JP2005054753A - Fuel injection control device for internal combustion engine - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a fuel injection control device for an internal combustion engine capable of optimally controlling the injection timing of main injection for surly preventing the misfire of main injection. <P>SOLUTION: This fuel injection control device for the internal combustion engine performing early stage injection before main injection is provided with a cylinder pressure sensor 29 detecting pressure in a combustion chamber and a heat release parameter calculation means calculating a heat release parameter relating to an in-cylinder heat release quantity generated in the combustion chamber by using actual pressure in the combustion chamber detected by the cylinder pressure sensor and engine crank angles. The injection timing of main injection is controlled for preventing the misfire of main injection based on a heat release parameter value calculated by the heat release parameter calculation means at timing before the start of main injection and after the completion of combustion corresponding to early stage injection. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a fuel injection control device for an internal combustion engine, and more particularly to a fuel injection control device that optimizes combustion in a diesel engine.

  Due to recent demands for exhaust gas regulations and noise reduction, there is an increasing demand for optimization of combustion in the combustion chamber of diesel engines. In order to optimize combustion, it is necessary to accurately control the injection amount, the injection timing, the injection period, and the like even in a diesel engine.

  However, in a diesel engine, the intake air amount is generally not adjusted and the engine load is controlled by the injection amount. Accordingly, in a diesel engine, combustion is performed in a lean air-fuel ratio region that is considerably higher than the stoichiometric air-fuel ratio, and the air-fuel ratio changes according to the load. For this reason, in conventional diesel engines, the air-fuel ratio is not strictly controlled as in gasoline engines, and conventionally, fuel injection characteristics such as injection amount and injection timing are also controlled as precisely as gasoline engines. Is not done. Conventionally, in a diesel engine, target values of fuel injection characteristic values such as injection amount, injection timing, injection pressure, etc. are determined from engine operating conditions (rotation speed, accelerator opening, etc.), and a fuel injection valve is determined according to the target values. However, with open-loop control, it is difficult to prevent the actual injection amount from causing an error with respect to the target injection amount, and it is difficult to accurately control the combustion state to the target state. there were.

Furthermore, in order to improve exhaust gas properties and reduce noise, multi-injection is effective in which fuel injection is performed multiple times before and after main injection during one cycle of each cylinder to optimally adjust the combustion state. . However, in order to perform multi-injection, it is necessary to precisely control each injection amount and injection timing of a plurality of fuel injections.
In addition, common rail type high-pressure fuel injection devices that have recently been adopted in diesel engines to improve combustion conditions have a short injection time and the injection pressure changes during injection. There are problems that are likely to occur. For this reason, in common rail high pressure fuel injection devices, measures such as setting the tolerance of the fuel injection valve to be small and improving the injection accuracy are taken. Since the fuel injection characteristic changes, it is difficult to always make the fuel injection characteristic value coincide with the target value accurately by performing the open loop control.

In this way, in a diesel engine, an error is likely to occur in the injection amount and the like, and even if a target value for obtaining an optimal combustion state can be set, it is difficult to actually match the injection amount with the target value.
In order to match the combustion state with the target combustion state, the actual combustion state is detected in some way, and fuel injection such as the injection amount and injection timing is made so that the actual combustion state matches the target combustion state It is effective to feedback control the characteristic value.

As described above, an example of a combustion control device for an internal combustion engine that detects the combustion state and feedback-controls the fuel injection characteristic value is disclosed in Patent Document 1.
The apparatus of Patent Document 1 detects the pressure in the combustion chamber by an in-cylinder pressure sensor, and detects the ignition timing of the fuel injected based on the detected pressure in the combustion chamber, or the timing when the pressure in the combustion chamber becomes maximum. The injection timing is controlled so that these detected timings coincide with the target timing. The target time is set based on the engine speed and the engine load (accelerator opening).

  In other words, the apparatus of Patent Document 1 attempts to feedback control the injection timing based on the actually measured pressure in the combustion chamber so as to match the actual combustion state with the target combustion state.

JP 09-68081 A JP 2002-276442 A Japanese Patent Laid-Open No. 11-257142

By the way, in many diesel engines, in order to accelerate the reaction of the main fuel injected by the main injection and shorten the ignition delay period of the main injection, in addition to the main injection, an auxiliary during the intake stroke of the internal combustion engine Intake during fuel injection for injecting fuel and pilot injection (hereinafter collectively referred to as “early injection”) for injecting fuel auxiliary during the compression stroke are performed.
The acceleration of the reaction of the main injection by the early injection occurs mainly due to the rise of the in-cylinder temperature by the early injection, and therefore depends on the in-cylinder heat generated by the combustion of the fuel injected by the early injection. Conversely, if the fuel by the main injection is not ignited even if the early injection is performed (hereinafter referred to as “misfire”), the cause is that the in-cylinder heat generation amount by the early injection is small.

On the other hand, whether or not misfire related to main injection occurs depends on the injection timing of main injection. That is, if the injection timing of the main injection is retarded too much, the fuel is injected after the in-cylinder temperature is lowered, and the fuel is not combusted.
From the above, in order to prevent the occurrence of misfire related to the main injection, it is necessary to appropriately adjust the in-cylinder heat generation amount due to the early injection and the injection timing of the main injection.

  In the apparatus of Patent Document 1 described above, it is not assumed that early injection is performed. Accordingly, the injection timing of the main injection is controlled based on the engine speed and the engine load, and naturally the in-cylinder heat generation amount due to the early injection is not a parameter for setting the injection timing of the main injection. For this reason, in the apparatus of Patent Document 1, in the internal combustion engine in which early injection is performed, the injection timing of the main injection is optimally controlled so as to prevent the occurrence of misfire related to the main injection according to the amount of heat generated in the cylinder by the early injection Can not do it.

  Therefore, the present invention provides a fuel injection control device for an internal combustion engine that can optimally control the injection timing of the main injection so as to reliably prevent the occurrence of misfire related to the main injection when early injection is performed. The purpose is to provide.

In order to solve the above-described problem, in the first invention, a fuel injection control for an internal combustion engine that includes a fuel injection valve that injects combustion into an engine combustion chamber and performs early injection before main injection in the vicinity of compression top dead center In the apparatus, the in-cylinder pressure sensor for detecting the pressure in the combustion chamber, and the heat generation parameter for calculating the heat generation parameter relating to the in-cylinder heat generated in the combustion chamber using the actual combustion chamber pressure and the engine crank angle detected by the cylinder pressure sensor. Parameter calculation means, and after the end of combustion corresponding to the early injection and before the start of main injection, the main injection based on the value of the heat generation parameter calculated by the heat generation parameter calculation means The injection timing of the main injection is controlled so that no misfire occurs.
According to the first aspect, since the heat generation parameter is calculated after the combustion of the fuel injected by the early injection is completed, the calculated heat generation parameter is the in-cylinder heat generation amount generated as a result of the combustion of the fuel injected by the early injection. In addition, since the heat generation parameter is calculated before the main injection is started, the calculated heat generation parameter is a value affected only by the early injection. Therefore, the calculated value of the heat generation parameter is a value corresponding only to the in-cylinder heat generation amount by the early injection.
Further, as described above, whether or not misfiring occurs with respect to main injection varies depending on the in-cylinder heat generation amount due to early injection and the injection timing of main injection. In other words, if the amount of heat generated in the cylinder by early injection is known, it is possible to know the injection timing of the main injection so that misfire does not occur with respect to the main injection. In the first invention, since the injection timing of the main injection is controlled based on the calculated value of the heat generation parameter, it is possible to reliably prevent the occurrence of misfire related to the main injection.
Note that “early injection” means injection performed before main injection in the same cycle. For example, in-intake injection that injects fuel supplementarily during the intake stroke of the internal combustion engine and auxiliary during the compression stroke And pilot injection for injecting fuel. Further, “misfire” means that the injected fuel does not ignite even though the fuel is injected.

According to a second invention, in the first invention, it is possible to execute main injection delay control for retarding the injection timing of the main injection after performing the early injection based on the engine operating state. During execution of the retard control, the misfire limit timing, which is the most retarded timing within the range in which no misfire occurs with respect to the main injection, based on the value of the heat generation parameter calculated by the heat generation parameter calculation means after the main injection. Or a limit timing calculating means for calculating a combustion deterioration limit timing which is the most retarded timing within a range in which the combustion deterioration of the main injection caused by retarding the main injection does not occur, and the injection of the main injection The timing is controlled so as to be the misfire limit time or the combustion deterioration limit time calculated by the limit time calculation means, or the advance side of these limit times.
When the main injection retardation control is performed, the energy converted into work for the piston among the thermal energy obtained by the combustion is reduced, so that the temperature of the exhaust gas can be raised. In order to increase the temperature rise effect of the exhaust gas, it is necessary to delay the injection timing of the main injection as much as possible and reduce the work on the piston as much as possible. On the other hand, as described above, if the injection timing of the main injection is retarded too much, misfire occurs with respect to the main injection.
According to the second aspect of the invention, the main injection is performed at the misfire limit time, which is the most retarded time within the range where no misfire occurs with respect to the main injection, or at the more advanced time. Alternatively, the main injection is performed at a time immediately before the range in which the misfire does not actually occur but the misfire does not actually occur and the combustion deterioration of the main injection occurs, or at a timing that is more advanced than that. For this reason, the temperature rise effect of the exhaust gas can be maximized while preventing the occurrence of misfire or the deterioration of the combustion of the main injection by retarding the injection timing of the main injection.

In a third aspect of the invention, in the first or second aspect, the heat generation parameter includes a product PV of a combustion chamber pressure P detected by the in-cylinder pressure sensor and a combustion chamber volume V determined from a crank angle θ, and a combustion chamber volume V. It is a difference ΔPV between the product PVbase of the pressure in the combustion chamber generated only by the compression of the piston and the combustion chamber volume determined from the crank angle when it is assumed that no combustion has occurred.
In the third invention, ΔPV represents the in-cylinder heat generated by the combustion in the combustion chamber. That is, the value PV of the product of pressure and volume is a value corresponding to the energy of the in-cylinder gas, and the change amount of PV per unit time is the energy applied to the in-cylinder gas, that is, the compression work due to the rise of the piston. And the amount of heat generated by combustion. Here, the amount of change in PV due to the compression work of the piston is calculated as the product PVbase of the pressure and the volume when it is assumed that combustion has not occurred in the combustion chamber, and thus the pressure in the combustion chamber detected by the in-cylinder pressure sensor. The difference ΔPV between the product PV of PV and the combustion chamber volume V determined from the crank angle θ and PVbase represents the amount of heat generated by combustion in the cylinder.

According to a fourth aspect, in the third aspect, after the value of the rate of change d (ΔPV) / dθ with respect to the crank angle θ of ΔPV exceeds a first predetermined value of zero or more during the compression stroke of the internal combustion engine. The timing that is equal to or smaller than the first predetermined value and smaller than the second predetermined value near zero is adopted as the timing after the end of combustion corresponding to the early injection and before the start of main injection.
As described above, ΔPV represents the amount of heat generated in the cylinder due to combustion, and the rate of change d (ΔPV) / dθ of ΔPV with respect to the crank angle θ is the amount of heat generated in the cylinder per unit crank angle due to combustion, that is, heat generated by combustion. It represents the incidence. Therefore, when the rate of change d (ΔPV) / dθ takes a positive value, combustion occurs in the combustion chamber, and therefore the rate of change d (ΔPV) / ΔPV with respect to the crank angle θ in the compression stroke of the internal combustion engine. When dθ takes a positive value for the first time, it is considered that the fuel injected by the early injection is burning. Therefore, the time when the rate of change d (PV) / dθ first takes a positive value and then returns to zero in the compression stroke of the internal combustion engine is considered to be the time when the combustion corresponding to the early injection is finished.
Thus, by calculating the value of ΔPV at the time when the combustion corresponding to the early injection is completed, the in-cylinder heat generation amount due to the combustion corresponding to the early injection can be accurately obtained.
However, the value of ΔPV represents the amount of heat generated in the cylinder when the combustion is completed, but decreases due to heat radiation (heat loss) from the wall surface of the combustion chamber. Therefore, after the combustion of the fuel injected by the early injection is completed, the value of ΔPV gradually decreases with the crank angle. Therefore, in the fourth invention, it is determined that the early injection is completed when the value of the rate of change d (ΔPV) / dθ is smaller than a second predetermined value that is a value near zero.
Further, when the combustion ΔPV is not in the combustion chamber, the value of the change rate d (ΔPV) / dθ is zero, but the above calculation is performed even if combustion does not occur due to a detection error of the in-cylinder pressure sensor. The value of change rate d (ΔPV) / dθ may not be equal to zero and may be blurred. Therefore, in the fourth aspect of the invention, when the rate of change d (ΔPV) / dθ exceeds the first predetermined value slightly larger than zero in consideration of such fluctuations, the combustion of the fuel injected by the early injection is performed. Is determined to have started.

ただし、実際には燃焼室壁面等からの燃焼室内の熱が奪われて熱損失が発生している。この熱損失分を考慮すると、早期噴射によって増大したＰＶ^κの値と、燃焼室内で燃焼が生じなかったと仮定した場合にＰＶ^κが到達していると考えられるＰＶ^κの値（ＰＶ^κｂａｓｅ）との差分、すなわちΔＰＶ^κが燃焼室内における燃焼によって発生した筒内発熱量を近似的に表していると考えられる。なお、κはポリトロープ指数である。 According to a fifth aspect of the invention, in the first or second aspect of the invention, the exothermic parameter is a value of the combustion chamber pressure P detected by the in-cylinder pressure sensor and a predetermined constant κ power of the combustion chamber volume V determined from the crank angle θ. a product PV ^kappa, difference ΔPV and the product PV ^kappa base with multiplication the constant kappa combustion chamber volume determined from the combustion chamber pressure and the crank angle caused by compression only of the piston when it is assumed that the combustion in the combustion chamber was no ^kappa It is.
In the fifth invention, the ΔPV ^κ value approximately represents the in-cylinder heat generation amount generated by the combustion in the combustion chamber. That is, the heat release rate dQ / dθ in the internal combustion engine can be expressed as the following formula (1). Here, in the early injection that is performed near the compression top dead center and has a short injection period, V ^1−κ in the equation (1) can be approximated to be constant. Therefore, the following equation (2) can be derived from the equation (1), and the in-cylinder heat generation amount dQ in a certain period can be taken out as a change amount (d (PV ^κ )) of PV ^κ in that period. Therefore, the in-cylinder heat generation amount by early injection can be calculated from the amount of change in PV ^κ before and after combustion by pilot injection.

However, in reality, heat in the combustion chamber is deprived from the wall surface of the combustion chamber and heat loss occurs. In view of this heat loss, the value of PV ^kappa was increased by early injection, the value of PV ^kappa is PV considered to have reached ^kappa in assuming the combustion in the combustion chamber has not occurred (PV ^kappa base) , That is, ΔPV ^κ is considered to approximately represent the in-cylinder heating value generated by the combustion in the combustion chamber. Note that κ is a polytropic index.

In the sixth aspect of the invention, in the fifth invention, during the compression stroke of the internal combustion engine, the value of the change rate d (ΔPV ^κ) / dθ with respect to the crank angle θ of the Pv ^kappa is greater than a first predetermined value or more zero After that, the timing that is equal to or less than the first predetermined value and becomes smaller than the second predetermined value near zero is adopted as the timing after the end of the combustion corresponding to the early injection and before the start of the main injection. .

According to a seventh invention, in the first to sixth inventions, further comprising a combustion state detecting means for detecting a combustion state relating to the main injection, and the misfire in which the injection timing of the main injection is calculated by the limit time calculating means. As a result of controlling to be the limit timing, the combustion deterioration limit timing, or the advance timing side of these limit timings, the combustion state detecting means retards the combustion state or main injection in which misfire related to main injection occurs When the combustion state in which the deterioration of the main injection combustion is detected is detected, the misfire limit timing or the combustion deterioration limit timing calculated by the limit timing calculation means is corrected to the advance side.
The limit time calculation means calculates the misfire limit time or the combustion deterioration limit time from the value of the heat generation parameter based on a predetermined relationship (calculation formula or map obtained experimentally). However, as a result of controlling the injection timing of the main injection to be the misfire limit timing or the combustion deterioration limit timing calculated by the limit timing calculation means, the combustion state in which misfire has occurred or the combustion in which combustion deterioration has occurred If a state is detected, it can be seen that there is an error in the predetermined relationship. The causes of such errors include variations among the fuel injection valve solids, changes over time in the heat radiation characteristics of the combustion chamber, and the like.
Therefore, according to the seventh invention, the misfire limit timing or the combustion deterioration limit timing calculated by the limit timing calculation means in the above-described case is corrected to the advance side. As a result, learning control of the relationship between the value of the heat generation parameter calculated by the heat generation parameter calculation means and the misfire limit time or the combustion deterioration limit time is performed, and misfires related to main injection occur even when this relationship changes. Or deterioration of combustion related to main injection is prevented.

  According to the present invention, it is possible to optimally control the injection timing of the main injection so as to prevent the occurrence of misfire related to the main injection in accordance with the in-cylinder heat generation amount by the early injection.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.
FIG. 1 is a diagram showing a schematic configuration of an embodiment when the fuel injection device of the present invention is applied to an automobile diesel engine.

  In FIG. 1, 1 is an internal combustion engine (in this embodiment, a four-cylinder four-cycle diesel engine having four cylinders # 1 to # 4 is used), 10a to 10d are # 1 to # 4 of the engine 1 A fuel injection valve that directly injects fuel into each cylinder combustion chamber is shown. The fuel injection valves 10a to 10d are each connected to a common pressure accumulation chamber (common rail) 3 via a fuel passage (high pressure fuel pipe). The common rail 3 has a function of storing the pressurized fuel supplied from the high-pressure fuel injection pump 5 and distributing the stored high-pressure fuel to the fuel injection valves 10a to 10d via the high-pressure fuel pipe.

  1 is an electronic control unit (ECU) that controls the engine. The ECU 20 is configured as a microcomputer having a known configuration in which a read only memory (ROM), a random access memory (RAM), a microprocessor (CPU), and an input / output port are connected by a bidirectional bus. In this embodiment, the ECU 20 performs fuel pressure control for controlling the discharge amount of the fuel pump 5 to control the common rail 3 pressure to a target value determined according to the engine operating conditions, and in addition to the fuel pressure according to the engine operating state. Sets the injection timing and injection amount of injection, and sets the injection parameter, injection timing, etc. so that the value of the combustion parameter calculated based on the cylinder pressure sensor output described later matches the target value determined according to the engine operating state Basic control of the engine such as fuel injection control for feedback control is performed.

  In order to perform these controls, in this embodiment, the common rail 3 is provided with a fuel pressure sensor 27 for detecting the fuel pressure in the common rail, and the accelerator pedal position (not shown) in the vicinity of the accelerator pedal of the engine 1 is provided. An accelerator opening sensor 21 that detects (a driver's accelerator pedal depression amount) is provided. In FIG. 1, reference numeral 23 denotes a cam angle sensor that detects the rotational phase of the cam shaft of the engine 1, and reference numeral 25 denotes a crank angle sensor that detects the rotational phase of the crank shaft. The cam angle sensor 23 is disposed near the cam shaft of the engine 1 and outputs a reference pulse every 720 degrees in terms of a crank rotation angle. The crank angle sensor 25 is disposed near the crankshaft of the engine 1 and generates a crank angle pulse at every predetermined crank rotation angle (for example, every 15 degrees).

  The ECU 20 calculates the engine speed from the frequency of the crank rotation angle pulse signal input from the crank angle sensor 25, and based on the accelerator opening signal input from the accelerator opening sensor 21 and the engine speed, the fuel injection valve 10a. To 10d, the injection timing and the injection amount are calculated.

  Further, in FIG. 1, 29a to 29d are known in-cylinder pressure sensors which are arranged in the cylinders 10a to 10d and detect the pressure in the cylinder combustion chamber. Each combustion chamber pressure detected by the in-cylinder pressure sensors 29 a to 29 d is supplied to the ECU 20 via the AD converter 30.

  In the present embodiment, the fuel pressure of the common rail 3 is controlled by the ECU 20 to a pressure corresponding to the engine operating state, and varies in a wide range, for example, at a high pressure of about 10 MPa to 150 MPa.

  By the way, in the present embodiment, a large amount of fuel injection (hereinafter referred to as “main injection”) performed near the compression top dead center and an auxiliary small amount of fuel performed before the main injection in the same cycle as the main injection. Fuel injection (hereinafter referred to as “early injection”) is performed. Furthermore, in the present embodiment, main injection retardation control is performed to retard the injection timing of the main injection after performing early injection according to the engine operating state. Hereinafter, main injection retardation control will be described.

  By the way, for example, when it is necessary to raise the temperature of the exhaust purification catalyst in order to activate the exhaust purification catalyst and improve the exhaust properties, it is conceivable to raise the temperature of the exhaust gas discharged from the internal combustion engine body. In order to raise the temperature of the exhaust gas, it is effective to retard the injection timing of the main injection. That is, when the injection timing of the main injection is retarded, the time taken from the start of the fuel injected in the main injection to the opening of the exhaust valve (not shown) is short. For this reason, the exhaust valve is opened while the energy converted into the work of the piston out of the thermal energy obtained by combustion in the combustion chamber is small, resulting in the exhaust gas exhausted from the combustion chamber. High thermal energy remains, and thus the exhaust gas temperature rises.

  However, if the injection timing of the main injection is retarded, the combustion corresponding to the main injection becomes unstable according to the retard width. That is, when the retard width is increased, an ignition delay occurs, and when the retard width is increased too much, the injected fuel eventually stops igniting (that is, misfire occurs with respect to the main injection).

For this reason, when retarding the injection timing of the main injection, early injection is also performed at the same time. When such early injection is performed, intermediate products such as aldehydes, ketones, peroxides, carbon monoxide, and the like are generated from the auxiliary fuel injected by the early injection due to the heat of compression during the compression stroke. The reaction of the main fuel injected by the main injection is accelerated, and the main fuel is easily ignited.
Further, before the combustion corresponding to the main injection is started, the auxiliary fuel injected by the early injection is ignited by compression, thereby increasing the temperature and pressure in the combustion chamber. Thus, the main fuel is easily ignited by injecting the main fuel into the combustion chamber whose temperature and pressure are increased by the main injection.
For this reason, when retarding the injection timing of the main injection, the retard width of the main injection can be increased by using the early injection at the same time.

  Examples of the early injection include the following two fuel injections. One is a small amount of auxiliary injection (hereinafter referred to as “in-intake during injection”) performed during the intake stroke (for example, in the vicinity of intake top dead center). In the intake air injection, since an intermediate product is generated from the start of the compression stroke, the generation of the intermediate product is further promoted and the reactivity of the main fuel is further increased. However, since this is performed when the temperature and pressure in the combustion chamber are low, if the injection amount is increased, the injected fuel reaches the cylinder wall in a liquid state and causes problems such as dilution of lubricating oil. For this reason, it is not possible to increase the injection amount, and it is also necessary to divide the required injection amount and divide the injection amount into small portions at a plurality of times to prevent liquid fuel from reaching the cylinder wall. is there.

  The other is a time much earlier than the main injection (for example, a time that is 20 degrees (20 ° CA) or more earlier in crank angle than the start of the main injection), and is a small amount of auxiliary injection (hereinafter, referred to as “subject injection”). This is referred to as “pilot injection”. In the pilot injection, since the intermediate product is generated after the injection is performed, the intermediate product is less likely to be generated as compared with the above-described in-intake injection, but at a time when the temperature and pressure in the combustion chamber are high. Since it is performed, the injection amount can be increased and the control of the injection amount is easy.

  By the way, in the main injection retardation control as described above, it is preferable to make the injection timing of the main injection as retarded as possible from the viewpoint of increasing the temperature rise effect of the exhaust gas. However, if the injection timing is retarded too much, even if early injection is performed, combustion deteriorates and eventually misfire occurs. Therefore, in the main injection delay angle control, combustion deterioration is caused by delaying the injection timing too much at the most retarded timing (hereinafter referred to as “misfire limit timing”) within the limit where no misfire occurs. It is effective to perform the main injection at the most retarded timing (hereinafter referred to as “combustion deterioration limit timing”) within a range that does not occur.

  The misfire limit time and the combustion deterioration limit time greatly depend on the amount of heat generated in the cylinder generated by the combustion corresponding to the early injection. That is, the misfire limit time and the combustion deterioration limit time are determined according to the in-cylinder heat generation amount by the early injection. Therefore, if the in-cylinder heat generation amount by early injection can be obtained, the misfire limit time and the combustion deterioration limit time can be obtained.

  Therefore, in the present embodiment, as the in-cylinder heat generation amount, the combustion chamber pressure P detected using the in-cylinder pressure sensors 29a to 29d (hereinafter collectively referred to as “in-cylinder pressure sensor 29”) and the combustion chamber volume V at that time The value of the heat generation parameter (a parameter related to the amount of heat generated in the cylinder generated in the combustion chamber) calculated using is used. Specifically, in this embodiment, ΔPV is used as a heat generation parameter. First, the physical meaning of the heat generation parameter ΔPV will be described.

  FIG. 2 is a diagram showing a change in the heat generation parameter with respect to the crank angle θ from the latter half of the compression stroke to the first half of the expansion stroke when pilot injection is performed as early injection (crank angle θ = 0 is the compression top dead center). Showing). In the drawing, the injection rate of fuel injection is shown, and the area of each mountain indicates the relative fuel injection amount of each fuel injection. As can be seen from the figure, main injection is performed in the vicinity of compression top dead center, and pilot injection is performed prior to this main injection.

In FIG. 2, the pressure P indicates a change in the actual combustion chamber pressure detected by the in-cylinder pressure sensor 29 with respect to the crank angle θ. Further, FIG. 2 shows a change in the heat generation parameter PV, which is obtained by multiplying the combustion chamber pressure P detected by the in-cylinder pressure sensor 29 by the combustion chamber volume V calculated based on the crank angle θ, and the in-cylinder pressure sensor 29. A change with respect to the crank angle θ of the heat generation parameter PV ^κ obtained by multiplying the combustion chamber pressure P detected in step κ by a value obtained by multiplying the combustion chamber volume V calculated based on the crank angle θ by κ is shown. Here, κ is a polytropic index.

  By the way, the energy of the gas in the combustion chamber is expressed by the product PV of pressure and volume, and the amount of change per unit time of this PV value is the energy applied to the cylinder interior gas, that is, the rise of the piston per unit time. It is the sum of the compression work due to and the amount of heat generated in the cylinder due to combustion.

  Here, as shown in FIG. 2, assuming that the product of the pressure in the combustion chamber and the volume of the combustion chamber when it is assumed that combustion has not occurred in the combustion chamber is PVbase, the change in PVbase is caused by the vertical movement of the piston in the cylinder. It represents a change in gas energy. Therefore, the difference ΔPV (= PV−PVbase) between PV and PVbase calculated based on the in-cylinder pressure sensor 29 means an in-cylinder heat generation amount due to combustion.

  Therefore, if the value of ΔPV after completion of combustion of the fuel injected by early injection is calculated, the calculated value of ΔPV becomes a value representing the total in-cylinder heating value generated by the combustion by early injection.

  Thus, in order for the value of ΔPV to represent the total in-cylinder heating value generated by the early injection, the calculation time is important. Below, with reference to FIG. 3, the calculation time of the value of ΔPV in the present embodiment will be described. FIG. 3 shows changes in the heat generation parameters with respect to the crank angle θ in the vicinity of the compression top dead center when early injection is performed.

  As described above, the value of ΔPV represents the in-cylinder heating value generated by the combustion in the combustion chamber. Therefore, in the compression stroke and the expansion stroke in which no gas enters and leaves the combustion chamber, basically, the value in the combustion chamber. ΔPV increases when combustion occurs, and ΔPV does not change and remains constant when combustion does not occur. That is, the value of change rate d (ΔPV) / dθ of ΔPV with respect to crank angle θ (hereinafter referred to as “ΔPV differential value”) is positive when combustion occurs in the combustion chamber, and is zero when combustion does not occur. It becomes.

  However, in reality, the energy (heat) in the cylinder generated by the combustion is released from the wall surface of the combustion chamber or the like. Therefore, when combustion does not occur in the combustion chamber after the completion of combustion corresponding to early injection. The value of ΔPV gradually decreases. Therefore, the ΔPV differential value d (ΔPV) / dθ becomes a negative value.

  Therefore, the ΔPV differential value is almost zero from the start of the compression stroke until the fuel by early injection burns, and when the fuel combustion by early injection starts, the ΔPV differential value takes a positive value. When the combustion of the fuel by the early injection is completed, the ΔPV differential value takes a negative value until the combustion of the fuel by the main injection is started.

  Therefore, in the present embodiment, the time when the ΔPV differential value takes a positive value for the first time from the start of the compression stroke is determined as the time when the combustion of fuel by early injection has started (CAstart). Actually, the ΔPV differential value is slightly less than zero because the detection value of the in-cylinder pressure sensor may take a positive value during the period from the start of the compression stroke until the fuel is burned by early injection. A time when the first predetermined value is exceeded may be determined as the time when combustion starts.

  The time when the ΔPV differential value first takes a value smaller than zero after the start of fuel combustion determination by early injection is determined as the time when fuel combustion by early injection ends (CAend). Note that the determination value may be a non-zero value near zero in consideration of a detection error of the in-cylinder pressure sensor.

  Therefore, in the present embodiment, the calculation time of ΔPV is the time when the ΔPV differential value takes a positive value at the start of the compression stroke and takes a value smaller than zero (the dashed line in FIGS. 2 and 3). reference). Thereby, the value of ΔPV at the time when the combustion of the fuel by the early injection is completed can be calculated, and the value of ΔPV becomes a value corresponding to the in-cylinder heat generation amount by the early injection.

  Note that, unlike the above embodiment, ΔPV may be calculated at the same timing as the main fuel injection start timing. At such a time, the value of ΔPV immediately before combustion corresponding to the main injection can be calculated. ΔPV calculated in this way is a value obtained by subtracting the amount of heat loss from the combustion chamber wall surface during the period between early injection and main injection from the in-cylinder heat generated by early injection. The amount of heat that contributes to the promotion of combustion of the fuel injected by the main injection. However, when the injection interval between the early injection and the main injection is short, or when the injection timing of the main injection is relatively early, a part of the fuel injected by the early injection at the start timing of the main injection is still burned. Therefore, it is effective to set the ΔPV calculation timing as the main injection start timing when the injection interval between the early injection and the main injection is relatively long.

By the way, as described above, in calculating the value of ΔPV, the product PVbase of the pressure in the combustion chamber and the volume of the combustion chamber when combustion does not occur in the combustion chamber is used. PVbase is obtained by approximating the compression and expansion caused by the movement of the piston in the cylinder by a polytropic change of the index κ, that is, by setting PV ^κ = C (constant value), and by early injection after the intake valve is closed. It is also considered that it can be calculated from the pressure in the combustion chamber before combustion of the injected fuel and the combustion chamber volume at this time.

  However, as described above, it is assumed that the polytropic change does not give or release energy other than work due to the compression / expansion of the gas in the cylinder. Since the energy (heat) in the room is deprived and heat loss occurs, even if no combustion occurs in the combustion chamber, the change in the polytropy is not strictly determined. Therefore, it is necessary to calculate the product PVbase of the pressure in the combustion chamber and the volume of the combustion chamber when combustion does not occur in the combustion chamber in consideration of such heat loss.

Therefore, in this embodiment, PVbase is calculated in consideration of heat loss. Hereinafter, the calculation principle of PVbase will be described.
As described above, PVbase can be obtained by subtracting the heat loss from PV calculated by approximating the compression and expansion due to the movement of the piston in the cylinder by the polytropic change of the index κ. In this embodiment, the combustion chamber pressure Pbase when combustion does not occur is calculated by subtracting the heat loss from the combustion chamber pressure calculated by approximating the change in the polytrope, and the crank angle θ is calculated. Multiply the combustion chamber volume V determined by

The combustion chamber pressure P (θ) ref calculated by approximating the change in the polytropy at a specific crank angle θ is calculated based on an in-cylinder pressure sensor at an arbitrary crank angle after the intake valve is closed and before pilot injection. Assuming that the combustion chamber pressure is P (0) and the combustion chamber volume at this time is V (0), the following equation (3) is obtained from the relationship of PV ^κ = C (constant value). Here, V (θ) is a combustion chamber volume determined from the specific crank angle θ.

Further, the decrease dPloss / dθ of the pressure per unit crank angle due to heat loss can be expressed as the following formula (4). Α in equation (4) is a proportionality constant and can be obtained experimentally. Then, the following equation (5) is obtained by integrating the equation (4).

Therefore, the combustion chamber pressure P (θ) ref at a specific crank angle θ approximated by a polytropic change of the index κ can be calculated from the combustion chamber pressure at an arbitrary crank angle after the intake valve is closed and before pilot injection. The heat loss P (θ) loss can be calculated by integrating αP from the above-mentioned arbitrary crank angle to a specific crank angle θ. The value of P (θ) base at a specific crank angle θ is the amount of heat loss calculated as described above from P (θ) ref at a specific crank angle approximated by a polytropic change as shown in Equation (6). Can be obtained by subtracting the decrease P (θ) loss of the pressure corresponding to.

  FIGS. 4 and 5 are flowcharts showing the calculation operation of the value of ΔPV (ΔPVend) at the end of combustion corresponding to the early injection. This operation is executed by the ECU 20 at every constant crank angle.

  3 and 4, at step 101, the current crank angle θ and the combustion chamber pressure P detected by the in-cylinder pressure sensor 29 are read. In step 102, the current combustion chamber volume V is calculated based on the crank angle θ. In the present embodiment, the relationship between θ and V is obtained in advance by calculation, and is stored in the ROM of the ECU 20 as a one-dimensional map using θ. In step 102, the combustion chamber volume V is obtained from this one-dimensional map using the value of θ read in step 101. Further, in step 101, the product PV is calculated by multiplying the calculated current combustion chamber volume V by the combustion chamber pressure P detected in step 101.

Next, in steps 103 to 109, the value of ΔPV is calculated.
In step 103, it is determined whether or not the value of the flag XQ is set to 1. The flag XQ is a flag indicating whether or not an initial value used in equation (7) of step 107 described later has already been acquired, and XQ = 1 indicates that it has already been acquired. If the initial value is acquired in step 103 (XQ = 1), steps 104 to 106 are not executed. On the other hand, if the initial value has not been acquired (ZQ ≠ 1), the process proceeds to step 104.

  In steps 104 to 106, initial values P (0) and V (0) used when calculating P (i) ref in equation (7) of step 107 described later are read. In step 104, it is determined whether or not the intake valve is closed. Whether or not the intake valve is closed is determined based on a command value of the closing timing from the ECU 20 to the intake valve. If it is determined in step 104 that the intake valve is not closed, the subsequent steps are not executed. On the other hand, if it is determined that the intake valve is closed, the routine proceeds to step 105. In step 105, the value of the combustion chamber pressure P read in step 101 is set to an initial value P (0), and the value of the combustion chamber volume V calculated in step 102 is set to an initial value V (0). Thereby, the combustion chamber pressure P and the combustion chamber volume V immediately after it is determined in step 104 that the intake valve is closed are set to the initial values. Next, at step 106, the value of the flag XQ is set to 1.

Next, at step 107, the combustion chamber pressure P (i) base when combustion does not occur in the combustion chamber is calculated by the following equation (7). Here, i represents the number of processes in step 107, and i = 1 when step 107 is executed after the intake valve is closed. Note that ΔP (i−1) refbase in the equation (7) is calculated by the following equation (8). Further, P (i) ref is calculated by the equation (3). Here, the initial values P (0) and V (0) acquired in step 105 are used as the initial values P (0) and V (0) in the equation (3).

In step 108, the combustion chamber pressure P (i) base when combustion does not occur in step 107, the current combustion chamber pressure P read in step 101, and the step 102 are calculated. From the current combustion chamber volume V, PVbase is calculated by the following equation (9).
PVbase = (PP (i) base) · V (9)
Next, at step 109, ΔPV is calculated by subtracting the PV base calculated at step 108 from the PV value calculated at step 102.

  Step 110 represents a calculation operation of the ΔPV differential value. In the present embodiment, the ΔPV differential value d (ΔPV) / dθ is calculated as a difference between the ΔPV value ΔPV (j) calculated this time in Step 109 and the ΔPV value ΔPV (j−1) calculated last time. Is done.

In steps 111 to 114, the time when the combustion of the fuel injected by the early injection is completed is calculated.
In step 111, it is determined whether or not the value of the flag XR is set to 1. The flag XR is a flag indicating whether or not combustion corresponding to early injection has started, and XR = 1 indicates that combustion has started. In step 111, when combustion has started (XR = 1), steps 112 and 113 are not executed, and the routine proceeds to step 114. On the other hand, if combustion has not started, the routine proceeds to step 112.

  In step 112, it is determined whether or not the ΔPV differential value d (ΔPV) / dθ calculated in step 110 is larger than a predetermined value C1. The predetermined value C1 is slightly larger than zero, and the ΔPV differential value d (ΔPV) / dθ is larger than the predetermined value C1 when combustion corresponding to the early injection is started. If it is determined in step 112 that the ΔPV differential value is equal to or smaller than the predetermined value C1, steps 113 to 116 are not executed. On the other hand, if it is determined that the ΔPV differential value is larger than the predetermined value C1, the routine proceeds to step 113. In step 113, the value of the flag XR is set to 1.

  In step 114, it is determined whether or not the ΔPV differential value d (ΔPV) / dθ is smaller than zero. When the ΔPV differential value is smaller than 0, it means that the combustion corresponding to the early injection is finished. If it is determined in step 114 that the ΔPV differential value is 0 or more, steps 115 and 116 are not executed. On the other hand, if it is determined that the ΔPV differential value is smaller than 0, the routine proceeds to step 115.

  In step 115, the value of ΔPV calculated in step 109 is set as a value ΔPVend of ΔPV when combustion corresponding to the early injection is completed. That is, as a result of performing the processing of steps 111 to 114, the case where the processing of step 115 is executed is immediately after the combustion for the early injection is completed. Therefore, the value of ΔPV calculated at this time is the early injection. This is because the value of ΔPV when the corresponding combustion is completed is shown. Next, in step 116, the values of the flags XQ and XR are reset to zero, and the parameter i indicating the number of processes used in step 107 is also reset to zero.

  As described above, by executing the operations of FIGS. 4 and 5, the value of ΔPV at the end of combustion corresponding to the early injection is calculated and stored for each cycle and for each cylinder.

  Next, the injection timing control of the main injection using the value of ΔPV at the combustion end timing corresponding to the early injection calculated as described above will be described. As described above, from the viewpoint of increasing the temperature rise effect of the exhaust gas while preventing the occurrence of misfire, when performing the main injection retarding control, the injection timing of the main injection is set to the misfire limit timing or the combustion deterioration It is preferable to control at the limit time. Therefore, in the present embodiment, the injection timing of the main injection is controlled to the misfire limit timing or the combustion deterioration limit timing. In the following description, the case where the injection timing of the main injection is controlled to the combustion deterioration limit timing will be described. However, the control can be similarly performed when the injection timing is controlled to the misfire limit timing.

  In the present embodiment, during the main injection retardation control, the injection timing of the main injection is determined in advance by using the engine speed, the accelerator opening, and the like by an injection timing setting operation (not shown) separately executed by the ECU 20. Is set to be the combustion deterioration limit time. However, the target injection timing set in this way is not set based on all the parameters related to the in-cylinder heat generation amount due to early injection. Changes in characteristics over time and the like are not taken into consideration, and therefore, the combustion deterioration limit time may not be reached.

  Therefore, in the present embodiment, the injection timing of the main injection is feedback-corrected so that the injection timing of the main injection becomes the combustion deterioration limit timing by using the value ΔPVend of ΔPV at the combustion end timing corresponding to the early injection. FIG. 6 shows a flowchart for explaining the procedure for correcting the injection timing of the main injection in the present embodiment. This operation is executed by the ECU 20 every time after the ΔPV calculation operation shown in FIGS. The procedure for the correction operation will be described below according to this flowchart.

  In step 121, the ECU 20 obtains from the ECU 20 the target injection timing CAtar calculated based on the engine speed and the throttle opening by the injection timing setting operation in the ECU 20 and the ΔPVend calculated by the above-described calculation operations of FIGS. Is done. Next, at step 122, based on the map stored in the ECU 20, the combustion deterioration limit timing CAlim is calculated from the value of ΔPVend acquired at step 121. The relationship between ΔPVend and combustion deterioration limit timing CAlim in the map may be experimentally obtained and stored in advance, or may be calculated by learning control as described later. In the present embodiment, in the above map, the combustion deterioration limit timing CAlim is calculated using only ΔPVend as an argument, but the map argument is not only ΔPVend but also one or more other parameters ( For example, the end timing of combustion corresponding to early injection, engine speed, intake gas temperature, engine coolant temperature, etc.) may be used as arguments.

  In step 123, it is determined whether or not the absolute value | CAlim−CAtar | of the value obtained by subtracting the target injection timing CAtar acquired in step 111 from the combustion deterioration limit timing CAlim calculated in step 122 is greater than a predetermined value C2. The Here, the predetermined value C2 is an absolute value equal to or greater than zero and is a value near zero. Therefore, in step 123, it is determined whether or not the difference between the calculated combustion deterioration timing and the target injection timing is within a certain range. When it is determined in step 123 that the absolute value is equal to or less than the predetermined value, the target injection timing is not changed as it is, and the main injection is injected into the fuel injection valve so that the main injection is performed at the target injection timing. A command is sent.

  On the other hand, if it is determined in step 123 that the absolute value is greater than the predetermined value, the process proceeds to step 124. In step 124, it is determined whether or not the combustion deterioration limit timing CAlim calculated in step 122 is smaller than the target injection timing CAtar acquired in step 111, that is, whether or not the target injection timing is earlier than the combustion deterioration limit timing. Is done. When it is determined that the target injection timing is earlier than the combustion deterioration limit timing (CAlim <CAlim), the routine proceeds to step 125, where a value obtained by adding the predetermined crank angle ΔCA1 to the value of the target injection timing CAtar, that is, the target injection timing. A new target injection timing is set after ΔCA1 of CAtar. The predetermined crank angle ΔCA1 may be a predetermined constant value or a value calculated based on the difference between CAlim and CAtar.

  On the other hand, if it is determined in step 124 that the target injection timing is later than the combustion deterioration limit timing (CAlim> CAlim), the routine proceeds to step 125, where a value obtained by subtracting the predetermined crank angle ΔCA2 from the value of the target injection timing CAtar. That is, a new target injection timing is set by ΔCA2 before the target injection timing CAtar. The predetermined crank angle ΔCA1 may be a predetermined constant value or a value calculated based on the difference between CAlim and CAtar. Then, an injection command for main injection is sent to the fuel injection valve so that main injection is performed at these new target injection timings.

  As described above, by performing the operation of FIG. 6, the injection timing of the main injection is controlled so as to be close to the combustion deterioration limit timing, and the temperature rise effect of the exhaust gas is enhanced while preventing the occurrence of misfire related to the main injection. be able to.

  In the above-described operation for correcting the injection timing of the main injection, the target injection timing is corrected based on the value of ΔPV. However, during the main injection retardation control, the value of ΔPV is used as the target injection timing of the main injection. The combustion deterioration limit time calculated based on the above may be adopted. Thereby, the temperature rise effect of exhaust gas can be maximized while preventing the occurrence of misfire related to the main injection.

  By the way, the relationship between the ΔPV value ΔPVend and the combustion deterioration limit timing CAlim at the end of combustion corresponding to early injection is not always constant. Change. Therefore, when the relationship between ΔPVend and CAlim is experimentally obtained in advance and stored in the ECU 20 as a map, and the relationship is not updated, the injection is performed by the main injection even if the injection timing control of the main injection shown in FIG. 6 is performed. Fuel combustion may continue to deteriorate.

  Therefore, in the present embodiment, if the combustion deterioration of the fuel injected by the main injection occurs as a result of performing the injection timing control of the main injection shown in FIG. 6, it is on the map stored in the ECU 20. Learning control is performed to correct the relationship between ΔPVend and CAlim so that combustion deterioration does not occur. That is, as a result of performing the injection timing control of the main injection, when the deterioration of the combustion corresponding to the main injection is detected, the combustion deterioration limit timing CAlim for ΔPVend is corrected to the advance side.

  This is shown in FIG. FIG. 7 shows the relationship between ΔPVend and CAlim stored as a map. As a result of performing the injection timing control of the main injection based on the relationship between ΔPVend and CAlim shown by the solid line in the figure, combustion deterioration occurs. In this case, it can be seen that CAlim for ΔPVend is corrected to the advance side. In the example shown in FIG. 7, CAlim is corrected to the advance side by the same advance amount for all ΔPVends, but different advance amounts may be corrected according to ΔPVend.

  FIG. 8 is a flowchart for explaining the procedure of the map correction operation used in step 112 of the operation shown in FIG. 6 in the present embodiment. This operation is performed every cycle. First, in step 141, it is determined whether or not the expansion stroke has ended. This is because the expansion stroke needs to be completed when determining the deterioration of combustion described later. If it is determined that the expansion stroke has not ended, Steps 142 to 144 are not executed. On the other hand, if it is determined that the expansion stroke has ended, the routine proceeds to step 142.

  In step 142, it is determined whether or not the combustion deterioration corresponding to the main injection has occurred. A method for determining this combustion deterioration will be described later. If it is determined in step 142 that combustion deterioration has occurred, the process proceeds to step 143. In step 143, the value of CAlim for all ΔPVends is corrected to the advance side by a predetermined value ΔCA3. The predetermined value ΔCA3 is a relatively small angle. On the other hand, if it is determined in step 142 that no deterioration in combustion has occurred, the process proceeds to step 144. In step 144, the value of CAlim for all ΔPVends is corrected to the retard side by a predetermined value ΔCA4. The predetermined value ΔCA4 is an extremely small angle and is smaller than the predetermined value ΔCA3. Step 144 is for executing only Step 143 to prevent the CAlim from being corrected only to the advance side.

  Next, a method for determining whether or not the deterioration of combustion corresponding to the main injection has occurred (hereinafter referred to as “determination method of combustion deterioration”) will be briefly described.

  When the maximum value of PV in each cycle is PVmax (see FIG. 2) and the crank angle when PV becomes PVmax is CApvmax, when the fuel injected by the main injection burns without causing combustion deterioration, These PVmax and CApvmax fall within the predetermined PV range and the predetermined crank angle range, respectively.

  However, when the combustion corresponding to the main injection is deteriorated, PVmax and CApvmax are not within the predetermined PV range and the predetermined crank angle range, PVmax is smaller than the predetermined PV range, and CApvmax is a predetermined crank angle. The time is retarded from the range. Therefore, in the present embodiment, PVmax and CApvmax are calculated, it is determined whether or not these PVmax and CApvmax are within a predetermined PV range or a predetermined crank angle range, respectively, and one or both of these are not within the predetermined range. It is assumed that the combustion corresponding to the main injection has deteriorated.

  In the present embodiment, the relationship between PVend and the combustion deterioration limit time is corrected, but the same can be done when correcting the relationship between PVend and the misfire limit time. However, in this case, regarding the misfire determination, PVmax occurs at the compression top dead center, and therefore CApvmax is a timing on the more advanced side than the predetermined crank angle range.

Next, a second embodiment of the present invention will be described. In the second embodiment, ΔPV ^κ is used as the heat generation parameter instead of ΔPV used in the first embodiment. First, the physical meaning of this ΔPV ^κ will be described.

ここで、圧縮上死点付近において行われ且つ噴射期間の短いパイロット燃料噴射では、Ｖ^1-κの変化が小さく、したがってＶ^1-κを一定と近似することができる。この場合、式（９）を積分して下記式（１０）を導き出すことができる。したがって、ＰＶ^κの値は筒内発熱量に相当する値となっている。

By the way, the in-cylinder heat generation amount per unit crank angle in the internal combustion engine, that is, the heat generation rate dQ can be expressed by the following equation (9).

Here, in the pilot fuel injection that is performed near the compression top dead center and has a short injection period, the change in V ^{1 -κ} is small, and therefore V ^{1 -κ} can be approximated to be constant. In this case, the following formula (10) can be derived by integrating the formula (9). Therefore, the value of PV ^{κ is} a value corresponding to the amount of heat generated in the cylinder.

Therefore, if the product of the multiplication kappa combustion chamber pressure and the combustion chamber volume when it is assumed that the combustion in the combustion chamber has not occurred and PV ^κ base, PV ^κ from PV ^kappa calculated based on the cylinder pressure sensor 29 A value ΔPV ^κ obtained by subtracting base is a value corresponding to the in-cylinder heating value due to combustion (ΔPV ^κ = PV ^κ −PV ^κ base). Therefore, if the ΔPV ^κ value at the pilot injection fuel combustion end timing is calculated, the calculated ΔPV ^κ value becomes a value corresponding to the total in-cylinder heating value generated by the combustion by the pilot fuel injection.

Here, a method for calculating PV ^κ base will be described. As described above, if the compression and expansion due to the movement of the piston in the cylinder can be approximated by a change in the polytropy of the index κ, the PV ^κ value is always constant when it is assumed that no combustion has occurred in the combustion chamber. However, as described above, energy (heat) in the combustion chamber is deprived from the wall surface of the combustion chamber and heat loss occurs, so that the PV ^κ value gradually decreases as the crank angle θ advances. Since this decrease in the PV ^κ value occurs at a constant rate with respect to the crank angle θ, the PV ^κ base can be approximated linearly as shown in FIG.

Therefore, in the present embodiment, the relationship between the PV ^κ value after the intake valve closing and before the pilot fuel injection and the crank angle at this time is acquired at two points having different crank angles. Then, based on the relationship between the PV ^κ value and the crank angle acquired in this way, a linear equation of PV ^κ base as shown in FIG. 2 is obtained by extrapolation.

In general, the change in the PV ^κ value when it is assumed that no combustion has occurred in the combustion chamber is slow, and in the above-described PV ^κ base calculation method, the above-described PV base is used to calculate the heat loss from the combustion chamber wall surface. Since the calculation is based on the actual detection value rather than by the perfect approximation as calculated, the PV ^κ base straight line is assumed to have caused no combustion in the combustion chamber with relatively high accuracy. The change of PV ^{kappa in} the case is shown.

In the PVbase calculation method described above, since the heat loss from the combustion chamber wall surface and the like is calculated by perfect approximation, the PV ^κ base calculation accuracy is higher than the PV base calculation accuracy. For this reason, in some cases, the calculation accuracy of ΔPV ^κ is higher than the calculation accuracy of ΔPV. However, as described above, ΔPV ^κ and the in-cylinder heat generation amount have a relationship as shown in Expression (10) only when pilot fuel injection is performed in the vicinity of the compression top dead center and the injection period is short. Therefore, it is preferable to use this embodiment when such pilot fuel injection is performed.

ΔPV ^κ calculated in this way can be used in the same manner as ΔPV in the first embodiment. Therefore, ΔPV the rate of change with respect to the crank angle θ of ^κ d ^{(ΔPV κ)} / value of dθ (hereinafter referred to as ^"ΔPV κ differential value") is, by early injection the time taken for the first time a positive value from the compression stroke start It is determined that the combustion of fuel has started (CAstart), and the time when the ΔPV ^κ differential value has taken a value smaller than zero for the first time since the start of combustion of fuel by period injection has been determined is the combustion of fuel by early injection Is determined to be the time when (CAend) is completed. Therefore, the calculation time of ΔPV ^{κ is set to} a time when the ΔPV ^κ differential value takes a value smaller than zero for the first time after taking a positive value at the start of the compression stroke.

In the injection timing control of the main injection, the injection of the main injection is performed so that the injection timing of the main injection becomes the combustion deterioration limit timing or the misfire limit timing using the value ΔPV ^κ end of ΔPV ^κ calculated at the above calculation timing. The timing is feedback corrected. Thereby, the temperature rise effect of exhaust gas can be heightened, preventing generation | occurrence | production of the misfire regarding main injection.

In the first embodiment and the second embodiment, the combustion deterioration limit timing or misfire limit timing is calculated based on ΔPV and ΔPV ^κ . However, the combustion deterioration limit timing or misfire limit timing also changes depending on the temperature in the combustion chamber immediately before main injection, that is, thermal energy. Therefore, instead of ΔPV or ΔPV ^κ , the combustion deterioration limit timing or misfire limit timing may be calculated based on PV.

1 is a diagram showing a schematic configuration of an embodiment in which the present invention is applied to a four-cylinder diesel engine for automobiles. It is a figure which shows the change with respect to the crank angle (theta) of the heat generation parameter at the time of performing early injection. It is a figure which shows the change with respect to crank angle (theta) of heat_generation | fever parameter PV, (DELTA) PV, d ((DELTA) PV) / d (theta). It is a part of flowchart which shows the calculation operation of the value of (DELTA) PV in the time when combustion of the fuel injected by the early injection was complete | finished. It is a part of flowchart which shows the calculation operation of the value of (DELTA) PV in the time when combustion of the fuel injected by the early injection was complete | finished. It is a flowchart explaining the procedure of correction | amendment operation of the injection timing of the main injection in this embodiment. The relationship between (DELTA) PVend memorize | stored as a map and CAlim is shown. The flowchart explaining the procedure of map correction operation is shown.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Diesel engine 3 ... Common rail 10a-10d ... In-cylinder fuel injection valve 20 ... Electronic control unit (ECU)
21 ... Accelerator opening sensor 25 ... Crank angle sensors 29a-29d ... In-cylinder pressure sensor

Claims (7)

In a fuel injection control device for an internal combustion engine that includes a fuel injection valve that injects combustion into an engine combustion chamber and performs early injection before main injection in the vicinity of compression top dead center,
In-cylinder pressure sensor for detecting the pressure in the combustion chamber, and heat generation parameter calculation means for calculating a heat generation parameter relating to the amount of heat generated in the cylinder in the combustion chamber using the actual pressure in the combustion chamber detected by the cylinder pressure sensor and the engine crank angle And at the time after the end of the combustion corresponding to the early injection and before the start of the main injection, misfire is caused with respect to the main injection based on the value of the heat generation parameter calculated by the heat generation parameter calculation means. A fuel injection control device for an internal combustion engine that controls the injection timing of the main injection so as not to occur.   It is possible to execute main injection delay control that retards the injection timing of the main injection after performing the early injection based on the engine operation state. During execution of the main injection delay control, the main injection delay control is executed after the main injection. Based on the value of the heat generation parameter calculated by the heat generation parameter calculation means, the main injection by retarding the misfire limit time, which is the most retarded time within the range where no misfire occurs with respect to the main injection, or the main injection Further includes a limit timing calculation means for calculating a combustion deterioration limit timing which is the most retarded timing within a range where no combustion deterioration occurs, and the misfire calculated by the limit timing calculation means for the injection timing of the main injection 2. The fuel injection control device for an internal combustion engine according to claim 1, wherein the control is performed so that the limit timing, the combustion deterioration limit timing, or the advance timing side of these limit timings is reached.   The heat generation parameter is generated only by the product PV of the combustion chamber pressure P detected by the in-cylinder pressure sensor and the combustion chamber volume V determined from the crank angle θ, and compression of the piston when it is assumed that no combustion has occurred in the combustion chamber. 3. The fuel injection control device for an internal combustion engine according to claim 1, wherein the difference ΔPV is a difference ΔPV between a product PVbase of a combustion chamber pressure and a combustion chamber volume determined from a crank angle.   During the compression stroke of the internal combustion engine, the value of the rate of change d (ΔPV) / dθ of the ΔPV with respect to the crank angle θ exceeds the first predetermined value that is greater than or equal to zero and is less than or equal to the first predetermined value. 4. The fuel injection control device for an internal combustion engine according to claim 3, wherein a time that is smaller than a second predetermined value is adopted as a time after the end of combustion corresponding to the early injection and before the start of main injection. The exothermic parameter is assumed to be that the product PV ^{κ of the} combustion chamber volume V determined by the combustion chamber pressure P detected by the in-cylinder pressure sensor and the crank angle θ and a predetermined constant κ power and combustion does not occur in the combustion chamber. The fuel injection of the internal combustion engine according to claim 1 or 2, which is a difference ΔPV ^κ of a product PV ^κ base of the pressure in the combustion chamber generated only by the compression of the piston and the constant κ power of the combustion chamber volume determined from the crank angle. Control device. During the compression stroke of the internal combustion engine, the value of the change rate d (ΔPV ^κ) / dθ with respect to the crank angle θ of the Pv ^kappa is equal to or less than said first predetermined value after exceeding the first predetermined value or more zero zero 6. The fuel injection control for an internal combustion engine according to claim 5, wherein a time that is smaller than a second predetermined value in the vicinity is adopted as a time after the end of combustion corresponding to the early injection and before the start of main injection. apparatus. Combustion state detection means for detecting a combustion state relating to the main injection is further provided, and the injection timing of the main injection is advanced from the misfire limit time or the combustion deterioration limit time calculated by the limit time calculation means or from these limit times. As a result, the combustion state in which a misfire related to the main injection is generated by the combustion state detecting means or the combustion state in which the deterioration of the main injection combustion is caused by retarding the main injection is detected. The internal combustion engine according to any one of claims 1 to 6, wherein when detected, the misfire limit time or the combustion deterioration limit time calculated by the limit time calculation means is corrected to the advance side. Fuel injection control device.

Images