JP2005090503A - Internal combustion engine driving method, computer program for internal combustion engine control device, data carrier and injection device for internal combustion engine - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve a known internal combustion engine driving method, a computer program corresponding thereto, and an injection device for a direct injection internal combustion engine for accurately injecting a target amount of fuel in consideration of pressure in a combustion chamber which is changed with a given change in rail pressure. <P>SOLUTION: In accordance with at least one rail pressure value detected right before an injection process and/or during the injection process or other parameters, if necessary, a value for an injection time calculated before is corrected at least once before operation to form a corrected value ti_K for a time for injection into a cylinder n under way. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a method for driving an internal combustion engine, for example in a vehicle with N cylinders into which fuel is directly injected, and a corresponding computer program. The invention also relates to a data carrier comprising the aforementioned computer program and an injection device for an internal combustion engine for carrying out the aforementioned method.

  A control device for a direct injection internal combustion engine is known from German Offenlegungsschrift 19645715. The control device described here calculates the basic injection time for a given cylinder of the internal combustion engine. This basic injection time is calculated so that a desired amount of injected fuel can be obtained under pressure characteristics that are considered to be steady in the fuel accumulator (hereinafter also referred to as rail) and the combustion chamber of the cylinder. In practice, however, the pressure characteristics are not steady, and tend to fluctuate particularly in the combustion chamber of the cylinder. Therefore, in the above-mentioned German Offenlegungsschrift 19645715, the previously calculated basic injection time is corrected with the calculated crank angle position depending on the pressure characteristics in the combustion chamber of the cylinder, and the timer for outputting the injection time is corrected in this way. It has been proposed to adjust to the value of the injection time. However, in the injection time correction calculation proposed here, the rail pressure is still considered to be stationary (that is, constant).

  Actually, not only the pressure in the combustion chamber but also the pressure in the rail fluctuates with time. For this reason, if calculation is performed based on steady pressure characteristics, there is always a risk of error. Moreover, since the pressure characteristics in the rail dynamically change depending on a large number of influences, the previous calculation is extremely complicated and can be performed only inaccurately.

  In view of these facts, German Patent No. 19726756 proposes another method and system for driving an internal combustion engine. Here, the rail pressure is detected while the injection process is in progress, and the amount of fuel injected from the start of injection is continuously calculated based on the detected rail pressure. If the amount of fuel injected during the current injection exceeds the set target value, the injection is interrupted.

However, even in the invention of the above-mentioned German Patent No. 19726756, the calculation of the injected fuel amount is very complicated, and the valve delay time in the ongoing injection process must be calculated separately from the opening and closing of the valve. There are drawbacks. Furthermore, this calculation must be performed at a very high time frequency, ie much more frequently than the 1 ms pattern. Otherwise, when the injection time is short (for example, 0.5 ms), a serious error occurs in the amount of injected fuel, or an error occurs in the entire process. Also, the method of the present invention does not take into account changing combustion chamber pressures.
German Publication No. 19645715 German Patent No. 19726756

  Therefore, starting from the prior art as described above, the present invention is known in the art so that the target fuel amount can be accurately injected in consideration of the combustion chamber pressure that changes with any change in rail pressure. To improve the driving method of the internal combustion engine, the corresponding computer program and the injection device for the direct injection internal combustion period.

  The task is to correct the pre-calculated injection time value at least once before operation based on at least one rail pressure value and / or possibly other parameters detected immediately before and / or during the injection process. This is solved by the configuration in which the correction value ti_K of the injection time of the cylinder n with respect to the ongoing injection is formed.

  Advantageously, the method according to the invention not only detects the rail pressure once just before the injection process, but also repeats it during the precalculated injection time, eg as often as possible in a narrow time pattern. Based on the continuously updated rail pressure value, the temporal variation in rail pressure and the variation in the amount of injected fuel resulting therefrom are captured very accurately and taken into account in the correction calculation for the currently ongoing injection time. As a result, the total amount of fuel injected into the cylinder is finally output very accurately.

  According to the above-described embodiments of the method of the present invention, any modulation or gradient setting of the rail pressure is possible. The rail pressure corresponding to the maximum pumping force of the high-pressure pump is quickly formed even at the restart, and the mixture preparation is improved and the raw emissions are reduced. Deviations from the desired or adjusted lambda values are also minimized if a rapid increase or decrease in rail pressure is required due to positive or negative load changes during the operation of the internal combustion engine. This is because the injection time is corrected so that the target fuel amount is always injected by the present invention. Pump pumping tolerances or other influences are eliminated by detecting rail pressure during the course of the injection process.

  Advantageously, the first correction of the pre-calculated injection time is made immediately before the start of the injection process, not after the injection process starts. This ensures that correction is performed even when the injection time is shorter than the rail pressure detection time pattern.

  Advantageously, the detection of the rail pressure takes place in a fixed time pattern, for example every 1 ms, before and / or during the injection process. It is also recommended that the injection time correction calculation be performed immediately after this detection.

  Advantageously, the pre-calculation of the injection time is performed at a trigger time that depends on a fixed crank angle pattern rather than the aforementioned time pattern. This trigger mark depends on the crank angle position preceding the start of injection by various time widths.

  Also advantageously, the pre-calculation of the injection time is based on the rail pressure value last detected in a predetermined time pattern before the trigger time synchronized with the angle, and takes into account the pre-calculated combustion chamber pressure value. This is done at the crank angle position.

  The correction calculation of the injection time becomes more accurate by taking into account the correction calculation at the time of rail pressure detection, not only the rail pressure but also the combustion chamber pressure of the cylinder in which injection is being performed at that time.

  Advantageously, the present invention provides a technique or mathematical algorithm that allows simple and quick injection time correction calculations. Since the algorithm is an additional calculation item for normal injection timing calculation, it may be omitted and simplified if the internal combustion engine speed is high and an operation time problem may occur. . Each time the last calculated injection time, i.e. the latest correction value of the injection time, is applied, the injection process is interrupted. This advantageously allows the output timer to be continuously updated with the latest correction value of the injection time each time during the calculation time pattern.

  Advantageously, for the pre-calculation and correction calculations of the injection time, rather than using true pressure characteristics in the rail with high frequency fluctuations, for example rail pressure values attenuated by low-pass filtering are used. .

  The above-mentioned problem is further solved by an injection device, a computer program, and a data carrier comprising the computer program for carrying out the method described above. The advantages of these solutions correspond to the advantages obtained from the method described above.

  The present invention will be described in detail below with reference to the drawings in accordance with various embodiments.

  FIG. 1 shows an injection device 100 of the present invention for an internal combustion engine having four cylinders (not shown). This injection device has a fuel accumulator (hereinafter referred to as rail) 110 that accumulates fuel and injects fuel into the combustion chamber of the cylinder via four injection valves 120. The injection device 100 further includes a high-pressure pump 130 that pumps fuel to the rail 110 at a high pressure. Fuel is supplied to the high-pressure pump 130 via the electric fuel pump 105. Inside the high-pressure pump 130, the fuel supplied from the electric fuel pump 105 is first supplied to the damper element 134. From there, the fuel reaches the pumping element 138 via the first check valve 136 when the fuel amount adjustment valve 132 as an actuator is in an ON state, that is, when the valve is blocked. The pumping element 138 has a piston 139, which is driven via a camshaft 160 of an internal combustion engine. In the state shown in FIG. 1, the piston 139 is in the lowest position, and the pumping element 138 has a chamber 137 for containing fuel supplied via the check valve 136. As the piston 139 of FIG. 1 moves upward based on the corresponding rotation of the camshaft 160 at a later point in time, the volume of the chamber 137 decreases and the fuel contained therein is compressed. The compressed fuel is supplied to the rail 110 via the second check valve 136. There, the fuel is stored under high pressure.

  The pressure of the fuel accumulated in the rail, that is, the rail pressure, is closed-loop controlled to a target pressure set via the injector 100. For this purpose, the injection device 100 has a control device 150. The actual rail pressure is supplied to the control device as an actual amount from the rail pressure sensors 140 and 140 '. When a control deviation between the set target pressure and the detected actual pressure is detected, the control device drives the fuel amount adjusting valve 132 in the high-pressure pump 130 and closes the loop so that the rail pressure becomes a desired target value. Control. To be precise, this closed-loop control turns off the fuel adjustment valve 132 between the pumping element 138 and the damper 134 shown in FIG. This is done by opening the connection pipe 133. At this time, the open connection pipe 133 allows the fuel flow from the chamber 137 to flow to the damper element 134, thereby achieving a desired pressure drop. On the contrary, when it is desired to increase the rail pressure, the fuel amount adjustment valve 132 is turned on correspondingly, and the connecting pipe 133 is closed.

  The opening of the injection valve 120 triggers a high-frequency pressure pulsation in the rail 110, which is superimposed on a slow pressure fluctuation. If this pressure pulsating flow is detected by the pressure sensors 140 and 140 ′ and supplied to the control device 150 as a component of the detected actual pressure, an erroneous measurement result is generated, which is not desirable for closed-loop control of the rail pressure. However, the time constant of this pulsating flow is much smaller than the time constant of the pressure increase due to the pressure stroke of the piston 139 in the pressure feeding element 138 of the high pressure pump 130 or the time constant of the pressure drop due to the amount of injected fuel. It can be removed through suitable filter means. As the first means, a low-pass filter 142 in the form of an RC element for the rail pressure sensor 140 is provided on the input side of the control device 150. This low-pass filter has a corresponding time constant for removing high frequency components.

  The frequency of the disturbing pressure pulsation is determined by the rail resonance frequency, in particular the rail geometry and volume. Rails are often configured large to achieve a pressure accumulation effect and have a low resonant frequency. As a result, the time constant of the disturbing pressure pulsating flow increases and is offset toward the time constant of the pressure change caused by the pumping stroke of the high-pressure pump 130 and the time constant caused by the injection. Therefore, instead of or in addition to the first means described above, a hydraulic low-pass filter that removes the high frequency of the disturbing pressure pulsating flow is realized as the second means. This second means is indicated by a broken line in FIG. By this second means, the rail is connected to the fuel vapor intermediate accumulator 146 serving as an additional volume via the throttle 144, so that the inherent rail volume is kept to a minimum. The rail pressure sensor 140 ′ is not directly connected to the rail 110 but is connected to the fuel vapor intermediate accumulator 146 serving as an additional volume portion. The additional volume advantageously extracts rail pressure characteristics that are significantly damped over the actual amount used in the controller 150. Such a device is called a hydraulic low-pass filter.

  A method of driving the internal combustion engine of the present invention using the injection device 100 shown in FIG. 1 will be described in detail below with reference to FIGS.

  FIG. 2 shows various processes performed in parallel in time. First, reference symbols t1_1. . . For example, 4 represents the injection timing of an internal combustion engine having four cylinders (not shown). The injection timing is related to the trigger mark tr shown above. The trigger mark corresponds to each cylinder 1 to 4 as well as the injection timing, and is obtained from the increment signal and the reference mark of the crankshaft sensor. The injection timing is related to a crankshaft reference mark BM representing an angular position of 60 ° in front of the top dead center OT of the cylinders 1 and 3, for example. The trigger mark sends out a calculation pattern for each cylinder for calculating the injection time of the present invention. This will be described in detail below.

  FIG. 2 further shows the stroke curve of the HDP stroke of the high-pressure pump 130 of FIG. 1 via the reference mark BM. This represents a state in which the piston 139 of the pumping element 138 is driven via the camshaft 160 of the internal combustion engine. Above the stroke curve, an adjustment signal MSV sent from the control device 150 for driving the fuel amount adjustment valve 132 is shown. A high level of this signal indicates that the fuel amount adjustment valve is on, that is, the connection pipe 133 is closed. Conversely, a low level of this signal indicates that the fuel amount adjustment valve is in an OFF state, that is, the connection pipe 133 is open. The opening and closing of the fuel amount adjustment valve 132 is set as a response to the rail pressure p_R detected by the control device, and control is performed so that the target value p_R_soll is set.

  Above the adjustment signal MSV for the fuel amount adjustment valve 132 in FIG. 2, the rail pressure p_R when the number of revolutions of the internal combustion engine is high and the flow rate of fuel is large is shown. The illustrated rail pressure characteristic is related to the adjustment signal MSV for the fuel amount adjustment valve 132, and from this, the operation of the above-mentioned closed loop control can be seen. During the stroke of the piston of the high-pressure pump, the pressure rises until the set rail pressure target value p_R_soll is achieved in the ON state of the fuel adjustment valve, that is, in the state where the connection pipe 133 is closed. I understand that. Thereafter, the fuel amount adjusting valve 132 is driven by the control device 150, that is, the connection pipe 133 is opened, and unnecessary fuel returns from the chamber 137 to the low-pressure circuit, particularly the damper element 134. The pressure characteristics shown in FIGS. 2 to 4 are theoretical values and do not consider the damping action on the rail. Actually, these pressure characteristic lines are not broken and are sinusoidal.

  As shown in FIG. 2, the fuel adjustment valve is turned on under high rotational speed, ie the connecting line 133 is closed, eg long during the piston stroke phase at the injection time t1_1 for the cylinder 1 When injection is performed, the rail pressure p_R increases to 3) more slowly than 1) when there is no injection or when the injection time is short. When the target rail pressure p_R_soll is achieved and the fuel amount adjustment valve 132 is turned off at a predetermined piston stroke position, the rail pressure is maintained over the current injection ti_1 from this point in time and in the subsequent injection t1_2 of the cylinder 2 as shown in FIG. The characteristic 4) shown in FIG. This is because the suction is performed in this phase of the piston 139 and the fuel is not pumped into the rail 110.

The calculation of the basic injection time ti is performed for each trigger mark tr. This is indicated in FIG. 2 by a vertical dashed trigger mark and a downward pointing arrow pointing thereto. The injection time is calculated based on the currently used rail pressure value, that is, the rail pressure value measured at the end of a predetermined time pattern. The rail pressure is preferably detected every ¹ ms in accordance with 6000 min ⁻¹ of the internal combustion engine. Specifically, the rail pressure characteristic shown in FIG. 2 is calculated based on the rail pressure value at which the injection time ti_2 for the cylinder 2 is currently used at the trigger mark tr_4, that is, the rail pressure value p_R_R2, for example. Represents. Accordingly, the injection time for the cylinder 3 is calculated based on the rail pressure value p_R_R3. As can be seen from FIG. 2, the rail pressure value used for this calculation does not coincide with the intermediate pressure value p_R_Mi for the injection time ti_i for the cylinder i.

  FIG. 3 basically shows the same physical quantity characteristics as in FIG. 2, but this is not for an internal combustion engine at high speed and high fuel flow, but for the starting phase of the internal combustion engine. . The same physical symbols as in FIG. 2 are given the same reference symbols. See the description of FIG. 2 for the basics of the time characteristics shown.

In addition to the already known characteristics, the lower part of FIG. 3 shows the engine speed n _{MOT in} the starting phase of the internal combustion engine in synchronism with the angular planes of other physical quantities. At the beginning of the starting phase, a relatively low rotational speed is first generated. Therefore, the low rail pressure p_ND of the low pressure system is generated here as shown in the upper and lower parts of FIG. During the start-up phase, especially when the rail pressure is still low, large errors may occur when calculating the injection time in the pressure formation at restart. In FIG. 3, for example, this is as follows. The intermediate rail pressure p_R_M3 during the cylinder 3 injection corresponds relatively well to the detected rail pressure value p_R_R3, but in the next injection, which is the final stage of the piston stroke, p_R_M4 is significantly higher than the calculated pressure p_R_R4. End up.

  As shown in FIGS. 2 and 3, during the course of the injection process, especially in the special situation where strong fluctuations in the rail pressure occur as shown in FIG. 3, the injection time is corrected to correspond to the true rail pressure. Must. Otherwise, the desired amount of fuel to be injected into the cylinder will not be accurately metered. For this purpose, two approaches are proposed according to the invention. First, these methods are derived mathematically.

  The target fuel amount qk_soll at the time of calculation of ti at the tr mark is calculated from the differential pressure Δp_RB_R between the rail pressure and the combustion chamber pressure existing at the time of calculation.

と表される。ここから噴射時間は

It is expressed. From here the injection time is

となる。ここでＫ＿ＥＶは噴射弁定数であり、ｔｉ＿Ｒはｔｒマークでのｔｉ計算時に計算された噴射時間である。なお前述の通り
Δｐ＿ＲＢ＿Ｒ＝ｐ＿Ｒ＿Ｒ−ｐ＿Ｂ＿Ｒ （３）
と表され、ここでｐ＿Ｒ＿Ｒはｔｉ計算時点でのレール圧であり、ｐ＿Ｂ＿Ｒはｔｉ計算時点での燃焼室圧力である。

It becomes. Here, K_EV is an injection valve constant, and ti_R is an injection time calculated at the time of calculating ti at the tr mark. As described above, Δp_RB_R = p_R_R−p_B_R (3)
Where p_R_R is the rail pressure at the time of ti calculation, and p_B_R is the combustion chamber pressure at the time of ti calculation.

  When injection is performed in the intake phase, the combustion chamber pressure is approximately equal to atmospheric pressure and can be considered constant. Furthermore, at the time of calculating ti at the tr mark, the fuel pressure is equal to the last detected pressure, and can be regarded as being constant over the entire injection time.

  The rail pressure changes as described with reference to FIGS. 2 and 3 until the time of injection and during the injection process. Therefore, the changing rail pressure is advantageously sampled every 1 ms, and the latest sampling value obtained in this way is used for the correction calculation of the injection time. Based on the rail pressure value at that time, not only the correction calculation is performed once immediately before the start of injection, but also the last calculated value during the injection process until the updated injection time has finally passed Continue to correct using. Therefore, according to the present invention, a simple correction formula is derived.

  It should be noted when calculating the correction formula that how much fuel quantity qK is injected from the rail under changing pressure. The amount of fuel is

のように計算される。ここでＫ＿ＥＶは噴射弁定数であり、Δｐ＿ＲＢ（ｔ）は時点ｔでのレール圧と燃焼室圧力との差圧である。

It is calculated as follows. Here, K_EV is an injection valve constant, and Δp_RB (t) is a differential pressure between the rail pressure and the combustion chamber pressure at time t.

  The above formula is digitized as a total formula using a sampling pattern Δt of 1 ms, for example. Ie

が成り立つ。噴射時間が近似的にΔｔの整数倍であるとすると、ｔｉ＝ｋ・ΔｔまたはΔｔ＝ｔｉ／ｋより

Holds. Assuming that the injection time is approximately an integer multiple of Δt, ti = k · Δt or Δt = ti / k

が得られる。ここでｋは１回の噴射中のサンプリング回数である。したがって目標燃料量ｑＫ＿ｓｏｌｌおよびこれに必要な補正量ｔｉ＿Ｋは

Is obtained. Here, k is the number of samplings during one injection. Therefore, the target fuel amount qK_soll and the correction amount ti_K necessary for this are

となる。

It becomes.

  If qK_soll is eliminated from the above-described equations (1) and (7), the injection time correction value ti_K is

となる。ここでＫＦは補正された実際の噴射時間ｔｉ＿Ｋとトリガマークｔｒで計算された元の噴射時間ｔｉ＿Ｒとの比を表す補正係数である。

It becomes. Here, KF is a correction coefficient representing the ratio between the corrected actual injection time ti_K and the original injection time ti_R calculated by the trigger mark tr.

  FIG. 4 shows how the injection time is corrected. In particular, it can be seen that the original injection time ti_R1 (Δp_RB_R1) obtained based on the differential pressure Δp_RB_R1 between the rail pressure and the combustion pressure obtained immediately before the trigger mark tr_3 for the cylinder 1 is corrected later. That's what it means. The calculated original injection time is shown as the second horizontal bar from the top in FIG. The correction value finally becomes a shorter injection time ti_K1. This is shown as the upper horizontal bar.

  FIG. 4 further shows that pressure detection in a 1 ms time pattern is performed without synchronizing with the trigger time point. The injection start time depends on a fixed crank angle mark. For example, it is not sufficient to perform the correction calculation after the injection is started so that the correction can be performed in the same manner as described above even with an extremely short injection time of 0.5 ms. In this case, the injection has already ended when the correction calculation is performed. Therefore, it is recommended to obtain the initial value of the injection time and the first correction value of the initial value before starting the original injection.

  The calculation of the correction equation (8) is advantageously performed such that the numerator terms other than the coefficient k are first obtained when calculating ti_R in the tr pattern. To create a total denominator, create individual quantities and add new pressure roots each time. The numerator is then divided by the new sum and the result is multiplied by the resulting index k. In this way, ti_K corrected for every 1 ms is advantageously obtained with a small calculation cost.

  To correct the ti output timer for driving the injection valve 120 according to the injection time calculated in FIG. 3, the injection time ti_K newly calculated every 1 ms as shown in the first column V1 of FIG. Or, instead, continue the correction during the injection process, or instead of the 1 ms pattern, when the current injection time ti_A is less than the correction value ti_K by the time increment δt_TK required for the output timer correction Correct with. See the second column V2 in FIG. 4 for the latter.

  Thus, in the ongoing injection process ti_R1 for the cylinder 1, an integral of the root value of the sampled pressure value is formed corresponding to the step curve along the pressure characteristic Δp_RB shown in FIG. The time is corrected to an appropriate value ti_K1. In FIG. 4, ti is shortened due to the pressure increase. When the pressure drop occurs, ti is extended.

  In order to reduce the error due to pressure sampling, when forming the root of the denominator of equation (8), it is not necessary to simply use a new sampling value, but to correspond to the step curve in the right part of the pressure characteristic of FIG. It is preferable to always add 1/2 of the difference between the new sampling value and the old sampling value. For example, when j = 5, (Δp5-4) / 2 = (p (j = 5) −p (j = 4)) / 2 is used. As a result, the sampling step curves are averaged to better approximate the true pressure characteristics, and a slower calculation pattern, for example, a 2 ms pattern can be selected. Ie

が用いられる。

Is used.

  The main purpose of the correction calculation according to equation (8) is to create pressure as quickly as possible during start-up. Since start-up and restart are usually performed in uniform mode injection, the combustion chamber pressure is constant, and Δp_RB can be regarded as the rail pressure minus atmospheric pressure.

  The correction calculation may be performed after a complete fuel pressure is formed in the “stratified charge” mode and the “uniform split” mode. At this time, instead of a constant combustion chamber pressure, the combustion chamber pressure at the time of injection is used. The combustion chamber pressure is the basis for the corresponding injection angle and is also used when calculating ti_R at the tr mark.

  Alternatively, the instantaneous combustion chamber pressure may be calculated from the compression ratio and crankshaft position in the same 1 ms time pattern as the rail pressure detection, or measured via a combustion chamber pressure sensor. Although this requires a small additional calculation cost, the pressure characteristics of the combustion chamber can be taken into account more accurately than when the injection time is pre-calculated.

  If the reserve operation time of the microcontroller used is limited, the above correction calculation can be cut off above the rotation speed threshold.

It is a figure which shows the structure of the injection apparatus of this invention. It is a 1st time chart which shows operation | movement of an injection apparatus. It is a 2nd time chart which shows operation | movement of the injection device in the starting process of an internal combustion engine. FIG. 4 is a third time diagram representing the method of the present invention.

Explanation of symbols

100 Injection Device 105 Electric Fuel Pump 110 Fuel Accumulator (Rail)
DESCRIPTION OF SYMBOLS 120 Injection valve 130 High pressure pump 132 Fuel quantity adjustment valve 133 Connection pipe line 134 Damper 136 Check valve 137 Chamber 138 Pressure feed element 139 Piston 140,140 'Pressure sensor 142 Low pass filter 144 Restriction part 146 Fuel vapor intermediate storage 150 Control device 160 Camshaft MSV control signal

Claims (19)

For example, to drive an internal combustion engine of a vehicle with N cylinders into which fuel is directly injected, detecting the rail pressure before the injection process to be triggered, and detecting the detected rail pressure and possibly other Precalculating the injection time ti of the cylinder n (n = 1... N) for the injection process to be triggered based on the parameters,
Based on at least one rail pressure value and / or possibly other parameters detected immediately before and / or during the injection process, the pre-calculated injection time value is corrected at least once before operation and in progress A method of driving an internal combustion engine, wherein a correction value ti_K of an injection time of a cylinder n with respect to the injection of   The method according to claim 1, wherein the detection of the rail pressure is carried out in a fixed time pattern, for example every 1 ms, before and / or during the injection process.   Pre-calculate the injection time ti of the cylinder n at each trigger time tr set by a fixed crank angle pattern, the pre-calculation being dependent on the injection angle position preceding the injection into the cylinder n by various time widths. 3. The method according to claim 1, wherein the method is preferably based on the last detected rail pressure value before the trigger time.   4. The method according to claim 1, wherein the injection time correction calculation performed immediately before the start of injection and / or during the injection process is performed immediately after the rail pressure is detected according to a predetermined time pattern. .   The method according to claim 1, wherein the rail pressure signal used for the injection time pre-calculation and / or correction calculation is a low-pass filtered rail pressure signal.   The method according to any one of claims 1 to 5, wherein a correction value ti_K for the injection time is calculated based on a rail pressure value obtained by subtracting a pressure value as a second parameter representing the combustion chamber pressure.   The method according to claim 6, wherein the pressure of the combustion chamber in the uniform mode of the internal combustion engine is constant at the magnitude of the atmospheric pressure.   The combustion chamber pressure in the injection mode in which the injection is performed in the compression stroke is calculated or estimated from the crankshaft position and the compression ratio in the same time pattern as the rail pressure detection, or is measured through the combustion chamber pressure sensor. 6. The method according to 6. Formula for correction value ti_K of injection time

Where ti_K is a correction value for the injection time, ti_R is the injection time previously calculated at the trigger time synchronized with the angle, and Δp_RB_R is the difference between the rail pressure and the combustion chamber pressure at the trigger time. Pressure, k is the number of samplings so far during one injection process, j is a time-synchronized sampling time immediately before and / or during the injection process, and Δp_RB (j) is a sampling time j The method according to claim 1, wherein the pressure is a differential pressure between the rail pressure and the combustion chamber pressure. Formula for correction value ti_K of injection time

Where ti_K is a correction value for the injection time, ti_R is the injection time previously calculated at the trigger time synchronized with the angle, and Δp_RB_R is the difference between the rail pressure and the combustion chamber pressure at the trigger time. Pressure, k is the number of samplings so far during one injection process, j is a time-synchronized sampling time immediately before and / or during the injection process, and Δp_RB (j) is a sampling time j The method according to claim 1, wherein p_R (j) is a rail pressure value at a sampling time j.   11. The ti output timer is updated with the latest value of the injection time each time the injection time correction calculation is performed until the injection time continuously updated by the repetition of correction ends. The method according to claim 1.   The method according to any one of claims 1 to 10, wherein the ti output timer is updated when a time obtained by subtracting a time taken to correct the output timer from an injection time continuously updated by the correction calculation has elapsed.   The method according to claim 1, wherein no correction calculation is performed above the rotation speed threshold value, or a plurality of correction calculations are converted into a single correction calculation. 14. A computer program for a control device for an internal combustion engine, comprising a program code for executing the driving method for the internal combustion engine according to claim 1. 15. A data carrier comprising a computer program for a control device for an internal combustion engine according to claim 14. N cylinders,
A rail or fuel accumulator (110) that accumulates fuel and delivers the fuel via an injection valve (120) to the combustion chamber of the internal combustion engine;
A high pressure pump (130) for pumping fuel to the fuel accumulator at high pressure;
A pressure sensor (140, 140 ′) for detecting the actual fuel pressure in the fuel accumulator;
The actual fuel pressure in the fuel accumulator detected by the pressure sensor is closed-loop controlled to the rail pressure value set by driving the actuator (132) of the high-pressure pump, and the cylinder n (n = 1... N of the internal combustion engine). A control device (150) for pre-calculating the injection time ti for
In an injection device (100) for an internal combustion engine,
Based on at least one rail pressure value detected and / or possibly other parameters immediately before and / or during the injection process, the controller (150) may determine a pre-calculated injection time value at least prior to operation. An injection apparatus for an internal combustion engine, which is configured to correct once and form a correction value ti_K of an injection time for the injection in progress in the cylinder n.   17. Apparatus according to claim 16, wherein the controller (150) is arranged to execute the instructions of a computer program for the controller of an internal combustion engine according to claim 14.   18. A filter for attenuating rail pressure characteristics, for example a low-pass filter (142) in the form of an RC element, is provided on the input side of the control device in order to evaluate the signal of the pressure sensor (140). The apparatus of any one of Claims.   A hydraulic low-pass filter comprising a throttle section (144) and a fuel vapor intermediate accumulator (146) is assigned to the fuel accumulator (110) to attenuate high frequency rail pressure fluctuations, and the damped rail pressure characteristics. 18. A device according to any one of claims 15 to 17, wherein a pressure sensor (140 ') for detecting is connected to the fuel vapor intermediate accumulator (146).

Images