generally rises to about 550K to 600K (other than in cold regions), it is not necessary to consider the temperature-limiting condition in this vaporization reaction.

The reference reaction amount efficiency in this vaporization reaction is, for example, −1.14 [J / mm ³ ].

  In addition, the effective injection amount in this vaporization reaction (the amount of fuel that contributes to the vaporization reaction) is determined from the fuel injection amount to the amount of wall surface adhesion and the amount of unburned floating fuel (the fuel that exists outside the spray mass and does not contribute to the reaction) Is the amount obtained by subtracting. Hereinafter, these fuel amounts are referred to as unburned fuel amounts. These unburned fuel amounts can be obtained experimentally according to the injection amount (which has a correlation with fuel penetration force) and the injection timing (which has a correlation with cylinder pressure).

  And as reaction amount in the said vaporization reaction, it calculates | requires by the following formula | equation (6).

Reaction amount in vaporization reaction = −1.14 × effective injection amount (6)
Since this vaporization reaction is an endothermic reaction, the reaction amount (generated heat amount) is a negative value. This vaporization reaction is hardly affected by the oxygen density because the amount of oxygen required for the reaction is small.

(B) Low-temperature oxidation reaction In the low-temperature oxidation reaction, a low-temperature oxidation reaction component (such as a fuel having a linear single bond composition such as n-cetane (C ₁₆ H ₃₄ )) contained in the diesel oil that is the fuel of the diesel engine 1 is burned. It is a reaction to. This low-temperature oxidation reaction component is a component that can be ignited even when the temperature in the cylinder is relatively low, and as the amount of n-cetane and the like increases (the higher the cetane fuel), The low temperature oxidation reaction easily proceeds, and the ignition delay is suppressed. Specifically, a low temperature oxidation reaction component such as n-cetane generally starts combustion (low temperature oxidation reaction) when the in-cylinder temperature reaches about 750K. Note that fuel components (high temperature oxidation reaction components) other than n-cetane do not start combustion (high temperature oxidation reaction) until the in-cylinder temperature reaches about 900K.

The reference reaction rate gradient (reference reaction rate efficiency) in this low-temperature oxidation reaction is, for example, 4.0 [J / CA ² / mm ³ ]. The standard reaction amount efficiency is, for example, 5.0 [J / mm ³ ].

The reaction rate and reaction amount of the low-temperature oxidation reaction are calculated based on the reaction rate gradient and reaction amount efficiency obtained by correcting the reference reaction rate gradient and reference reaction amount efficiency with the oxygen density ( For example, it is calculated by multiplying the effective injection amount). Further, in calculating the reaction rate of the low-temperature oxidation reaction, a coefficient (rotation speed correction coefficient = (reference value) corresponding to the value obtained by multiplying the reaction speed gradient by the effective injection amount (reference reaction speed). Rotation speed / actual rotation speed) ² ) is multiplied. An arbitrary rotation speed (for example, 2000 rpm) can be set as the reference rotation speed for obtaining the rotation speed correction coefficient. Thereby, even if a gas composition etc. change, reaction rate can be calculated | required as a value depending on time.

The rotational speed correction coefficient may be obtained from the rotational speed correction coefficient map shown in FIG. The rotation speed correction coefficient map shown in FIG. 9 is obtained by setting the reference rotation speed to 2000 rpm. In a region where the actual rotational speed of the engine 1 is equal to or higher than the reference rotational speed (2000 rpm), a value corresponding to “(reference rotational speed / actual rotational speed) ² ” (a value corresponding to the engine rotational speed indicated by a one-dot chain line in the figure). ) As the rotation speed correction coefficient. On the other hand, in a region where the actual rotational speed of the engine 1 is less than the reference rotational speed (2000 rpm), the value corresponding to “(reference rotational speed / actual rotational speed) ² ” is corrected by a predetermined ratio (to the lower side). The corrected value is obtained as the rotation speed correction coefficient (see the solid line in the region below the reference rotation speed). In this case, the correction ratio is obtained by experiment or simulation.

  The reference rotation speed is not limited to the above-described value, and is preferably set to a rotation speed range where the engine 1 is used most frequently.

  Since this low-temperature oxidation reaction is an exothermic reaction, the reaction amount (generated heat amount) is a positive value.

(C) Thermal decomposition reaction The thermal decomposition reaction is a reaction that performs thermal decomposition of a fuel component. In general, the reaction temperature is about 800K.

The reference reaction rate gradient in this thermal decomposition reaction is, for example, −0.2 [J / CA ² / mm ³ ]. The standard reaction amount efficiency is, for example, 5.0 [J / mm ³ ].

  Further, the reaction rate and reaction amount of the thermal decomposition reaction are also calculated based on the reaction rate gradient and reaction amount efficiency obtained by correcting the reference reaction rate gradient and reference reaction amount efficiency with the oxygen density ( For example, it is calculated by multiplying the effective injection amount). Further, even when calculating the reaction rate of the thermal decomposition reaction, a value (reference reaction rate) obtained by multiplying the reaction rate gradient by an effective injection amount is multiplied by the rotation rate correction coefficient corresponding to the engine rotation rate. .

  In the present embodiment, this thermal decomposition reaction is treated as an endothermic reaction. That is, the reaction amount (generated heat amount) is a negative value.

(D) High-temperature oxidation reaction by premixed combustion The reaction temperature of the high-temperature oxidation reaction by premixed combustion is generally about 900K. That is, the reaction that starts combustion when the in-cylinder temperature reaches 900K is the high-temperature oxidation reaction by this premixed combustion.

Further, the reference reaction rate gradient in the high temperature oxidation reaction by this premixed combustion is, for example, 4.3 [J / CA ² / mm ³ ]. The standard reaction amount efficiency is, for example, 30.0 [J / mm ³ ].

  Further, the reaction rate and reaction amount of the high temperature oxidation reaction by this premixed combustion are also based on the reaction rate gradient and reaction amount efficiency obtained by correcting the reference reaction rate gradient and reference reaction amount efficiency by the oxygen density. Calculated (for example, calculated by multiplying the effective injection amount). Further, in the high temperature oxidation reaction by this premixed combustion, the reaction rate gradient (reaction rate gradient set based on the oxygen density) is corrected by the fuel density, thereby correcting the reaction rate gradient (based on the fuel density). The reaction rate gradient set in the above is calculated. Further, even when calculating the reaction rate of the high-temperature oxidation reaction by the premixed combustion, the rotation speed correction according to the engine rotation speed with respect to the value obtained by multiplying the corrected reaction speed gradient by the effective injection amount (reference reaction speed). The coefficient is multiplied.

  Since the high temperature oxidation reaction by this premixed combustion is an exothermic reaction, the reaction amount (generated heat amount) is a positive value.

(E) High temperature oxidation reaction by diffusion combustion The reaction temperature of the high temperature oxidation reaction by diffusion combustion is generally about 1000K. That is, the reaction in which the fuel injected into the cylinder having a temperature of 1000 K or more starts combustion immediately after the injection is a high-temperature oxidation reaction by this diffusion combustion.

  Moreover, the reaction rate in the high temperature oxidation reaction by this diffusion combustion changes according to the common rail pressure, and is obtained from the following equations (7) and (8).

GrdB = A × common rail pressure + B (7)
Grd = GrdB × (reference engine speed / actual engine speed) ²
X (d / reference d) x (N / reference N) (8)
GrdB: reference reaction rate, Grd: reaction rate, d: nozzle hole diameter of the injector 23, N: number of nozzle holes of the injector 23, A, B: constants obtained by experiments, etc. Multiplying the ratio of the actual nozzle hole diameter to the reference nozzle hole diameter and the ratio of the actual nozzle hole number to the reference nozzle hole number of the injector 23 is a generalized equation. Further, the equation (8) is obtained by multiplying the rotation speed correction coefficient, thereby obtaining a response speed corrected in accordance with the engine rotation speed.

  In calculating the reaction rate of the high-temperature oxidation reaction by diffusion combustion, a corrected reaction rate gradient (reaction rate gradient set based on the fuel density) is calculated by correcting the reference reaction rate gradient based on the fuel density. The reaction rate may be calculated based on this corrected reaction rate gradient.

The standard reaction amount efficiency of the high temperature oxidation reaction by diffusion combustion is, for example, 30.0 [J / mm ³ ]. The reaction amount of the high temperature oxidation reaction by diffusion combustion is also the above standard reaction amount efficiency. It is calculated on the basis of the reaction amount efficiency obtained by correcting by the oxygen density or the fuel density (for example, it is calculated by multiplying the effective injection amount).

  Since the high temperature oxidation reaction by diffusion combustion is also an exothermic reaction, the reaction amount (generated heat amount) is a positive value.

  As described above, the reaction form of the fuel can be separated.

(F) Influence of Oxygen Density and Fuel Density on Each Reaction As described above, the oxygen density depends on the reaction start timing, reaction rate, reaction amount, heavy component ideal heat release rate waveform model and light component ideal in each reaction of fuel. It affects the phase difference from the heat generation rate waveform model and the ratio (division ratio) between the reaction amount in the heavy ideal heat generation rate waveform model and the reaction amount in the light ideal heat generation rate waveform model. Fuel density also affects the reaction rate in each reaction of fuel.

  Hereinafter, the influence of the oxygen density and the fuel density on each reaction will be specifically described.

<Reaction start time>
As described above, the reaction start timing shifts to the retard side as the oxygen density decreases. The reaction start time in this case is calculated by the following formula (9).

Reaction start time = reference temperature arrival time + oxygen density decrease correction retardation amount (9)
Here, the reference temperature reaching time is about 750 K for the low temperature oxidation reaction, about 800 K for the thermal decomposition reaction, about 900 K for the high temperature oxidation reaction by premixed combustion, and about 1000 K for the high temperature oxidation reaction by diffusion combustion, as described above. It is time to reach the temperature (crank angle position).

  The oxygen density decrease correction retardation amount is a correction amount for the reaction start time due to the influence of the oxygen density. In setting the oxygen density reduction correction retardation amount, a correction retardation amount map previously obtained by experiment or simulation is stored in the ROM, and the oxygen density reduction correction delay amount is stored from the correction retardation amount map. The angular amount is extracted.

  FIG. 10 shows an example of a corrected retardation amount map for a certain reaction (for example, a low temperature oxidation reaction). Similar maps are stored in the ROM for other reactions (thermal decomposition reaction and high temperature oxidation reaction).

  This correction retardation amount map is obtained by simplifying the change in the retardation amount of the reaction start timing (oxygen density decrease correction retardation amount) with respect to the change in oxygen density by the Wiebe function.

  In the case shown in FIG. 10, when the oxygen density changes from ρ1 to ρ2, the retardation amount when the oxygen density is ρ1 is CA1, and the retardation amount when the oxygen density is ρ2 is “0”. The terms a and m, which are the shape parameters of the Wiebe function, are set.

  The ignition delay temperature may be obtained, and the corresponding crank angle (reaction start time) may be calculated. That is, the relationship between the oxygen density and the ignition delay temperature is defined in advance by experiments and simulations, and the ignition delay temperature obtained from the oxygen density is added to the reference temperature, and when the corrected temperature after this addition is reached. Is calculated as a reaction start time.

  The correction retardation amount map shown in FIG. 10 described above represents the change in the oxygen density reduction correction retardation amount with respect to the change in the oxygen density by a Wiebe function. However, the present invention is not limited to this, and a correction retardation amount map in which a change in the oxygen density decrease correction retardation amount with respect to a change in oxygen density is expressed by a linear function may be used.

  In addition, when considering the retard amount of the reaction start time due to the influence of the spray distribution (for example, the influence of the spray distribution due to the change of the fuel injection pressure), the retard amount of the reaction start time with respect to the change of the spray distribution is considered. A corrected retardation amount map that prescribes the change is created in advance, and the retardation amount of the reaction start time is obtained according to this corrected retardation amount map.

<Reaction rate gradient>
As described above, the reaction rate decreases as the oxygen density decreases. That is, the reaction rate gradient is reduced. The reaction rate gradient in this case is calculated by the following equation (10).

Reaction rate gradient = (reference reaction rate gradient × gradient correction coefficient) × (2000 / NE) ² (10)
Here, the reference reaction rate gradient is about 40 [J / CA ² / mm ³ ] in the low-temperature oxidation reaction and about −0.2 [J / CA ² / mm ³ ] in the thermal decomposition reaction. NE is the engine speed at the calculation timing of the oxygen density. In this equation (10), the reference rotational speed is set to 2000 rpm, and the reaction speed gradient at the calculation timing of the oxygen density is obtained.

  The gradient correction coefficient is a correction amount of the reaction rate gradient due to the influence of the oxygen density. In setting the gradient correction coefficient, a gradient correction coefficient map previously obtained by experiment or simulation is stored in the ROM, and the gradient correction coefficient is extracted from the gradient correction coefficient map.

  FIG. 11 shows an example of a gradient correction coefficient map for a certain reaction (for example, a high temperature oxidation reaction by premixed combustion). This gradient correction coefficient map is obtained by simplifying the change of the gradient correction coefficient with respect to the change of the oxygen density by using the Wiebe function.

  In the case shown in FIG. 11, when the oxygen density changes from ρ3 to ρ4, the gradient correction coefficient when the oxygen density is ρ3 is set to “0”, and the gradient correction coefficient when the oxygen density is ρ4. The terms a and m, which are the shape parameters of the Wiebe function, are set to be “1”.

  Fuel density also affects the reaction rate. That is, the lower the fuel density, the lower the reaction rate. That is, the reaction rate gradient is reduced.

  In the present embodiment, the final reaction rate gradient (the corrected reaction rate gradient) is obtained by correcting the reaction rate gradient calculated by the equation (10) according to the fuel density (multiplying the gradient correction coefficient). I am doing so.

  The relationship between the fuel density and the reaction rate gradient correction amount is stored in the ROM in a fuel density gradient correction coefficient map that has been obtained in advance by experiments and simulations, and is corrected from this fuel density gradient correction coefficient map. Coefficients are extracted.

  FIG. 12 shows an example of a gradient correction coefficient map (fuel density gradient correction coefficient map) for a certain reaction (for example, high temperature oxidation reaction by premixed combustion). This gradient correction coefficient map is obtained by simplifying the change of the gradient correction coefficient (fuel density gradient correction coefficient) with respect to the change of the fuel density using the Wiebe function.

  In the case shown in FIG. 12, when the fuel density changes from ρ5 to ρ6, the gradient correction coefficient when the fuel density is ρ5 is set to “0”, and the gradient correction coefficient when the oxygen density is ρ6. The terms a and m, which are the shape parameters of the Wiebe function, are set to be “1”.

  The gradient correction coefficient maps shown in FIGS. 11 and 12 described above represent the change of the gradient correction coefficient with respect to the change of the oxygen density and the change of the gradient correction coefficient with respect to the change of the fuel density, respectively, by the Wiebe function. However, the present invention is not limited to this, and a gradient correction coefficient map representing a change in gradient correction coefficient with respect to a change in oxygen density and a change in gradient correction coefficient with respect to a change in fuel density may be used.

<Reaction amount>
As described above, the amount of reaction decreases as the oxygen density decreases. The reaction amount efficiency in this case is calculated by the following equation (11).

Reaction amount efficiency = reference reaction amount efficiency × oxygen density correction coefficient (11)
Here, the oxygen density correction coefficient is a correction amount of the reaction amount efficiency due to the influence of the oxygen density. In setting the oxygen density correction coefficient, an oxygen density correction coefficient map previously obtained by experiments and simulations is stored in the ROM, and the oxygen density correction coefficient is extracted from the oxygen density correction coefficient map. . This oxygen density correction coefficient map represents the same tendency as the gradient correction coefficient map described above (FIG. 11). That is, the change of the oxygen density correction coefficient with respect to the change of the oxygen density is expressed by the Wiebe function.

<Phase difference>
As described above, the phase difference of each ideal heat generation rate waveform model (light component ideal heat generation rate waveform model and heavy component ideal heat generation rate waveform model) increases as the oxygen density decreases. In other words, the reaction start timing shifts to the retard side as the oxygen density decreases, but the shift amount is larger in the heavy component ideal heat generation rate waveform model than in the light component ideal heat generation rate waveform model. to be influenced.

  FIG. 13 shows an example of a phase difference map for a certain reaction (for example, low temperature oxidation reaction). This phase difference map is obtained by simplifying the change in the phase difference with respect to the change in the oxygen density by using the Wiebe function.

  In the case shown in FIG. 13, as the oxygen density is lowered, the reaction start of the heavy ideal heat release rate waveform model rather than the amount of shift of the light start ideal heat release rate waveform model to the retard side of the reaction start timing. The terms a and m, which are the shape parameters of the Wiebe function, are set so that the phase difference increases as the amount of shift to the retard side of the timing increases.

  The phase difference map shown in FIG. 13 represents the change of the phase difference with respect to the change of the oxygen density by the Wiebe function. However, the present invention is not limited to this, and a gradient correction coefficient map in which a change in phase difference with respect to a change in oxygen density is expressed by a linear function may be used.

<Reaction split ratio>
As described above, the reaction split ratio of each ideal heat release rate waveform model decreases as the oxygen density decreases. In other words, as the oxygen density decreases, the amount of reaction in the heavy ideal heat release rate waveform model increases, so the reaction amount in the heavy ideal heat release rate waveform model and light ideal heat release rate waveform The ratio with the reaction amount in the model becomes smaller. For example, the ratio of the reaction amount in the light component ideal heat release rate waveform model to the reaction amount in the heavy component ideal heat release rate waveform model approaches “3: 2” from the state of “5: 0”. To go.

  FIG. 14 shows an example of a division ratio map for a certain reaction (for example, a low temperature oxidation reaction). This split ratio map simplifies the change of the split ratio (reaction amount in the light component ideal heat release rate waveform model / reaction amount in the heavy component ideal heat release rate waveform model) with the Wiebe function. Is.

  In the case shown in FIG. 14, the reaction amount in the heavy ideal heat release rate waveform model increases as the oxygen density decreases, and the reaction in the heavy ideal heat release rate waveform model increases. The terms a and m, which are the shape parameters of the Wiebe function, are set so that the ratio between the amount and the reaction amount in the light component ideal heat generation rate waveform model becomes smaller.

  The division ratio map shown in FIG. 14 described above represents the change in the division ratio with respect to the change in the oxygen density by the Wiebe function. However, the present invention is not limited to this, and a division ratio map in which a change in the division ratio with respect to a change in oxygen density is expressed by a linear function may be used.

(1-B) Creation of Ideal Heat Release Rate Waveform Model for Each Separated Reaction Form Next, creation of ideal heat release rate waveform models for each separated reaction form will be described.

  By separating the reaction forms as described above, it is possible to create an ideal heat release rate waveform model in each reaction form. That is, an ideal heat release rate waveform model can be created for each of a vaporization reaction, a low temperature oxidation reaction, a thermal decomposition reaction, a high temperature oxidation reaction by premixed combustion, and a high temperature oxidation reaction by diffusion combustion. Especially for low-temperature oxidation reactions, thermal decomposition reactions, and high-temperature oxidation reactions, the light component ideal heat generation rate waveform model and heavy component ideal heat generation rate waveform model are created for each single reaction. Is possible. More specifically, in the ideal heat generation rate waveform model in the low-temperature oxidation reaction and thermal decomposition reaction, the light component ideal heat generation rate waveform model and the heavy component ideal heat generation rate waveform model when the oxygen density is low. Are separated and each is created separately. In the ideal heat generation rate waveform model for the high temperature oxidation reaction by premixed combustion and the high temperature oxidation reaction by diffusion combustion, the ideal heat generation rate for the light component is obtained regardless of whether the oxygen density is low or the fuel density is low. The waveform model and the heavy component ideal heat release rate waveform model are separated and each is created individually. The reaction gradient, phase difference, and reaction split ratio of the light component ideal heat generation rate waveform model and the heavy component ideal heat generation rate waveform model are obtained from each map and calculation formula as described above.

  In this embodiment, the ideal heat release rate waveform model is approximated to an isosceles triangle for each reaction. That is, based on the reaction start temperature described above, the reaction rate (reaction rate corrected according to the oxygen density) is the slope of the hypotenuse of the isosceles triangle, and the reaction amount (reaction amount corrected according to the oxygen density) is An ideal heat release rate waveform model is created in which the area of an isosceles triangle is used and the reaction period is the length of the base of the isosceles triangle. As the reaction start time, the creation of the following ideal heat release rate waveform model is applied to each of the reaction modes described above. This will be specifically described below.

(A) Reaction rate (reaction rate gradient)
The reaction rate is set based on the reaction rate gradient, and when the ideal heat generation rate waveform model is approximated to an isosceles triangle, the reaction rate during the period in which the heat generation rate increases and the period in which the heat generation rate decreases Their absolute values are consistent with the reaction rate at.

  In addition, when the reaction rate in the period in which the heat generation rate decreases is lower than the reaction rate in the period in which the heat generation rate increases (when the ideal heat generation rate waveform model is an inequilateral triangle), The descending gradient is obtained by multiplying the ascending gradient by a predetermined value α (<1).

  In the ideal heat release rate waveform model in the high temperature oxidation reaction by diffusion combustion, the reaction rate is proportional to the injection rate waveform gradient, and the reaction rate is constant if the fuel injection pressure (common rail internal pressure) is constant. In the ideal heat release rate waveform model in other reactions (for example, high temperature oxidation reaction by premixed combustion), the reaction speed is proportional to the fuel injection amount.

(B) Generated heat (area)
The reaction amount efficiency [J / mm ³ ] in each reaction can be regarded as a constant (for example, 30 J / mm ^{3 in} the case of a high temperature oxidation reaction) if the combustion period is optimized. For this reason, the generated heat amount is obtained by multiplying the reaction amount efficiency by the fuel injection amount (the effective injection amount).

  However, the low-temperature oxidation reaction is completed together with the high-temperature oxidation reaction, and the high-temperature oxidation reaction by diffusion combustion is completed alone.

  The amount of heat generated in this way corresponds to the area of a triangle that is an ideal heat generation rate waveform model.

(C) Combustion period (base)
The combustion period corresponding to the length of the base of the triangle is obtained from the gradient of the triangle (reaction rate) and the area of the triangle (amount of generated heat).

  As shown in FIG. 15, the triangular area (corresponding to the amount of generated heat) is S, the base length (corresponding to the combustion period) is L, and the height (corresponding to the heat generation rate at the peak of the heat generation rate) is H. , A is the period from the start of combustion to the peak of the heat generation rate, and B is the period from the peak of the heat generation rate to the end of combustion (B = A when the ideal heat generation rate waveform model is an isosceles triangle), G is the ascending gradient (corresponding to the reaction rate during the period when the heat generation rate increases), and α (≦ 1) is the ratio of the descending gradient (corresponding to the reaction rate during the period when the heat generation rate decreases) to this ascending gradient. In this case, the following relationship holds. FIG. 15A shows a case where the ideal heat generation rate waveform model is an isosceles triangle, and FIG. 15B shows a case where the ideal heat generation rate waveform model is an unequal triangle.

H = A × G = B × α × G
From this, B = A / α.

S = A ² × G / 2 + A × G × B / 2 = (1 + 1 / α) × A ² × G / 2
Therefore, A = SQRT [2S / {(1 + 1 / α) G}].

Therefore, the base length L is
L = A + B = A (1 + 1 / α)
= (1 + 1 / α) × SQRT [2S / {(1 + 1 / α) G}]
When the ideal heat release rate waveform model is an isosceles triangle, α = 1.
L = 2 × SQRT (S / G) = 2 × SQRT (30 × Fq / G).

(Fq is the fuel injection amount (effective injection amount). As described above, when the heat generation amount per 1 mm ^{3 of} fuel is 30 J, “30 × Fq” is the triangular area S)
In this way, the combustion period is determined when the injection amount (injection amount command value: a value correlated with the generated heat amount) and the gradient (reaction rate) are given.

  Hereinafter, the reason why the ideal heat generation rate waveform model can be approximated to a triangle (especially an isosceles triangle) will be described. FIG. 16A shows the relationship between the elapsed time when fuel is injected from the injector 23 and the amount of fuel supplied into the cylinder in one reaction mode (the amount of fuel used in that reaction mode). ing. In FIG. 16A, the fuel injection period in which the fuel supply amount is obtained is divided into ten periods. That is, the fuel injection period is divided into ten periods in which the fuel supply amounts are equal to each other, and the period numbers from the first period to the tenth period are assigned to each. That is, after the fuel injection in the first period is completed, the fuel injection in the second period is started without interruption, and after the fuel injection in the second period is completed, the fuel injection is interrupted. The fuel injection is continued until the end of the tenth period in such an injection form that the fuel injection in the third period is started without any change.

  FIG. 16B shows the reaction amount of the fuel injected in each period (the one shown in FIG. 16B is the exothermic amount in the exothermic reaction). As shown in FIG. 16B, during the period from when the fuel injection in the first period is started until the fuel injection in the second period is started (period t1 in FIG. 16B), Only the reaction of the fuel injected in the first period is performed. The fuel injected in the first period is the period from the start of fuel injection in the second period until the start of fuel injection in the third period (period t2 in FIG. 16B). And the reaction of the fuel injected in the second period are performed together. In this way, every time a new injection period is reached, the total reaction amount of fuel gradually increases (the total reaction amount increases by the amount of fuel in the period during which new injection is started). This increase period corresponds to the positive gradient period (advanced period relative to the peak position of the reaction) of the ideal heat release rate waveform model.

  Thereafter, the reaction of the fuel injected in the first period ends. At this time (timing T1 in FIG. 16B), the reaction of the fuel injected in the second period and thereafter is not completed, and the reaction of the fuel injected in the tenth period from the second period is completed. continuing. Then, when the reaction of the fuel injected in the second period ends (timing T2 in FIG. 16B), the reaction of the fuel injected in the third period and thereafter has not ended, so the third period The reaction of the fuel injected in the tenth period will continue. In this way, the reaction of the fuel injected in each period ends in sequence, so that the total reaction amount of the fuel gradually decreases (the total reaction amount decreases by the amount of fuel for which the reaction has ended). Go). This decrease period (a period in which the reaction amount is indicated by a broken line in FIG. 16B) is a negative gradient period (a period retarded from the peak position of the reaction) of the ideal heat release rate waveform model. Equivalent to.

  Since the fuel reaction is performed in the above-described manner, the ideal heat generation rate waveform model can be approximated as a triangle (isosceles triangle).

  The above is the procedure for creating the ideal heat release rate waveform model for each reaction mode of fuel.

  As described above, the light component ideal heat release rate waveform model and the heavy component ideal heat release rate waveform model are affected by the oxygen density. And the influence degree of this oxygen density changes with reaction of a fuel. For example, in the low temperature oxidation reaction, the phase difference, reaction gradient, and split ratio of each ideal heat generation rate waveform model are affected as the oxygen density decreases. On the other hand, in the high temperature oxidation reaction, the reaction gradient and the split ratio of each ideal heat generation rate waveform model are affected as the oxygen density decreases.

  FIG. 17 shows a light component ideal heat release rate waveform model (a waveform indicated by a solid line in the figure) and a heavy component ideal heat release rate waveform model (a broken line in the figure) at low oxygen density in each of the low temperature oxidation reaction and the high temperature oxidation reaction. It is a figure for demonstrating the change of (waveform shown by). FIG. 17A shows a light component ideal heat generation rate waveform model and a heavy component ideal heat generation rate waveform model when the oxygen density is sufficiently high. 17 (b) shows a light component ideal heat release rate waveform model and a heavy component ideal heat release rate waveform model of the low-temperature oxidation reaction at low oxygen density. 17 (c) shows a light component ideal heat release rate waveform model and a heavy component ideal heat release rate waveform model of the high-temperature oxidation reaction at low oxygen density. As shown in FIG. 17 (b), in the low temperature oxidation reaction at low oxygen density, a phase difference at the start of the reaction occurs between the light component ideal heat release rate waveform model and the heavy component ideal heat release rate waveform model. The reaction split ratio is reduced. In addition, as shown in FIG. 17 (c), in the high temperature oxidation reaction at low oxygen density, there is no phase difference at the start of the reaction between the light component ideal heat generation rate waveform model and the heavy component ideal heat generation rate waveform model. The reaction split ratio becomes small. Further, the ratio of the reaction periods of the light component ideal heat generation rate waveform model and the heavy component ideal heat generation rate waveform model is, for example, “1: 1.5”. These characteristics are not limited to this. Further, not only the low temperature oxidation reaction and the high temperature oxidation reaction but also the similar characteristics can be mentioned in the thermal decomposition reaction.

(1-C) Creation of ideal heat generation rate waveform model by filtering of ideal heat generation rate waveform model After creating the ideal heat generation rate waveform model as described above, this ideal heat generation rate waveform model is subjected to well-known filter processing ( For example, an ideal heat generation rate waveform is created by performing smoothing by processing using a Wiebe filter. This will be specifically described below.

  FIG. 18 shows an ideal heat release rate waveform model (isosceles triangle corresponding to each reaction) when one fuel injection is performed. In FIG. 18, in order to facilitate understanding of the present invention, an ideal heat generation rate waveform model in the case where a vaporization reaction, a low temperature oxidation reaction, a thermal decomposition reaction, and each high temperature oxidation reaction are sequentially performed by one fuel injection. (Isosceles triangles corresponding to each reaction) are shown. Specifically, in the figure, I is the ideal heat release rate waveform model of the vaporization reaction, II (solid line) is the light component ideal heat release rate waveform model in the low temperature oxidation reaction, and II '(dashed line) is the heavy component in the low temperature oxidation reaction. The ideal heat release rate waveform model, III is the ideal heat release rate waveform model of the pyrolysis reaction (pyrolysis reaction that is endothermic), IV (solid line) is the light ideal heat release rate waveform model in the high temperature oxidation reaction by premixed combustion, IV '(dashed line) is the heavy ideal heat release rate waveform model for high temperature oxidation reaction by premixed combustion, V (solid line) is the light ideal heat release rate waveform model for high temperature oxidation reaction by diffusion combustion, V' (dashed line) Is a heavy ideal heat release rate waveform model of high temperature oxidation reaction by diffusion combustion. In the ideal heat generation rate waveform model shown in FIG. 18, one waveform model indicates that there is no significant difference between the ideal heat generation rate waveform models of light and heavy components in the pyrolysis reaction. In addition, in the high-temperature oxidation reaction by premixed combustion, a phase difference occurs at the start of the reaction of each ideal heat release rate waveform model (for example, a phase difference occurs due to the influence of fuel density). A case where no phase difference occurs at the start of the reaction of the ideal heat generation rate waveform model is shown as an example. The relationship between the ideal heat generation rate waveform models for light and heavy components in each reaction is not limited to that shown in FIG.

  FIG. 19 shows an ideal heat generation rate waveform obtained by smoothing the ideal heat generation rate waveform model by filtering. In this way, the ideal heat release rate waveform model (isosceles triangle) corresponding to each reaction (vaporization reaction, low temperature oxidation reaction, thermal decomposition reaction, each high temperature oxidation reaction) is smoothed by the filter processing, and the ideal heat release rate A waveform will be created.

  In the actual engine 1, pilot injection, after injection, and the like are performed in addition to the main injection. Therefore, for these pilot injections and after-injections, an ideal heat release rate waveform model is created in the cylinder in the same way as described above, and this is smoothed by filtering to create an ideal heat release rate waveform. Is done.

  Then, an ideal heat generation rate waveform for the entire cylinder in the main injection and these ideal heat generation rate waveforms (ideal heat generation rate waveforms for pilot injection and after injection) are combined to make one cycle. An ideal heat generation rate waveform for the target is created.

  Even when the main injection is divided into a plurality of times and executed (divided main injection), the ideal heat generation rate for one cycle is obtained by synthesizing the ideal heat generation rate waveforms in each main injection. A waveform will be created.

  When multiple injections are performed in this way, in synthesizing the ideal heat generation rate waveforms, in-cylinder temperature at the timing of fuel injection at the preceding stage (advance side), and thereafter ( It is necessary to take into account that the in-cylinder temperature at the timing of fuel injection is different from each other (on the retard side). Specifically, in the steady operation state of the engine, when the preheating or the like is not performed at the timing of fuel injection on the advance side, fresh air sucked from the outside, residual gas in the cylinder, and EGR The reaction is started based on the compressed gas temperature resulting from the temperature rise of the gas such as gas as the piston 13 moves. When starting the engine or returning the fuel injection from the fuel cut, the reaction starts based on the compressed gas temperature due to the rise in temperature of the fresh air drawn from the outside as the piston 13 moves. Will be. On the other hand, when the fuel is injected at the retarded angle side, the temperature rises by adding the temperature of burned gas (combustion gas of fuel injected at the advanced angle side) to the compressed gas temperature. Since the fuel is injected with respect to the temperature field, the reaction start timing shifts to the advance side as compared with the case where there is no temperature increase due to the burned gas. Taking this into account, the ideal heat release rate waveform due to the reaction of fuel injected on the advance side and the ideal heat release rate waveform due to the reaction of fuel injected on the retard side are considered the temperature changes described above. Ask. That is, the start time of each reaction in each injection is defined by temperature management. Thereby, it is possible to appropriately obtain the start time of each reaction in each injection. As a result, it is possible to properly define the reaction start order, the period in which the reactions are paralleled, etc., and the ideal heat generation rate waveform by synthesizing the ideal heat generation rate waveform created for each injection Can be created with high accuracy.

(2) Creation of actual heat generation rate waveform The actual heat generation rate waveform to be compared with the ideal heat generation rate waveform is generated according to a change in the in-cylinder pressure detected by the in-cylinder pressure sensor 4A. In other words, there is a correlation between the heat generation rate in the cylinder and the in-cylinder pressure (the higher the heat generation rate, the higher the in-cylinder pressure). Therefore, from the in-cylinder pressure detected by the in-cylinder pressure sensor 4A. An actual heat generation rate waveform can be created. Since the process of creating the actual heat generation rate waveform from the detected in-cylinder pressure is known, the description thereof is omitted here.

(3) Diagnosis of combustion state by comparing ideal heat generation rate waveform and actual heat generation rate waveform As a diagnosis of combustion state (diagnosis of reaction mode), the deviation of the actual heat generation rate waveform from the ideal heat generation rate waveform Done based on size. For example, when there is a reaction form in which the divergence is equal to or greater than a preset threshold value (abnormality judgment divergence amount in the present invention), it is diagnosed that the reaction form is abnormal. Become. For example, when there is a reaction form in which the deviation of the heat generation rate is 10 [J / ° CA] or more, or the deviation of the actual heat generation rate waveform to the crank angle side (advance side or When there is a reaction form in which the delay angle deviation is 3 ° CA or more, it is diagnosed that the reaction form is abnormal. These values are not limited to this, and are set as appropriate through experiments and simulations.

  For example, the case where the ideal heat generation rate waveform shown in FIG. 19 is created will be described as an example. As shown in FIG. 20, the ideal heat generation rate waveform ( The actual heat generation rate waveform in each high-temperature oxidation reaction (high-temperature oxidation reaction by premixed combustion and high-temperature oxidation reaction by diffusion combustion) is shifted to the retard side with respect to the high-temperature oxidation reaction (the waveform shown in FIG. If it exceeds, it is diagnosed that an abnormality has occurred in each high-temperature oxidation reaction, that is, an abnormality has occurred in the reaction start timing of each high-temperature oxidation reaction.

  Moreover, like the actual heat generation rate waveform indicated by the one-dot chain line in FIG. 20, the peak value of the heat generation rate waveform in each high-temperature oxidation reaction is higher than the ideal heat generation rate waveform indicated by the solid line, and the deviation thereof has a threshold value. If it exceeds, it is diagnosed that an abnormality has occurred in each high-temperature oxidation reaction, that is, an abnormality has occurred in the reaction amount in each high-temperature oxidation reaction. Such diagnosis is not limited to the high temperature oxidation reaction, but is similarly performed for the vaporization reaction, the low temperature oxidation reaction, and the thermal decomposition reaction.

  The parameters for diagnosing whether or not an abnormality has occurred in the reaction form are not limited to the above-described reaction time deviation (ignition delay, etc.) or the peak value deviation of the heat release rate waveform, but the reaction rate. Deviation, reaction period deviation, peak phase, and the like.

(4) Correction of control parameter of engine 1 according to diagnosis result In the diagnosis of the combustion state by comparing the ideal heat generation rate waveform and the actual heat generation rate waveform, the actual heat generation rate with respect to the ideal heat generation rate waveform as described above. When there is a reaction form in which the waveform deviation exceeds a preset threshold, it is diagnosed that an abnormality has occurred in the reaction form, and the control parameter of the engine 1 is corrected so as to reduce this deviation. .

  For example, when the actual heat generation rate waveform is shown by a broken line in FIG. 20, it is determined that fuel ignition delay has occurred and oxygen is insufficient, and the intake air cooling capacity by the intercooler 61 is determined. The shortage of oxygen is eliminated by reducing the EGR gas amount by decreasing the opening degree of the EGR valve 81 or by increasing the intake supercharging rate.

  Further, when the actual heat generation rate waveform is shown by a one-dot chain line in FIG. 20, it is determined that the reaction amount of the fuel is too large, and the fuel injection amount reduction correction, the EGR gas increase correction, etc. I do.

  As other corrective action, when the reaction start time in the actual heat generation rate waveform is on the retard side with respect to the ideal heat generation rate waveform, the intake supercharging rate is increased or the pilot in the cylinder is piloted. It is also possible to make corrections such as increasing the amount of preheating by injection.

  Control parameters for bringing the actual heat generation rate waveform closer to the ideal heat generation rate waveform are not limited to those described above, but include fuel injection timing, gas composition in the cylinder, intake air amount (gas amount), and various learning values. (A learned value of the fuel injection amount or fuel injection timing) may be used. For example, when the oxygen density in the cylinder is excessive or insufficient, learning is performed so as to correct the EGR gas or the intake supercharging rate as the learning value. Further, when the fuel density in the cylinder is excessive or insufficient, learning is performed such that the fuel injection timing, the fuel injection pressure, and the fuel injection amount are corrected as the learning value.

  Such correction of the control parameter is executed when the actual heat generation rate waveform can be substantially matched with the ideal heat generation rate waveform by the correction of the control parameter. Specifically, it is executed when the deviation amount of the actual heat generation rate waveform with respect to the ideal heat generation rate waveform is equal to or less than a predetermined correctable deviation amount. This correctable deviation amount is set in advance by experiment or simulation. When the deviation amount of the actual heat generation rate waveform with respect to the ideal heat generation rate waveform exceeds the correctable deviation amount, the correction amount of the control parameter exceeds the predetermined guard value. Diagnose that a failure has occurred in a part of the devices constituting the system 1. Specifically, when the lower limit values of the in-cylinder temperature, oxygen density, and fuel density are set in advance, and any of these in-cylinder temperature, oxygen density, and fuel density is lower than the lower limit value, If the correction amount of the control parameter of the engine 1 exceeds a predetermined guard value, it is diagnosed that the engine 1 has failed.

  In this case, without correcting the control parameter, for example, the MIL (warning lamp) on the meter panel in the passenger compartment is lit to urge the driver to warn, and the diagnosis provided in the ECU 100 has abnormality information. Will be written.

  As described above, in this embodiment, when the shape of the ideal heat generation rate waveform changes under the influence of the properties of the fuel, such as the light and heavy components of the fuel, the ideal heat generation according to the change The shape of the rate waveform can be defined, and an appropriate ideal heat generation rate waveform can be created. For this reason, it becomes possible to obtain high reliability in the created ideal heat generation rate waveform. In view of the fact that the influence of the fuel properties on the ideal heat release rate waveform changes according to the oxygen density and fuel density, which are environmental parameters in the cylinder, the influence of the fuel properties on the shape of the ideal heat release rate waveform The shape of the ideal heat release rate waveform is defined by changing according to the oxygen density and fuel density in the cylinder. This makes it possible to obtain even higher reliability in the created ideal heat generation rate waveform.

  Further, in this embodiment, an ideal heat generation rate waveform for the entire cylinder is created by filtering the ideal heat generation rate waveform model created for each of the plurality of reaction forms of the fuel. The combustion state is diagnosed using the ideal heat release rate waveform. For this reason, when the actual heat generation rate waveform deviates from the ideal heat generation rate waveform by a predetermined amount or more for each of the plurality of fuel reaction forms, it is possible to diagnose that the reaction form is abnormal. it can. That is, each reaction form can be handled individually and the presence or absence of abnormality can be diagnosed for each. For this reason, it is possible to specify the reaction form in which an abnormality has occurred with high accuracy, and to improve the diagnostic accuracy. And by taking improvement measures (correction of control parameters) for the reaction form diagnosed as abnormal (when the divergence is equal to or less than a predetermined correctable deviation amount), the reaction state of the reaction form is optimized. Therefore, it is possible to correct the optimal control parameter for performing the correction, and an effective correction operation can be performed. As a result, it becomes possible to bring the entire reaction of the fuel closer to the ideal reaction (the actual heat generation rate waveform of each reaction approaches the ideal heat generation rate waveform), and the controllability of the engine 1 is greatly improved. be able to.

  Further, when it is diagnosed that an abnormality has occurred in the reaction, whether or not the abnormality can be resolved is determined based on a deviation amount of the actual heat generation rate waveform from the ideal heat generation rate waveform. Therefore, it is possible to accurately determine a state in which a normal reaction state is obtained by correcting the control parameter and a state in which maintenance such as component replacement is necessary.

-Other embodiments-
In the embodiment described above, the case where the present invention is applied to the in-line four-cylinder diesel engine 1 mounted on an automobile has been described. The present invention is applicable not only to automobiles but also to engines used for other purposes. Further, the number of cylinders and the engine type (separate type engine, V-type engine, horizontally opposed engine, etc.) are not particularly limited. Further, the present invention is not limited to diesel engines using light oil as fuel, but can also be applied to engines using gasoline or other fuels.

  In the above embodiment, the combustion state diagnosis apparatus according to the present invention is stored in the ROM of the in-vehicle ECU 100 (mounted on the vehicle), and the combustion state is diagnosed in the operating state of the engine 1. The present invention is not limited to this, and the combustion state diagnosis device is provided in an experimental device (engine bench tester), and the combustion state is diagnosed when the engine 1 is tested on the experimental device in the design stage of the engine 1. It is also possible to apply to a usage pattern in which an appropriate value of the control parameter is obtained by performing the above.

  In the above embodiment, an ideal heat generation rate waveform model for the entire cylinder is created on the assumption that the oxygen density and the fuel density in the outer cavity region and the inner cavity region are substantially the same. The ideal heat release rate waveform was created by filtering the rate waveform model. The present invention is not limited to this, and it is assumed that at least one of the oxygen density and the fuel density in the region outside the cavity and the region in the cavity is different, and an ideal heat release rate waveform model is created for each region. An ideal heat generation rate waveform for the entire cylinder may be created by filtering the heat generation rate waveform model and synthesizing each ideal heat generation rate waveform.

  In the embodiment, when creating the ideal heat generation rate waveform model, two ideal heat generation rate waveform models (light component ideal heat generation rate waveform model and heavy component ideal heat generation rate waveform model are based on the fuel properties and the like. ). The present invention is not limited to this, and three or more ideal heat generation rate waveform models may be defined based on fuel properties and the like, and ideal heat generation rate waveform models may be created from these ideal heat generation rate waveform models.

  Furthermore, the usage form of the created ideal heat generation rate waveform is not limited to the diagnosis of the combustion state, but may be an engine design and a value for adapting control parameters.

  In addition, in the above-described embodiment, the engine 1 to which the piezo injector 23 that changes the fuel injection rate by being in a fully opened valve state only during the energization period is described. However, the present invention applies a variable injection rate injector. It can also be applied to the engine.

  INDUSTRIAL APPLICABILITY The present invention is applicable to creation of a heat release rate waveform for each reaction of fuel and diagnosis of each reaction in a diesel engine mounted on an automobile.

1 engine (internal combustion engine)
12 Cylinder bore 13 Piston 13b Cavity 23 Injector (fuel injection valve)
3 Combustion chamber 4A In-cylinder pressure sensor 100 ECU
I Waveform model of ideal heat release rate for vaporization reaction
II Light-weight ideal heat release rate waveform model of low-temperature oxidation reaction
II 'Waveform model of ideal heat release rate for heavy components of low-temperature oxidation reaction
III Ideal heat release rate waveform model of pyrolysis reaction
IV Light-weight ideal heat release rate waveform model of high-temperature oxidation reaction by premixed combustion
IV 'Heavy-weight ideal heat release waveform model V of high-temperature oxidation reaction by premixed combustion V Light-weight ideal heat release waveform model V of high-temperature oxidation reaction by diffusion combustion V' Heavy-weight ideal heat generation of high-temperature oxidation reaction by diffusion combustion Rate waveform model

Claims (6)

An apparatus for creating a heat release rate waveform of the fuel reaction in an internal combustion engine that burns fuel injected into a cylinder from a fuel injection valve,
When creating the ideal heat generation rate waveform of the reaction of the fuel injected from the fuel injection valve by defining the shape of the ideal heat generation rate waveform based on the fuel properties ,
Of the low-temperature oxidation reaction, thermal decomposition reaction, and high-temperature oxidation reaction that are the reactions of the fuel, the low-temperature oxidation reaction and the thermal decomposition reaction are performed according to only the oxygen density in the cylinder and the oxygen density in the cylinder. The ideal heat generation rate waveform of the minute and the ideal heat generation rate waveform of the heavy component, the lower the oxygen density in the cylinder, the larger the phase difference of these ideal heat generation rate waveforms, and the reaction amount of the heavy component While reducing the ratio of the reaction amount of the light component to the light, and reducing the reaction gradient of these ideal heat generation rate waveforms to define the shape of the ideal heat generation rate waveform,
For the high-temperature oxidation reaction, the oxygen in the cylinder is compared with the ideal heat generation rate waveform for light and heavy ideal heat generation rates for heavy fuel according to both the oxygen density and fuel density in the cylinder. The lower the density, the larger the phase difference of these ideal heat generation rate waveforms, the smaller the ratio of light reaction amount to the heavy reaction amount, and the smaller the reaction gradient of these ideal heat generation rate waveforms, Further, the heat generation rate waveform generating apparatus for an internal combustion engine is characterized in that the lower the fuel density is, the smaller the reaction gradient of these ideal heat generation rate waveforms is to define the shape of the ideal heat generation rate waveform. . In the internal combustion engine heat release rate waveform creation device according to claim 1,
The ideal heat generation rate waveform is an ideal heat generation rate waveform model composed of triangles with the reaction rate as the base point, the reaction rate as the slope, the reaction amount as the area, and the reaction period as the base length. An apparatus for creating a heat release rate waveform of an internal combustion engine, which is created by smoothing an ideal heat release rate waveform model of each reaction by filtering. In the internal combustion engine heat release rate waveform creation device according to claim 2,
The ideal heat generation rate waveform model has a light component ideal heat generation rate waveform model corresponding to the reaction of the light component of the fuel and a heavy component ideal heat generation rate waveform model corresponding to the reaction of the heavy component of the fuel. ,
The ideal heat generation rate waveform of the internal combustion engine is created by smoothing the light component ideal heat generation rate waveform model and the heavy component ideal heat generation rate waveform model by filtering. Creation device. Comparing the ideal heat generation rate waveform obtained by the heat generation rate waveform generating device for an internal combustion engine according to claim 1, 2 or 3, and the actual heat generation rate waveform when the fuel actually reacts in the cylinder; Combustion of an internal combustion engine characterized by diagnosing that an abnormality has occurred in the fuel reaction when the deviation of the actual heat generation rate waveform from the ideal heat generation rate waveform is a predetermined amount or more Condition diagnosis device. The combustion state diagnosis apparatus for an internal combustion engine according to claim 4,
When there is a reaction in which the deviation of the actual heat generation rate waveform from the ideal heat generation rate waveform is greater than or equal to a predetermined abnormality determination deviation amount, and it is diagnosed that the reaction is abnormal, When the deviation of the actual heat generation rate waveform with respect to the generation rate waveform is equal to or less than a predetermined correctable deviation amount, the control parameter of the internal combustion engine is corrected to control the deviation to be less than the abnormality determination deviation amount. A combustion state diagnosis of an internal combustion engine characterized by diagnosing that a failure has occurred in the internal combustion engine when the deviation of the actual heat generation rate waveform from the ideal heat generation rate waveform exceeds the correctable deviation amount apparatus. The combustion state diagnosis apparatus for an internal combustion engine according to claim 4 or 5,
A combustion state diagnostic device for an internal combustion engine, which is mounted on a vehicle or mounted on an experimental device.

Images