JP2000240497A - Fuel injection amount detecting device for internal combustion engine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To high-precisely detect a fuel injection amount from a differential pressure between a reference pressure and a cylinder pressure by calculating a reference pressure (a pressure of cylinder air excluding the increase of a pressure by fuel) corresponding to an engine operation condition during running of an engine without lowering operability. SOLUTION: The waveform of a cylinder pressure (a motoring pressure) Pm equivalent to one cycle during cut of fuel injection is detected by a cylinder pressure sensor and stored at a memory. By determining a pressure ratio between a detecting pressure Pk of a cylinder pressure sensor and a motoring pressure Pm by a crank angle θ0 before ignition during operation of an engine, a pressure ratio between a reference pressure Pb on a present engine operation condition and the motoring pressure Pm is determined. The reference pressure Pb is calculated by multiplying the motoring pressure Pm by a pressure ratio at a crank angle following the crank angle θ0, and from a differential pressure between a detecting pressure Pk (θQ) at an injection amount detecting timing θQ after a lapse of a given time starting from an ignition timing θf and the reference pressure Pb(θQ), a fuel injection amount is calculated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a fuel injection amount detecting device for an internal combustion engine which detects an in-cylinder pressure of an internal combustion engine (engine) and detects a fuel injection amount from the detected pressure.

[0002]

2. Description of the Related Art The fuel injection amount of an engine is a control parameter that directly affects the engine output, fuel consumption and emission. In general, the amount of fuel injection varies depending on the individual difference of fuel injectors and aging, and in particular, in a diesel engine that injects fuel at a high pressure, it is easily affected by the individual difference of fuel injectors and aging, and the fuel injection amount is high. The amount tends to vary. Further, when performing a plurality of fuel injections in the same cycle, a change in the fuel injection amount has a large effect on the engine output, fuel consumption, and emission.
Therefore, in order to improve engine output, fuel efficiency and emission, it is necessary to precisely control the fuel injection amount.

Therefore, in a diesel engine, as shown in Japanese Patent Application Laid-Open No. H10-184420, attention is paid to the characteristic that when fuel is burned, a combustion pressure corresponding to the fuel injection amount is generated. A technology that detects the actual fuel injection amount by detecting the pressure increase due to combustion from this detection value and adjusts the injection parameter so that this fuel injection amount matches the target fuel injection amount is studied. Have been.

However, since a diesel engine has a large compression ratio and draws a large amount of air into a cylinder, the compression pressure of the cylinder air (hereinafter referred to as "reference pressure") is relatively higher than the pressure increase due to combustion. Become larger. For this reason, unless the reference pressure is removed from the detected value of the in-cylinder pressure, the pressure rise due to combustion cannot be accurately detected.

Therefore, the above publication discloses a fuel injection system in which a pilot injection is performed prior to a main injection for generating an engine output, the pilot injection is stopped during operation of the engine, and the in-cylinder pressure at that time is detected as a reference pressure. Then, the reference pressure is subtracted from the in-cylinder pressure detected at the time of performing the pilot injection, a pressure increase amount due to combustion of the pilot injection fuel is obtained, and the fuel injection amount of the pilot injection is calculated from the pressure increase amount. I have.

[0006]

Incidentally, the reference pressure is not constant but changes depending on the engine operating conditions and the like. In particular, the change tends to be large in an engine with a supercharger. Therefore, in order to accurately determine the pressure increase due to combustion, it is necessary to subtract the reference pressure according to the engine operating condition at that time from the detected in-cylinder pressure.

However, according to the technique disclosed in the above publication, when detecting the reference pressure during the operation of the engine, the pilot injection must be stopped. The pilot injection must be stopped every time the air pressure changes, which has the disadvantage that the engine output becomes unstable and the drivability deteriorates.

If the reference pressure at each crank angle is calculated in advance for each engine operating condition at the design stage and stored in a memory such as a map, the pilot injection can be performed without stopping the pilot injection. Although the corresponding reference pressure can be obtained from the memory, it is practically difficult to calculate and store in advance all the reference pressures for all the ever-changing engine operating conditions. In addition, it is necessary to store a large amount of data relating to the reference pressure for each engine operating condition, and there is a disadvantage that a large-capacity memory is required and the cost is increased.

The present invention has been made in view of such circumstances. Therefore, it is an object of the present invention to provide a method for operating an engine without having to pre-calculate and store a reference pressure for each engine operating condition in a map or the like. It is possible to easily obtain a reference pressure according to the engine operating conditions without deteriorating drivability, and accurately detect the fuel injection amount from the detected value of the in-cylinder pressure using this reference pressure. It is an object of the present invention to provide a fuel injection amount detecting device for an internal combustion engine which can perform the above.

[0010]

According to a first aspect of the present invention, there is provided a fuel injection amount detecting apparatus for an internal combustion engine according to the present invention. Based on the internal pressure (hereinafter referred to as “motoring pressure”), the current pressure of the in-cylinder air (hereinafter referred to as “reference pressure”) excluding the pressure increase due to combustion is calculated by reference pressure calculation means, and the in-cylinder pressure is detected The fuel injection amount is calculated by the injection amount calculating means based on the current in-cylinder pressure (hereinafter, referred to as “detected pressure”) detected by the means and the reference pressure.

In this case, the motoring pressure is the in-cylinder pressure during non-combustion under a certain engine operating condition, that is, the compression pressure of in-cylinder air excluding a pressure increase due to combustion. Therefore, since the motoring pressure is equivalent to the reference pressure under the engine operating conditions when the motoring pressure is detected, the motoring pressure is based on the relationship between the engine operating conditions at the time of detecting the motoring pressure and the current engine operating conditions. The reference pressure under the current engine operating conditions can be calculated as data. For this reason, in the present invention, it is possible to easily calculate the reference pressure according to the engine operating condition at the time during the engine operation, without having to previously calculate the reference pressure for each engine operating condition and store it in a map or the like. The fuel injection amount can be accurately calculated from the comparison between the reference pressure and the detected pressure. In addition, when obtaining the reference pressure for each engine operating condition, it is not necessary to stop the fuel injection, so that deterioration of drivability can be avoided.
Further, since there is no need to store a large amount of data relating to the reference pressure for each engine operating condition, a large-capacity memory is not required, and the cost can be reduced accordingly. In addition, since the motoring pressure serving as base data when calculating the reference pressure is detected by the in-cylinder pressure detection means during operation of the engine, it is possible to cope with differences in motoring pressure characteristics due to individual differences between individual engines. .

Here, the fuel injection cut performed at the time of deceleration of the vehicle, at the time of high rotation, or the like is performed because the inside of the cylinder is in a non-combustion state.
It is preferable that the in-cylinder pressure detecting means detects the in-cylinder pressure at the time of fuel injection cut as the motoring pressure. By doing so, it is not necessary to create a non-combustion state for detecting the motoring pressure during the operation of the engine, and without impairing drivability, the motoring pressure can be reduced by utilizing the fuel injection cut during vehicle deceleration. Can be detected.

It is preferable that the reference pressure is calculated by multiplying the motoring pressure by a coefficient obtained from the pressure ratio between the detected pressure and the motoring pressure.
In other words, the motoring pressure is the detected pressure under the engine operating conditions at the time of detecting the motoring pressure (= the reference pressure at the time of detecting the motoring pressure). The pressure ratio with the reference pressure at the time of detecting the pressure is a powerful parameter for estimating the pressure ratio between the reference pressure under the current engine operating conditions and the reference pressure at the time of detecting the motoring pressure. Therefore, by multiplying the motoring pressure (= the reference pressure at the time of detecting the motoring pressure) by the coefficient obtained from the pressure ratio, the reference pressure under the current engine operating conditions can be easily calculated.

In this case, it is preferable that the pressure ratio between the detected pressure and the motoring pressure is calculated at at least one crank angle before the fuel is ignited.
As shown in FIG. 2, before the fuel ignition, the in-cylinder pressure does not increase due to the combustion, so that the detected pressure and the reference pressure substantially match. Therefore, if the pressure ratio is calculated before the fuel is ignited, it is possible to calculate the pressure ratio which is not affected by the pressure increase due to the combustion, and to accurately calculate the reference pressure even at the crank angle after the fuel is ignited.

By the way, the in-cylinder pressure sensor used as the in-cylinder pressure detecting means may have an offset error in the output characteristics depending on operating conditions such as temperature, which causes a decrease in the accuracy of detecting the fuel injection amount.

As a countermeasure, an offset error of the output characteristic of the in-cylinder pressure detecting means is calculated based on a plurality of detected pressures detected at a plurality of crank angles by the in-cylinder pressure detecting means. The output characteristic of the in-cylinder pressure detecting means may be corrected by the offset error.
In this way, even if an offset error occurs in the output characteristics of the in-cylinder pressure detecting means, the fuel injection amount is accurately calculated using the correction value excluding the offset error from the output of the in-cylinder pressure detecting means. can do.

It is preferable that the motoring pressure is detected by the in-cylinder pressure detecting means under a predetermined condition every time fuel injection is cut, and the stored value of the motoring pressure is updated. In this way, even if the characteristics of the internal combustion engine and the output characteristics of the in-cylinder pressure detecting means change over time, the reference pressure can be accurately calculated based on the latest motoring pressure updated according to the change over time. Thus, it is possible to prevent the detection accuracy of the fuel injection amount from lowering due to a change with time.

In the output characteristics of the in-cylinder pressure detecting means, a gain (output sensitivity) with respect to a pressure change may change due to a use condition, a temporal change, or the like. As a countermeasure, the gain error of the output characteristic of the in-cylinder pressure detecting means is obtained by comparing the motoring pressure detected at at least one crank angle by the in-cylinder pressure detecting means with its standard value. The output characteristic of the in-cylinder pressure detecting means may be corrected by the gain error. With this configuration, even if a gain error occurs in the output characteristic of the in-cylinder pressure detecting means, the fuel injection amount can be accurately calculated using the correction value excluding the gain error from the output of the in-cylinder pressure detecting means. Can be.

The fuel injection amount may be calculated, for example, by calculating the fuel injection amount based on a ratio between a detected pressure after fuel ignition and a reference pressure. The fuel injection amount may be calculated based on the pressure difference between the detected pressure and the reference pressure after a predetermined time has elapsed from the fuel ignition timing. Since the pressure difference between the detected pressure and the reference pressure after fuel ignition corresponds to the pressure increase due to combustion, the fuel injection amount can be calculated with high accuracy from this pressure difference. Moreover, since the differential pressure is obtained after a predetermined time has elapsed from the ignition timing, the combustion time from the ignition timing to the detection of the differential pressure can be kept constant irrespective of the engine speed, and the influence of the engine speed is affected. It is possible to detect the fuel injection amount with high accuracy.

Further, the differential pressure between the detected pressure and the reference pressure from when the fuel is ignited until a predetermined time has elapsed is integrated in a predetermined cycle, and the fuel injection amount is calculated based on the integrated value. You may do it. In this way, even if noise or the like is superimposed on the output of the in-cylinder pressure detecting means and the differential pressure data temporarily fluctuates, the effect can be suppressed and the fuel injection amount can be accurately calculated.

Alternatively, the fuel injection amount may be calculated based on a differential pressure between a detected pressure and a reference pressure at a predetermined crank angle. This eliminates the need to calculate the timing (crank angle) for obtaining the differential pressure, so that the fuel injection amount calculation program can be simplified and the calculation amount (CPU load) can be reduced.

Further, the differential pressure between the detected pressure from the ignition timing of the fuel to the predetermined crank angle and the reference pressure is integrated in a predetermined cycle, and the fuel injection amount is calculated based on the integrated value. You may do it. Also in this case, it is possible to accurately calculate the fuel injection amount while suppressing the influence of the variation of the differential pressure data due to noise or the like.

Further, the fuel injection amount may be calculated based on the rate of increase of the differential pressure between the detected pressure and the reference pressure after a predetermined time has elapsed from the ignition timing of the fuel. If an offset error occurs in the output characteristic of the in-cylinder pressure detecting means, the offset error is also included in the differential pressure data. However, when the rate of increase of the differential pressure is obtained, the offset error can be almost canceled. Therefore, if the fuel injection amount is calculated from the rate of increase in the differential pressure, the effect of the offset error can be suppressed to a small value and the fuel injection amount can be calculated with high accuracy.

In a fuel injection system that performs a main injection for generating an engine output and a pilot injection preceding the same within one cycle, a difference after ignition of the main injection fuel is calculated when calculating a fuel injection amount of the pilot injection. When the pressure is used, the pressure is increased by the combustion of the main injection fuel.

Therefore, when calculating the fuel injection quantity of the pilot injection, the pilot injection of the pilot injection is performed based on the differential pressure between the detected pressure and the reference pressure at or before the ignition timing of the main injection fuel. It is preferable to calculate the fuel injection amount. With this configuration, it is possible to accurately calculate the fuel injection amount of the pilot injection using the differential pressure (the pressure increase amount due to only the fuel combustion during the pilot injection) that is not affected by the pressure increase due to the combustion of the main injection fuel. it can.

In the case where a plurality of fuel injections are performed in one cycle, the detected pressure and the reference pressure at or before the ignition timing of the Nth injection fuel (N is an integer of 2 or more) are set. The fuel injection amount of the (N-1) th or earlier injection may be calculated based on the pressure difference from the pressure.
This makes it possible to accurately calculate the fuel injection amount of the (N-1) th or earlier injection based on the differential pressure that is not affected by the pressure increase due to the Nth injection fuel combustion.

According to a fifteenth aspect, the differential pressure between the detected pressure used for calculating the fuel injection amount and the reference pressure is reduced by the detected pressure before ignition to make it dimensionless, and the fuel pressure is calculated based on the value. The injection amount may be calculated. If a gain error is included in the output characteristic of the in-cylinder pressure detecting means, the differential pressure between the detected pressure and the reference pressure and the detected pressure each include the same gain error. If subtracted by
The gain error can be almost canceled, and the fuel injection amount can be accurately calculated from this value.

[0028]

DESCRIPTION OF THE PREFERRED EMBODIMENTS [Embodiment (1)] An embodiment (1) in which the present invention is applied to a four-cylinder diesel engine will be described below with reference to FIGS.

First, the configuration of the entire engine control system will be described with reference to FIG. An electromagnetic valve type fuel injection valve 12 is attached to each cylinder of the diesel engine 11 which is an internal combustion engine. Each of the fuel injection valves 12 is provided with a common rail fuel that is stored at a high pressure from a high pressure pump (not shown). 13
Distributed through The common rail 13 has a fuel pressure (common rail fuel pressure) distributed to the fuel injection valves 12.
Is installed. Further, one representative cylinder of the diesel engine 11 is provided with an in-cylinder pressure sensor 15 (in-cylinder pressure detecting means) for detecting an in-cylinder pressure.

Further, a crank angle sensor 16 for outputting a pulse signal at predetermined crank angles is provided near the crankshaft 20 of the engine 11, and a cylinder discriminating sensor 17 is provided near the camshaft (not shown). Is installed. Also,
A load sensor 18 such as an accelerator sensor is provided on an accelerator pedal (not shown).

The output signals of the various sensors described above are input to an engine electronic control circuit (hereinafter referred to as “ECU”) 19. The ECU 19 is mainly composed of a microcomputer, and calculates a fuel injection amount and a fuel injection timing based on an engine operating state detected by various sensors.
The fuel injection valve 12 is controlled based on the calculation result.

Further, the ECU 19 calculates the non-combustion in-cylinder pressure (motoring pressure) Pm detected by the in-cylinder pressure sensor 15 under predetermined conditions at the time of fuel injection cut.
In addition to functioning as a reference pressure calculating means for calculating the current in-cylinder air pressure (reference pressure) Pb excluding the pressure increase due to combustion, the in-cylinder pressure (detected pressure) Pk detected by the in-cylinder pressure sensor 15 and the reference pressure It functions as an injection amount calculating means for calculating a fuel injection amount based on a pressure difference from Pb (a pressure increase amount due to combustion). Then, the ECU 19 performs feedback control of the fuel injection amount of the fuel injection valve 12 so that the detected fuel injection amount matches the target fuel injection amount.

Here, a method of detecting the fuel injection amount by the ECU 19 will be described. The waveform of the motoring pressure Pm (θ) shown in FIG. 2 is obtained when a predetermined operating condition is satisfied during the fuel injection cut (for example, when the engine speed is a predetermined engine speed N).
), The output of the in-cylinder pressure sensor 15 is read for one cycle, and the backup RA in the ECU 19 is read.
It is stored in a non-volatile memory (not shown) such as M. The waveform of the motoring pressure Pm (θ) is newly detected every time a predetermined running time T1 (for example, 100 hours) elapses and a predetermined operating condition is satisfied during the fuel injection cut.
Update stored values.

The detected pressure Pk (θ) is detected by the in-cylinder pressure sensor 15 at a predetermined sampling interval Δθ (for example, 1 ° C.). On the other hand, the reference pressure Pb (θ) is
It is calculated as follows from the motoring pressure Pm (θ) and the detected pressure Pk (θ).

Detected pressure Pk (θ) before fuel ignition
Does not include the pressure rise due to combustion, so the reference pressure Pb
(Θ) substantially coincides with (θ). Therefore, the detected pressure Pk (θ at the calculated crank angle θ0 set in the compression stroke before ignition
By calculating the pressure ratio H between 0) and the motoring pressure Pm (θ0) according to the following equation, it is possible to calculate the pressure ratio H that is not affected by the pressure rise due to combustion at all. H = Pk (θ0) / Pm (θ0)

Here, the calculated crank angle θ0 is determined by the pressure ratio H
In order to increase the calculation accuracy of BTDC, it is preferable that the temperature be as short as possible immediately before ignition, for example, BTDC is set to 10 ° C. (10 ° A before compression top dead center). Alternatively, the pressure ratio H may be calculated only at one crank angle θ0 before ignition, but the pressure ratio H may be calculated at a plurality of crank angles before ignition and an average value of the plurality of pressure ratios H may be used. You may do it.

By multiplying the motoring pressure Pm (θ) by the pressure ratio H for each crank angle θ after the calculated crank angle θ0, the reference pressure P at each crank angle θ is obtained.
b (θ) is calculated. Pb (θ) = H × Pm (θ) Thereby, the reference pressure Pb can be easily calculated.

As shown in FIG. 3, at the injection amount detection timing θQ after a predetermined time t0 has elapsed from the fuel ignition timing θf,
Detected pressure Pk (θQ) and calculated reference pressure Pb (θQ)
Is calculated by the following equation. ΔP (θQ) = Pk (θQ) -Pb (θQ)

The method of detecting the fuel ignition timing θf is, for example, that the differential pressure ΔP (θ) between the detected pressure Pk (θ) and the reference pressure Pb (θ) is larger than the ignition determination value F (for example, 100 kPa). The changed crank angle θf may be used as the ignition timing. As another method, the ignition timing may be detected using, for example, a combustion light sensor, a knock sensor, an ion current sensor, or the like.

This differential pressure ΔP (θQ) corresponds to the pressure increase due to combustion. Also, a predetermined time t from the ignition timing θf
Since the differential pressure ΔP (θQ) is calculated at the injection amount detection timing θQ after the lapse of 0, the combustion time from detection of the ignition timing θf to detection of the differential pressure ΔP (θQ) is always constant regardless of the engine speed. be able to. As a result, as shown in the experimental results of FIG. 4, the differential pressure ΔP (θ
The relationship between Q) and the fuel injection amount Q can be approximated by a straight line (proportional expression), and the fuel injection amount Q can be easily calculated by the following equation. Q = α × ΔP (θQ) In the above equation, α is a proportionality constant (slope of a straight line).
In order to further improve the accuracy of detecting the fuel injection amount Q, the relationship between the differential pressure ΔP (θQ) and the fuel injection amount Q may be approximated by a higher-order function (curve) having a quadratic function or more.

The detection of the fuel injection amount by the ECU 19 described above is executed by a fuel injection amount detection program shown in FIG. This program is executed every predetermined time or every predetermined crank angle, and the fuel injection amount of the representative cylinder provided with the in-cylinder pressure sensor 15 is detected. When the present program is started, first, in step 101, it is determined whether or not the accumulated traveling time since the previous update of the waveform of the motoring pressure Pm (θ) has exceeded a predetermined time T1 (for example, 100 hours). I do. If the accumulated traveling time has not reached the predetermined time T1, the process proceeds to step 105 without updating the waveform of the motoring pressure Pm (θ).

On the other hand, if the integrated running time exceeds the predetermined time T1, the routine proceeds to step 102, where the predetermined operating condition is satisfied during the fuel injection cut (for example, when the engine speed reaches the predetermined speed N). ), The output of the in-cylinder pressure sensor 15 is read as the motoring pressure Pm (θ) for one cycle, and the data of the motoring pressure Pm (θ) stored in the nonvolatile memory of the ECU 19 is updated.
Thereafter, in step 103, it is determined whether or not the updated peak pressure Pmax of the motoring pressure Pm (θ) is higher than a predetermined value Ps. If the peak pressure Pmax is equal to or less than the predetermined value Ps, it is determined that the in-cylinder pressure is abnormally reduced.
Proceeding to step 104, an abnormal display is made by turning on a warning lamp (not shown) or the like to notify the driver of the abnormal decrease in the in-cylinder pressure, and the program is terminated.

On the other hand, if it is determined in step 103 that the peak pressure Pmax is higher than the predetermined value Ps,
It is determined that the in-cylinder pressure is normal, and the routine proceeds to step 105. In this step 105, the waveform of the motoring pressure Pm (θ) stored in the nonvolatile memory of the ECU 19 is read, and in the next step 106, the current crank angle θ
Is compared with the calculated crank angle θ0 (for example, BTDC 10 ° C.), and until the calculated crank angle θ0 is reached, step 1 is executed.
Wait at 06. Thereafter, when the calculated crank angle θ0 is reached, the routine proceeds to step 107, where the detected pressure Pk (θ0) and the motoring pressure Pm at the calculated crank angle θ0 are obtained.
The pressure ratio H to (θ0) is calculated by the following equation. H = Pk (θ0) / Pm (θ0)

In the next step 108, the calculated crank angle θ
0, for each crank angle θ, the motoring pressure Pm
By multiplying (θ) by the pressure ratio H, the reference pressure Pb (θ) at each crank angle θ is calculated by the following equation. Pb (θ) = H × Pm (θ) The data of the reference pressure Pb (θ) is temporarily stored in a memory such as a RAM of the ECU 19 until this program ends.

Thereafter, the routine proceeds to step 109, where the injection amount detection timing θQ is calculated. The calculation method is, for example, a crank angle θQ corresponding to a predetermined time t0 after the ignition timing θf.
Is calculated.

Thereafter, the routine proceeds to step 110, where the current crank angle θ is compared with the injection amount detection timing θQ, and the routine stands by at step 110 until the injection amount detection timing θQ is reached. Thereafter, when the injection amount detection timing θQ is reached, step 11
1 and the differential pressure ΔP (θQ
) Is calculated by the following equation. ΔP (θQ) = Pk (θQ) -Pb (θQ)

In the next step 112, the previously obtained proportional constant α is read from the memory, the fuel injection amount Q is calculated by the following equation, and the program is terminated. Q = α × ΔP (θQ) The proportionality constant α may be corrected according to the engine operating conditions.

As shown by the broken line in FIG. 6, for example, when a reference pressure corresponding to a low engine load is calculated and stored in advance, and this reference pressure is applied to all engine operating conditions, a high pressure when the engine is under a high load is obtained. In this case, the reference pressure is greatly different from the actual reference pressure, and the fuel injection amount may be erroneously detected. As a countermeasure, if the reference pressure is calculated in advance for each engine operating condition and stored in a map or the like, and the reference pressure corresponding to the engine operating condition is obtained from the map or the like, the accuracy of detecting the fuel injection amount is improved. Although it is possible, it is practically difficult to calculate and store all the reference pressures in advance for all the ever-changing engine operating conditions. In addition, it is necessary to store a large amount of data relating to the reference pressure for each engine operating condition, and there is a disadvantage that a large-capacity memory is required and the cost is increased.

On the other hand, in this embodiment (1), FIG.
Focusing on the point where the detected pressure Pk (θ0) at the calculated crank angle θ0 before the fuel ignition and the reference pressure Pb (θ0) at the calculated crank angle θ0 of the engine operating conditions at that time substantially match as shown in FIG. By calculating the pressure ratio H between the detected pressure Pk (θ0) and the motoring pressure Pm (θ0) at the calculated crank angle θ0 before ignition, the pressure between the reference pressure Pb and the motoring pressure Pm under the current engine operating conditions is calculated. The ratio H is obtained, and the reference pressure Pb (θ) at each crank angle θ is calculated by multiplying the motoring pressure Pm (θ) by the pressure ratio H for each crank angle θ after the calculated crank angle θ0. Thereby, using the pressure ratio H which is not affected at all by the pressure rise due to combustion,
It is possible to accurately calculate the reference pressure Pb (θ) of the crank angle after fuel ignition.

After a lapse of a predetermined time t0 from the ignition timing θf, the detected pressure Pk (θQ) and the reference pressure Pb (θQ)
, The combustion time from the ignition timing θf to the detection of the differential pressure ΔP (θQ) can be kept constant irrespective of the engine speed. θQ), the fuel injection amount Q can be accurately calculated. Moreover, unlike the related art, it is not necessary to stop fuel injection when detecting the reference pressure, so that drivability can be prevented from deteriorating. Furthermore, since it is not necessary to calculate the reference pressure for each engine operating condition in advance and store it in a map or the like, a large-capacity memory is not required, and the cost can be reduced accordingly.

In the above embodiment (1), the in-cylinder pressure sensor 1 is used when the fuel injection is cut, for example, when the vehicle is decelerated.
5, the motoring pressure Pm is detected, so that the motoring pressure Pm can be detected without impairing drivability, and it is possible to cope with differences in motoring pressure characteristics due to individual differences of individual engines. In addition, it is possible to reduce the variation in the accuracy of detecting the fuel injection amount due to the individual difference of the engine. Moreover, since the motoring pressure Pm is updated every time the predetermined time T1 has elapsed, even if the engine characteristics and the output characteristics of the in-cylinder pressure sensor 15 change over time, the motoring pressure Pm is updated based on the updated motoring pressure Pm. The reference pressure Pb can be calculated with high accuracy, and it is possible to prevent a decrease in the accuracy of detecting the fuel injection amount due to a temporal change.

[Embodiment (2)] Next, an embodiment (2) of the present invention will be described with reference to FIGS. In the above embodiment (1), after a lapse of a predetermined time t0 from the ignition timing θf, a differential pressure ΔP (θQ) between the detected pressure Pk (θQ) and the reference pressure Pb (θQ) is calculated, and from this differential pressure ΔP (θQ). Although the fuel injection amount Q is calculated, the present embodiment (2)
Then, as shown in FIG. 7, a predetermined time t
During the period up to the injection amount detection timing θQ after the lapse of 0, the differential pressure ΔP (θ) between the detected pressure Pk (θ) and the reference pressure Pb (θ) is integrated at every predetermined sampling interval Δθ, and this differential pressure is calculated. The fuel injection amount Q is calculated from the integrated value ΣΔP (θ).

As shown in the experimental results of FIG.
differential pressure integrated value ΣΔP from f to injection amount detection timing θQ
The relationship between (θ) and the fuel injection amount Q can be approximated by a straight line (proportional expression), and the fuel injection amount Q can be easily calculated by the following equation. Q = α × ΣΔP (θ) In the above equation, α is a proportionality constant (slope of a straight line).
In order to further improve the detection accuracy of the fuel injection amount Q, the relationship between the differential pressure integrated value ΣΔP (θ) and the fuel injection amount Q may be approximated by a quadratic function or higher-order function (curve). good.

The fuel injection amount detection program shown in FIG. 9 executed in the embodiment (2) performs the processing in steps 110 to 112 in FIG.
Steps 3 to 117 have been changed, and the other steps are the same as those in FIG. In this program, after calculating the injection amount detection timing θQ (step 109), when the crank angle θ reaches the ignition timing θf, the process proceeds from step 113 to step 114, and the differential pressure ΔP (θ) is calculated by the following equation. ΔP (θ) = Pk (θ) -Pb (θ)

In the next step 115, the integrated differential pressure value ΣΔP (θi) up to this time is replaced with the integrated differential pressure value ΣΔP
(Θi−1) is obtained by adding the present differential pressure ΔP (θi). ΣΔP (θi) = ΣΔP (θi-1) + ΔP (θi) Thereafter, the routine proceeds to step 116, where the crank angle θ is compared with the injection amount detection timing θQ, and if the injection amount detection timing θQ has not been reached, the above-mentioned step 114 is performed. And the differential pressure ΔP (θ)
Is repeated (steps 114 and 115).

After that, the crank angle θ becomes equal to the injection amount detection timing θ.
When Q is reached, steps 116 to 117
To read the previously obtained proportional constant α from the memory,
The proportional constant α is multiplied by the integrated differential pressure value ΣΔP (θ) to obtain the fuel injection amount Q, and the program ends. Q = α × ΣΔP (θ) The proportionality constant α may be corrected according to the engine operating conditions.

In the fuel injection amount detection process of the embodiment (2) described above, the differential pressure ΔP (θ) is integrated during the period from the ignition timing θf to the injection amount detection timing θQ after a lapse of a predetermined time t0, and Since the fuel injection amount Q is calculated from the differential pressure integrated value ΣΔP (θ), even if noise or the like is superimposed on the output of the in-cylinder pressure sensor 15 and the data of the differential pressure ΔP (θ) temporarily varies, It is possible to accurately calculate the fuel injection amount Q while suppressing the influence thereof.

If the pressure difference ΔP (θ) at the crank angle before the ignition timing θf can be regarded as substantially 0, the processing of step 113 is omitted, and after the calculation of the injection amount detection timing θQ (step 109), the ignition is started. Without waiting for the timing θf, the differential pressure Δ
The integration process of P (θ) (steps 114 and 115) may be started.

[Embodiment (3)] In the embodiment (1), the crank angle after a predetermined time t0 has elapsed from the ignition timing θf is set as the injection amount detection timing θQ. However, in the embodiment (3) of the present invention, As shown in FIG. 10, the injection amount detection timing θQ
Is set to a predetermined crank angle after the ignition timing which is determined in advance regardless of the ignition timing, and from the differential pressure ΔP (θQ) between the detected pressure Pk (θQ) and the reference pressure Pb (θQ) at the injection amount detection timing θQ. The fuel injection amount Q is calculated.

This eliminates the need to calculate the injection amount detection timing θQ, so that the processing in step 109 can be omitted in the fuel injection amount detection program of FIG. 5 and the amount of calculation can be reduced. And ECU1
9 can be reduced. In the present invention, the crank angle obtained by adding a predetermined crank angle to the detected crank angle of the ignition timing θf is used as the injection amount detection timing θ.
Q is good.

[Embodiment (4)] In the embodiment (2), the differential pressure Δ from the ignition timing θf until a predetermined time t0 has elapsed.
Although P (θ) is integrated, in the embodiment (4) of the present invention,
As shown in FIG. 11, the injection amount detection timing θQ is set to a predetermined crank angle after a predetermined ignition timing regardless of the ignition timing, and during the period from the ignition timing θf to the injection amount detection timing θQ, Differential pressure ΔP for each sampling interval Δθ
(Θ) is integrated, and the fuel injection amount Q is calculated from the differential pressure integrated value ΣΔP (θ).

In this way, in the fuel injection amount detection program of FIG. 9, the processing of step 109 can be omitted, and the amount of calculation can be reduced. In this case, the crank angle obtained by adding a predetermined crank angle to the detected crank angle of the ignition timing θf may be used as the injection amount detection timing θQ.

[Embodiment (5)] Next, an embodiment (5) of the present invention will be described with reference to FIGS. In the embodiment (1), the differential pressure ΔP between the detected pressure Pk (θQ) and the reference pressure Pb (θQ) at the injection amount detection timing θQ.
Although the fuel injection amount was calculated from (θQ), in the present embodiment (5), as shown in FIG.
, The detected pressure Pk (θQ) and the reference pressure Pb (θQ
) And the rate of increase dΔP (θQ) / d of the differential pressure ΔP (θQ)
t is calculated, and the fuel injection amount Q is calculated from the differential pressure increase rate dΔP (θQ) / dt.

As shown in the experimental results of FIG. 13, the relationship between the differential pressure increase rate dΔP (θQ) / dt and the fuel injection amount Q at the injection amount detection timing θQ after a lapse of a predetermined time t0 from the ignition timing θf is represented by a straight line ( (Proportional formula), and the fuel injection amount Q can be easily calculated by the following formula. Q = α × dΔP (θQ) / dt In the above equation, α is a proportional constant (slope of a straight line).
In order to further improve the detection accuracy of the fuel injection amount Q, the relationship between the differential pressure increase rate dΔP (θQ) / dt and the fuel injection amount Q is approximated by a quadratic function or higher-order function (curve). May be.

Even if an offset error occurs in the output characteristic of the in-cylinder pressure sensor 15 and the differential pressure ΔP (θQ) includes the offset error, the offset error is obtained by calculating the differential pressure increase rate dΔP (θQ) / dt. Since it can be almost canceled, the differential pressure increase rate dΔ as in the present embodiment (5).
If the fuel injection amount is calculated from P (θQ) / dt, the effect of the offset error of the in-cylinder pressure sensor 15 can be reduced to calculate the fuel injection amount Q with high accuracy.

[Embodiment (6)] By the way, in a fuel injection system for performing main injection for generating engine output and pilot injection preceding it in one cycle, any one of the above-described embodiments (1) to (5) When the fuel injection amount of the pilot injection is calculated by using the method described above, if the above-described injection amount detection timing θQ is set after the ignition of the main injection fuel, the influence of the pressure increase due to the combustion of the main injection fuel may occur. Therefore, the fuel injection amount of the pilot injection cannot be accurately calculated.

As a countermeasure, the embodiment (6) of the present invention
Then, when calculating the fuel injection amount Qp of the pilot injection by using any of the methods (1) to (5), as shown in FIG.
Qp is set at or before the ignition timing θfm of the main injection fuel, and the fuel injection amount Qp of the pilot injection is calculated based on the differential pressure ΔP (θQp) which is not affected by the pressure increase due to the combustion of the main injection fuel. ing.

In this way, the fuel injection amount Qp of the pilot injection can be calculated with high accuracy without being affected by the pressure increase due to the combustion of the main injection fuel. Note that the main injection amount detection timing θQm may be set by any one of the above-described embodiments (1) to (5).

[Embodiment (7)] In the embodiment (7) of the present invention, in the system in which the fuel is injected a plurality of times (for example, three times) in one cycle, the embodiment (1) is used.
When calculating the fuel injection amount by using any one of the methods (1) to (5), as shown in FIG.
Set.

More specifically, the first injection amount detection timing θQ1
Is set at or before the ignition timing θf2 of the second injection fuel, and the fuel injection amount Q1 of the first injection is determined based on the differential pressure ΔP (θQ1) which is not affected by the pressure increase due to the combustion of the second injection fuel. Is calculated.

Similarly, the second injection amount detection timing θQ2 is set to 3
Set the ignition timing of the second injected fuel θf3 or earlier,
Based on the differential pressure ΔP (θQ2) which is not affected by the pressure rise due to the third injection fuel combustion, the fuel injection amount Q2 is calculated by combining a part of the first injection and the second injection.
Note that the third injection amount detection timing θQ3 may be set by any of the above-described embodiments (1) to (5).

In general, when a plurality of fuel injections are performed within one cycle, the (N-1) th injection amount detection is performed at or before the Nth (N is an integer of 2 or more) ignition timing of the injected fuel. It is sufficient to set the timing θQN−1, so that (N) is not affected by the pressure increase due to the combustion of the N-th injected fuel.
-1) It is possible to accurately calculate the fuel injection amount QN-1 of the injection before the first injection.

A NOx catalyst (not shown) for purifying NOx (nitrogen oxide) in the exhaust gas is provided in the exhaust pipe, and post-injection is performed in the expansion stroke after the main injection to supply fuel (HC) to the NOx catalyst. In the system for supplying as a reducing agent, by applying the embodiment (7), the combustion amount of the post-injection fuel can be detected, and the post-injection timing is controlled so as to reduce the combustion amount. , NOx purification rate can be improved.

[Embodiment (8)] Next, an embodiment (8) of the present invention will be described with reference to FIGS. The in-cylinder pressure sensor 15 may have an offset error in its output characteristics depending on operating conditions such as temperature (see FIG. 16A), which causes a decrease in fuel injection amount detection accuracy. This offset error can be obtained as follows.

Here, the offset error is b, the detected pressures at crank angles θ1 and θ2 (where θ1 <θ2 <θ0) before ignition are Ps (θ1) and Ps (θ2), respectively, and the true in-cylinder pressure is Assuming that Pt (θ1) and Pt (θ2), they can be expressed as follows. Ps (θ1) = Pt (θ1) + b (1) Ps (θ2) = Pt (θ2) + b (2)

Assuming that the state change of the in-cylinder air from the crank angle θ1 to the crank angle θ2 is an adiabatic change, the following equation can be obtained. Pt (θ1) × ｛V (θ1)｝ γ = Pt (θ2) × {V (θ2)} γ Pt (θ2) / Pt (θ1) = {V (θ1) / V (θ2)} γ = K (3) where V (θ) is the cylinder volume, γ is the specific heat ratio, and K is V
(Θ) and γ.

By solving the above equations (1) to (3), the offset error b can be calculated by the following equation. b = 1 / (K−1) × {K × Ps (θ1) −Ps (θ2)} (4) If this offset error b is subtracted from the output of the in-cylinder pressure sensor 15, the in-cylinder pressure sensor 15 Can be corrected.

The output characteristics of the in-cylinder pressure sensor 15 are as follows:
The gain (output sensitivity) with respect to the pressure change may change due to use conditions, aging, and the like (see FIG. 17), which also reduces the accuracy of detecting the fuel injection amount. This gain error can be obtained as follows.

Here, the gain error is a and the standard value of the motoring pressure at the calculated crank angle θ0 is Pmt (θ0
), Assuming that the detected value of the motoring pressure is Pms (θ0), the following expression can be obtained. The standard value Pmt (θ0) of the motoring pressure is calculated from the calculated crank angle θ0.
Is the standard motoring pressure in, set in advance based on design data, or in the initial state (before deterioration)
The motoring pressure detected by the in-cylinder pressure sensor 15 may be used. Pms (θ0) = a × Pmt (θ0)

Therefore, the gain error a can be calculated by the following equation. a = Pms (θ0) / Pmt (θ0) (5) By dividing the output of the in-cylinder pressure sensor 15 by the gain error a, the output of the in-cylinder pressure sensor 15 can be corrected for a gain error. . Since the detected value Pms (θ0) of the motoring pressure changes depending on the engine operating conditions such as the engine speed, the standard value Pmt (θ0) is set for each engine operating condition, and is set according to the engine operating condition at that time. The standard value Pmt (θ0) may be selected.

In the present embodiment (8), the ECU 19 executes the fuel injection amount detection program shown in FIGS. 18 and 19, so that the ECU 19 detects the in-cylinder pressure sensor 15 using the above equations (4) and (5). It functions as an offset error correction means and a gain error correction means for correcting the offset error and the gain error of the output characteristics, and calculates the fuel injection amount using the differential pressure in which the offset error and the gain error have been corrected.

The fuel injection amount detection program shown in FIGS. 18 and 19 is executed every predetermined time or every predetermined crank angle. When this program is started, first, in step 30
At 1, the waveform of the motoring pressure Pm (θ) is updated.
As a result, the motoring pressure Pm (θ) is updated in accordance with the temporal change in the output characteristics of the in-cylinder pressure sensor 15.
It should be noted that the motoring pressure Pm (θ) is updated by one cycle when the predetermined operating condition is satisfied during the fuel injection cut, as in the embodiment (1).
And updates the data of the motoring pressure Pm (θ) stored in the nonvolatile memory of the ECU 19.

Thereafter, at step 302, the motoring pressure Pm (θ
1), Pm (θ2) is read out and the motoring pressure Pm
The offset error bm of (θ) is calculated by the following equation. bm = 1 / (K-1) × ｛K × Pm (θ1) −Pm (θ
2)｝

In the next step 303, the offset error of the motoring pressure Pm (θ) is corrected using the offset error bm by the following equation. Pm ′ (θ) = Pm (θ) −bm The motoring pressure Pm ′ (θ) after the offset error correction obtained in this way is temporarily stored in a memory such as a RAM of the ECU 19 until this program ends. Remember.

Thereafter, when the crank angle θ becomes the crank angle θ1, the detected pressure Pk (θ1) is detected.
When the crank angle θ becomes the crank angle θ2, the detected pressure Pk (θ2) is detected, and the offset error bk of the detected pressure Pk (θ) is calculated by the following equation (steps 304 to 3).
08). bk = 1 / (K−1) × ｛K × Pk (θ1) −Pk (θ
2)｝

After that, the crank angle θ becomes the calculated crank angle θ.
When the value becomes 0, the process proceeds from step 309 to step 310 in FIG. 19, and the motoring pressure Pm '(θ0) after the offset error correction at the calculated crank angle θ0 and the standard value Pm of the motoring pressure at the calculated crank angle θ0.
t (θ0) is read from the nonvolatile memory of the ECU 19, and the gain error a is calculated by the following equation. a = Pm '(θ0) / Pmt (θ0)

In the next step 311, the offset error b
Using k, the offset error of the detected pressure Pk (θ0) is corrected by the following equation. Pk '(θ0) = Pk (θ0) -bk

Thereafter, at step 312, the pressure ratio H 'between the detected pressure Pk' (. Theta.0) after the offset error correction and the motoring pressure Pm '(. Theta.0) after the offset error correction.
Is calculated by the following equation. H '= Pk' (. Theta.0) / Pm '(. Theta.0) Thereafter, the routine proceeds to step 313, where the injection amount detection timing .theta.Q is calculated. The calculation method is the same as in the embodiment (1).
A crank angle .theta.Q corresponding to a predetermined time t0 after the ignition timing .theta.f is calculated.

After that, the crank angle θ becomes the injection amount detection timing θ
When it becomes Q, the process proceeds from step 314 to step 315, where the detected pressure Pk (θQ
) Is corrected by the following equation. Pk '(θQ) = Pk (θQ) -bk

In the next step 316, the reference pressure Pb '(θQ) after the offset error correction is calculated by the following equation. Pb ′ (θQ) = H ′ × Pm ′ (θQ)

In the next step 317, using the gain error a, the detected pressure Pk '(θQ) after offset error correction.
The gain error of the pressure difference ΔP ′ (θQ) between the pressure difference and the reference pressure Pb ′ (θQ) is corrected by the following equation. ΔP ′ (θQ) = 1 / a × {Pk ′ (θQ) −Pb ′
(ΘQ)｝ The differential pressure ΔP ′ (θQ) calculated in this way is a value in which both the offset error and the gain error have been corrected.

Thereafter, at step 318, the corrected pressure difference ΔP '(θQ) is multiplied by the proportionality constant α to obtain the fuel injection amount Q, and the program ends. Q = α × ΔP '(θQ)

In the embodiment (8) described above, even if an offset error or a gain error occurs in the output of the in-cylinder pressure sensor 15 due to a use condition or a change with time, the offset error or the gain error is obtained and the detected pressure is detected. Since the motoring pressure and the differential pressure are corrected, the fuel injection amount can be calculated using the data from which the offset error and the gain error of the output of the in-cylinder pressure sensor 15 have been removed, and the fuel injection amount detection with higher accuracy Is possible.

In the embodiment (8), the standard value Pmt of the motoring pressure at one crank angle θ0 is set.
(Θ0) is stored, and the gain error a is obtained at one crank angle θ0, but the standard value Pmt of the motoring pressure is obtained.
The waveform of (θ) may be stored, a gain error may be obtained at two or more crank angles, and the average value thereof may be used as the gain error.

In the above embodiment (8), after the offset error of both the detected pressure and the motoring pressure is corrected, the gain error of the differential pressure between the detected pressure after the offset error correction and the reference pressure is corrected. However, conversely, the offset error may be corrected after the gain error is corrected, and both the offset error and the gain error are read at the stage of reading the output of the in-cylinder pressure sensor 15. May be corrected. In short, the offset error and the gain error may be corrected before the fuel injection amount is calculated using the differential pressure. Alternatively, only one of the offset error and the gain error may be corrected.

[Embodiment (9)] When a gain error occurs in the output characteristic of the in-cylinder pressure sensor 15, the differential pressure ΔP
And the detected pressure Pk also include the same gain error, so that the gain error can be almost canceled by dividing the differential pressure ΔP by the detected pressure Pk.

Focusing on this point, in the embodiment (9) of the present invention, in each of the above embodiments (1) to (7), instead of the differential pressure ΔP, the differential pressure ΔP is replaced by the crank angle θZ before ignition.
The value .DELTA.P / Pk (.theta.Z) obtained by reducing the dimension by the detected pressure Pk (.theta.Z) in step (2) and using the dimensionless value is used. In this way, even when the in-cylinder pressure sensor 15 having a large gain error is used, the fuel injection amount can be accurately calculated based on the dimensionless data in which the gain error is almost canceled.

In the above embodiments, the motoring pressure Pm is updated every time the predetermined time T1 elapses.
It may be updated every time a predetermined traveling distance has elapsed. The in-cylinder pressure sensor 15 may be of a type that is integrated with the fuel injection valve 12 and a glow plug (not shown).

Further, in each of the above embodiments, the fuel injection amount is detected for the representative cylinder provided with the in-cylinder pressure sensor 15, but the in-cylinder pressure sensor 15 is provided for all cylinders, Alternatively, the fuel injection amount detection program may be executed to detect the fuel injection amount for each cylinder.

In each of the above embodiments, the present invention is applied to a four-cylinder diesel engine having a common race type injection system. However, the present invention is applied to a diesel engine having an injection system other than the common rail type and a diesel engine other than the four cylinder type. The invention may be applied.

[Brief description of the drawings]

FIG. 1 is a diagram showing a schematic configuration of an entire system according to an embodiment (1) of the present invention.

FIG. 2 is a diagram showing waveforms of a detected pressure, a reference pressure, and a motoring pressure.

FIG. 3 is a diagram for explaining a method of calculating a fuel injection amount according to the embodiment (1).

FIG. 4 is a diagram showing a relationship between a differential pressure and a fuel injection amount.

FIG. 5 is a flowchart showing a processing flow of a fuel injection amount detection program according to the embodiment (1).

6A and 6B are diagrams illustrating characteristics of a detected pressure and a reference pressure for the embodiment (1) and a comparative example, where FIG. 6A is a diagram when the engine is under a low load, and FIG.

FIG. 7 is a diagram for explaining a method of calculating a fuel injection amount according to the embodiment (2).

FIG. 8 is a diagram showing a relationship between a differential pressure integrated value and a fuel injection amount.

FIG. 9 is a flowchart showing a processing flow of a fuel injection amount detection program according to the embodiment (2).

FIG. 10 is a diagram for explaining a method of calculating a fuel injection amount according to the embodiment (3).

FIG. 11 is a diagram for explaining a method of calculating a fuel injection amount according to the embodiment (4).

FIG. 12 is a diagram for explaining a method for calculating a fuel injection amount according to the embodiment (5).

FIG. 13 is a diagram showing a relationship between a differential pressure increase rate and a fuel injection amount.

FIG. 14 is a diagram illustrating a method for calculating a fuel injection amount of pilot injection according to the embodiment (6).

FIG. 15 is a diagram for explaining a method of calculating a fuel injection amount in a system for performing fuel injection a plurality of times within one cycle according to the embodiment (7).

16A is a diagram illustrating output characteristics of an in-cylinder pressure sensor before offset error correction, and FIG. 16B is a diagram illustrating output characteristics of an in-cylinder pressure sensor after offset error correction.

FIG. 17 is a diagram for explaining a gain error of an output characteristic of the in-cylinder pressure sensor.

FIG. 18 is a flowchart (part 1) illustrating a processing flow of a fuel injection amount detection program according to the embodiment (8).

FIG. 19 is a flowchart (part 2) illustrating a processing flow of a fuel injection amount detection program of the embodiment (8).

[Explanation of symbols]

11: diesel engine (internal combustion engine), 12: fuel injection valve, 15: in-cylinder pressure sensor (in-cylinder pressure detecting means), 1
6 ... Crank angle sensor, 19 ... ECU (reference pressure calculating means, fuel injection amount calculating means, offset error correcting means, gain error correcting means).

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Kanehito Nakamura 1-1-1, Showa-cho, Kariya-shi, Aichi Prefecture Inside of Denso Co., Ltd. (72) Inventor Daisuke Kojima 1-1-1, Showa-cho, Kariya-shi, Aichi Prefecture Co., Ltd. F term in DENSO (reference) 3G084 AA01 BA33 CA00 DA04 DA13 EA05 EA06 EA07 EB06 EB12 EB22 EB24 EC04 FA13 FA21 FA38 3G301 HA02 JA19 JA20 JB10 KA26 MA24 NA04 NA08 NC01 NC08 ND02 PB03Z PC01Z PE03Z

Claims (15)

[Claims] 1. A fuel injection amount detecting device for an internal combustion engine, wherein the in-cylinder pressure of an internal combustion engine is detected by an in-cylinder pressure detecting means, and a fuel injection amount is detected based on the detected value. Based on the in-cylinder pressure during non-combustion detected in the past (hereinafter, referred to as "motoring pressure"), the current in-cylinder air pressure (hereinafter, referred to as "reference pressure") excluding a pressure increase due to combustion is calculated. A reference pressure calculating means, and an injection amount calculating means for calculating a fuel injection amount based on the current in-cylinder pressure (hereinafter referred to as “detected pressure”) detected by the in-cylinder pressure detecting means and the reference pressure. A fuel injection amount detecting device for an internal combustion engine. 2. The fuel injection amount detecting device for an internal combustion engine according to claim 1, wherein the in-cylinder pressure detecting means detects the in-cylinder pressure when fuel injection is cut off as the motoring pressure. 3. The reference pressure calculating means calculates the reference pressure by multiplying the motoring pressure by a coefficient obtained from a pressure ratio between the detected pressure and the motoring pressure. 3. The fuel injection amount detecting device for an internal combustion engine according to claim 1 or 2. 4. The fuel injection amount detecting device for an internal combustion engine according to claim 3, wherein the reference pressure calculating means calculates the pressure ratio based on at least one crank angle before fuel ignition. 5. An offset error of an output characteristic of said in-cylinder pressure detecting means is calculated based on a plurality of detected pressures detected at a plurality of crank angles by said in-cylinder pressure detecting means. 5. The fuel injection amount detecting device for an internal combustion engine according to claim 1, further comprising an offset error correcting unit that corrects an output characteristic of the internal pressure detecting unit. 6. A motoring pressure updating means for detecting the motoring pressure by the in-cylinder pressure detecting means under a predetermined condition for each fuel injection cut and updating a stored value of the motoring pressure. The fuel injection amount detecting device for an internal combustion engine according to any one of claims 1 to 5, wherein: 7. A gain error of an output characteristic of the in-cylinder pressure detecting means is obtained by comparing the motoring pressure detected at at least one crank angle by the in-cylinder pressure detecting means with a standard value thereof. 7. The fuel injection amount detecting device for an internal combustion engine according to claim 1, further comprising a gain error correcting unit that corrects an output characteristic of the in-cylinder pressure detecting unit by an amount corresponding to the error. 8. The fuel injection amount calculating unit calculates a fuel injection amount based on a differential pressure between the detected pressure and the reference pressure after a lapse of a predetermined time from a fuel ignition timing. 8. The fuel injection amount detecting device for an internal combustion engine according to any one of claims 1 to 7. 9. The fuel injection amount calculating means integrates a differential pressure between the detected pressure and the reference pressure during a predetermined time period from a fuel ignition timing until a predetermined time has elapsed, and calculates a fuel injection amount based on the integrated value. The fuel injection amount detection device for an internal combustion engine according to any one of claims 1 to 7, wherein 10. The fuel injection amount calculation unit calculates the fuel injection amount based on a pressure difference between the detected pressure and the reference pressure at a predetermined crank angle.
8. The fuel injection amount detection device for an internal combustion engine according to any one of claims 1 to 7. 11. The fuel injection amount calculating means integrates a differential pressure between the detected pressure and a reference pressure from a fuel ignition timing to a predetermined crank angle in a predetermined cycle, and calculates a fuel injection amount based on the integrated value. The fuel injection amount detection device for an internal combustion engine according to any one of claims 1 to 7, wherein 12. The fuel injection amount calculating means calculates a fuel injection amount based on a rate of increase in a differential pressure between the detected pressure and the reference pressure after a predetermined time has elapsed from a fuel ignition timing. A fuel injection amount detecting device for an internal combustion engine according to any one of claims 1 to 7. 13. A fuel injection device for performing a main injection for generating an engine output and a pilot injection preceding the main injection within one cycle, wherein the injection amount calculating means is configured to perform the main injection fuel at the ignition timing or before the ignition timing. The fuel injection amount detecting device for an internal combustion engine according to any one of claims 1 to 12, wherein a fuel injection amount of the pilot injection is calculated based on a pressure difference between a detected pressure and the reference pressure. 14. A fuel injection device for performing a plurality of fuel injections in one cycle, wherein the injection amount calculating means is configured to perform the Nth (N is an integer of 2 or more) injection fuel ignition timing or before the ignition timing. 13. The fuel injection for an internal combustion engine according to claim 1, wherein a fuel injection amount of the (N-1) th or earlier injection is calculated based on a pressure difference between the detected pressure and the reference pressure. Quantity detection device. 15. The fuel injection amount calculating means reduces the differential pressure between the detected pressure used for calculating the fuel injection amount and the reference pressure by the detected pressure before ignition to make it dimensionless, The fuel injection amount detecting device for an internal combustion engine according to any one of claims 1 to 14, wherein the fuel injection amount is calculated based on the fuel injection amount.