Method for determining one or more characteristic times of a fuel injection

Abstract

The present invention relates to a method for determining one or more characteristic times (NOS, NOE, NCS, NCE) of a fuel injection carried out by means of a fuel injector of an internal combustion engine, and to a computing unit and a computer program for carrying out said method. In the method according to the invention, a signal (S) is received from a sensor (36) for detecting fuel pressure changes for an injection of the fuel injector and is evaluated accordingly by means of the method.

Description

[0001] Description

[0002] title

[0003] Method for determining one or more characteristic points in time of a

[0004] The present invention relates to a method for determining one or more characteristic times of a fuel injection carried out by means of a fuel injector of an internal combustion engine, as well as a computing unit and a computer program for carrying out the method.

[0005] Background of the invention

[0006] Modern internal combustion engines feature fuel injectors that allow fuel to be injected precisely into the combustion chambers. For precise control and compliance with emissions and performance requirements of an internal combustion engine, precise metering of the fuel injection quantity is necessary. However, fuel injectors have component tolerances and are also subject to wear, resulting in tolerances in the injection quantity both between different injectors and within an individual injector over the course of its service life. Therefore, it is advantageous to record characteristic times of the injection processes, in particular the opening and closing of a nozzle needle of the fuel injectors, as accurately as possible in order to determine the actual amount of fuel injected.

[0007] If the fuel is introduced into the combustion chambers of the internal combustion engine at high pressure, fuel injectors can be used whose nozzle needle is controlled servo-hydraulically, i.e. via fuel pressure in a control chamber of the fuel injector. In this case, there is a time delay between electrical activation and the movement of the nozzle needle of the fuel injector, so that electrical control variables, such as a control current and / or a control voltage, cannot be used directly to determine the opening and closing of the nozzle needle. Therefore, additional sensors that detect, for example, the fuel pressure in the control chamber of the fuel injector, can be used with such fuel injectors.

[0008] DE 10 2021 203 572 A1, for example, shows a servo-hydraulically actuated fuel injector in which a pressure chamber is formed, which is hydraulically connected to the control chamber of the injector. The pressure chamber is sealed in a fluid-tight manner by a membrane, and a pressure sensor is arranged above the membrane, which is configured to measure the pressure in the pressure chamber.

[0009] Disclosure of the invention

[0010] According to the invention, a method for determining one or more characteristic points in time of a fuel injection performed by a fuel injector of an internal combustion engine, as well as a computing unit and a computer program for implementing the method, are proposed, having the features of the independent patent claims. Advantageous embodiments are the subject of the dependent claims and the following description.

[0011] Servo fuel injectors or fuel injectors with a switching valve are particularly suitable as fuel injectors, since direct determination of the characteristic timing from their electrical control variables is not possible. The method is particularly suitable for a fuel injector according to DE 10 2021 203 572 A1.

[0012] A sensor for detecting changes in fuel pressure is arranged in the fuel injector. The sensor is preferably arranged in the fuel injector in such a way that it can detect a pressure in a control chamber of the fuel injector. The sensor can preferably be a piezoelectric sensor, the signal from which is sent to a computing unit, which can preferably be the engine control unit. To carry out the fuel injection, the fuel injector receives a control signal that includes a start of energizing (SOE) and an end of energizing (EOE). In particular, the fuel injector can receive the control signal from a computing unit, which can preferably be the engine control unit.

[0013] The characteristic times of a fuel injection are defined as times during the opening and closing process of the fuel injector (or its nozzle needle). In particular, the needle opening start (NOS), at which the nozzle needle moves out of a nozzle needle seat, the needle opening end (NOE), at which the nozzle needle reaches an upper nozzle needle stop, the needle closing start (NCS), and the needle closing end (NCE), at which the nozzle needle reaches the nozzle needle seat again.

[0014] An analysis of the sensor signal with respect to one or more of the above-mentioned characteristic points in time allows direct conclusions to be drawn about the actual behavior of the nozzle needle and thus about the actual amount of fuel injected into the combustion chamber of the internal combustion engine. The analysis can preferably be performed across the entire operating range of the engine, i.e., at various fuel pressures and fuel injector activation times, under different external conditions (signal interference, varying contact resistance).

[0015] Specifically, a signal is received from the sensor for detecting fuel pressure changes for an injection of the fuel injector. In other words, a signal is received from the sensor that represents / reflects a pressure change in the fuel injector, preferably in the control chamber of the fuel injector, during a period of an injection (control chamber pressure signal). The received signal is subsequently filtered with a first filter and a second filter, wherein the second filter has different filter characteristics (in particular, a larger filter width) than the first filter. This means that the signal filtered with the second filter (second filtered signal) is filtered differently (in particular, more strongly) than the signal filtered with the first filter (first filtered signal).In particular, the received signal is filtered with the first and second filters in a predetermined time interval from a start of control to a predetermined time after the start of control.

[0016] Subsequently, a difference signal is formed by subtracting the first filtered signal from the second filtered signal. Furthermore, at least one characteristic time point is determined in the received signal and / or the second filtered signal and / or the difference signal.

[0017] In one embodiment, a first characteristic time is determined based on a local maximum of the difference signal. In particular, the first characteristic time can be the beginning of needle opening, at which the nozzle needle moves out of the nozzle needle seat.

[0018] In one embodiment, a second characteristic time is determined based on a local maximum of the first derivative of the second filtered signal. The second characteristic time can preferably be the needle closing point, at which the nozzle needle reaches the nozzle needle seat again.

[0019] In one embodiment, a third characteristic time is determined based on a local maximum of the received signal. The third characteristic time can preferably be the beginning of needle closing, at which the nozzle needle moves away from the upper nozzle needle stop. In one embodiment, a fourth characteristic time is determined based on a minimum of the second derivative of the received signal. The fourth characteristic time can preferably be the end of needle opening, at which the nozzle needle reaches the upper nozzle needle stop.

[0020] Based on the four characteristic points in time described above, for example, an injection rate of the fuel injection can be determined, from which, in turn, the fuel quantity introduced into the combustion engine can be determined. This can be compared with a pre-controlled fuel quantity, and the difference between the two values ​​can be used to control the fuel injection quantity. Thus, by determining the characteristic points in time, the accuracy of fuel metering can be increased at all operating points of the combustion engine and under different external conditions.

[0021] According to one embodiment, to determine the first characteristic time, preferably all local maxima of the difference signal that exceed both a first threshold and a second threshold are determined starting from a first predetermined time, and a first specific local maximum of the difference signal is determined as the first characteristic time. In other words, the first characteristic time, in particular the beginning of needle opening, can be determined based on a local maximum of the difference signal that occurs first after the first predetermined time and exceeds both a first and a second threshold.

[0022] The first predetermined time can advantageously be a time from which the first characteristic time can physically occur. In the preferred case where the first characteristic time is the beginning of needle opening, this can only occur after a time delay after the start of actuation in a preferably used servo injector. This time delay can, for example, be measured on one or more fuel injectors on the test bench, and the predetermined time can be determined based on the measured time delay. Particularly preferably, the first predetermined time can be set to approximately 300 ps after the start of actuation of the fuel injector.

[0023] According to one embodiment, it is determined that no characteristic first point in time, in particular no needle opening start, exists if the first determined local maximum of the difference signal has the highest value of all determined local maxima. In particular, in this case, it can also be determined that the nozzle needle will no longer open. The maxima in the difference signal result from several effects that are always visible in the signal, whereby these effects generally always generate a higher maximum in the difference signal than the needle opening start. If the first maximum is already the highest, it can be assumed that the needle will no longer open.

[0024] According to one embodiment, threshold values ​​are determined relative to a reference amplitude determined in the respective signal. In particular, the reference amplitude can be a signal amplitude of the first filtered signal, which temporally occurs as the first negative signal amplitude after the start of activation of the fuel injector. To determine the reference amplitude as accurately as possible, an initial level of the first filtered signal (e.g., a zero line of the signal) can be determined, for example, by averaging a predetermined number of signal values ​​before the start of activation of the fuel injector.In addition, a level of the first negative signal amplitude can be determined based on a first minimum of the first filtered signal, for example by averaging a predetermined number of signal values ​​in a predetermined range around a minimum value of the first filtered signal, which occurs as the first minimum value after the start of activation of the fuel injector. Preferably, all signal values ​​for determining the minimum of the first filtered signal are located in a time interval between the start of activation of the fuel injector and a second predetermined time. Particularly preferably, the second predetermined time can be 0.8 ms after the start of activation of the fuel injector. The reference amplitude can then be determined by forming the difference between the determined initial level and the determined first minimum of the first filtered signal.

[0025] According to one embodiment, the first threshold value is formed from a predetermined first value and the reference amplitude. The first predetermined value can preferably be determined as a function of the system pressure (rail pressure). In particular, the predetermined first value can be a first predetermined factor for a minimum height of the local maxima in the difference signal, by which the reference amplitude can be multiplied. Thus, the number of determined local maxima of the difference signal is limited to those whose height exceeds a minimum height relative to the reference amplitude.

[0026] Analogous to the first threshold, the second threshold can also be formed from a predetermined second value and the reference amplitude. The second predetermined value can advantageously also be determined as a function of the system pressure (rail pressure). In particular, the predetermined second value can be a second predetermined factor for a minimum prominence of the local maxima in the difference signal, by which the reference amplitude can be multiplied. Thus, the number of local maxima of the difference signal can be additionally limited to those whose prominence exceeds a minimum prominence relative to the reference amplitude.

[0027] The prominence of a local maximum is a value that indicates how prominent the local maximum is in terms of its height and position compared to other local maxima. To determine the prominence of a local maximum in the difference signal, a horizontal line can be drawn from the local maximum to the left and right. Where the horizontal line intersects a curve of the difference signal to the left and right (e.g. at another local maximum or at the beginning or end of the signal), the outer end points of a left and right interval can be marked. A vertical distance between a largest negative signal amplitude within the two intervals and the local maximum can then be determined as the prominence of the local maximum.

[0028] According to one embodiment, to determine the second characteristic time, all local maxima of the first derivative of the second filtered signal that exceed a third threshold are determined from the end of the actuation of the fuel injector, and a last determined local maximum of the first derivative of the second filtered signal is determined as the second characteristic time. In other words, the second characteristic time, in particular the needle closing end, can be determined based on a local maximum of the first derivative of the second filtered signal, which occurs temporally as the last maximum in the respective signal and exceeds a third threshold.

[0029] According to one embodiment, the third threshold value is formed from a predetermined third value and the reference amplitude. The third predetermined value can advantageously also be determined as a function of the system pressure (rail pressure). In particular, the predetermined third value can be a third predetermined factor for a minimum prominence of the local maxima in the first derivative of the second filtered signal, by which the reference amplitude can be multiplied. Thus, the number of local maxima of the first derivative of the second filtered signal can be limited to those whose prominence exceeds a minimum prominence relative to the reference amplitude.

[0030] According to one embodiment, the third characteristic time is determined as a global maximum of the first filtered signal from the end of the actuation of the fuel injector up to the second characteristic time. In other words, the third characteristic time, in particular the start of needle closing, can be determined based on a global maximum of the first filtered signal, which lies in time between the end of the actuation of the fuel injector and the second characteristic time, in particular the end of the needle closing. According to one embodiment, an absolute minimum of the second derivative of the first filtered signal can be determined as the fourth characteristic time, in particular as the end of the needle opening, if an absolute minimum of the first derivative of the first filtered signal lies before the third characteristic time, in particular before the end of the needle closing.In this case, the nozzle needle is opened so far that it reaches the nozzle needle stop.

[0031] The absolute minimum of the second derivative corresponds to the maximum negative curve curvature of the first filtered signal and thus to the sought fourth characteristic time point, in particular the needle opening end.

[0032] If the absolute minimum of the first derivative of the received signal occurs after the third characteristic time, in particular after the beginning of needle closing, it can be determined that there is no fourth characteristic time, in particular no end of needle opening. In this case, the nozzle needle is operating in the ballistic range, i.e., the nozzle needle is not opened far enough to reach the upper nozzle needle stop.

[0033] According to one embodiment, the fuel injector is controlled, particularly in subsequent fuel injections, taking into account the one or more determined characteristic points in time. This means that, for example, based on an actual fuel quantity determined from the characteristic points in time, the control start, the control end, and / or the control duration of a subsequent fuel injection can be adjusted in order to inject the desired / required fuel quantity at the current operating point. This enables an improvement in the accuracy of fuel metering at all operating points of the internal combustion engine under different external conditions.

[0034] A computing unit according to the invention, e.g., a control unit of an internal combustion engine, which may preferably be the engine control unit, is configured, particularly in terms of programming, to carry out a method according to the invention. Implementing a method according to the invention in the form of a computer program or computer program product with program code for carrying out all method steps is also advantageous, as this entails particularly low costs, particularly if an executing control unit is also used for other tasks and is therefore already present. Finally, a machine-readable storage medium is provided with a computer program stored thereon, as described above. Suitable storage media or data carriers for providing the computer program are, in particular, magnetic, optical, and electrical memories, such as hard disks, flash memories, EEPROMs, DVDs, and others.Downloading a program via computer networks (internet, intranet, etc.) is also possible. Such a download can be done via a wired connection or wirelessly (e.g., via a Wi-Fi network, a 3G, 4G, 5G, or 6G connection, etc.).

[0035] Further advantages and embodiments of the invention will become apparent from the description of the accompanying drawings.

[0036] The invention is illustrated schematically in the drawings using exemplary embodiments and is described below with reference to the drawings.

[0037] Short description of the drawings

[0038] Figure 1a shows schematically a longitudinal section of a fuel injector on which the method according to the invention can be used.

[0039] Figure 1 b shows schematically a current curve, a nozzle needle stroke curve, an injection rate and a control chamber pressure signal of a fuel injector according to Figure 1a.

[0040] Figure 2 schematically shows a section of the control chamber pressure signal from Figure 1b, as well as two control chamber pressure signals, each filtered with a first and a second filter. Figure 3 schematically shows a difference signal formed from the two filtered control chamber pressure signals shown in Figure 2.

[0041] Figure 4 schematically shows the first filtered control chamber pressure signal, the difference signal and a first threshold value, above which local maxima of the difference signal are taken into account to determine the beginning of needle opening.

[0042] Figure 5 shows schematically the current curve of the control as well as the first filtered control chamber pressure signal, which has a reference amplitude that can be used, for example, to determine the second threshold value shown in Figure 4.

[0043] Figure 6 schematically shows the current profile of the fuel injector as well as a first derivative of the second filtered control chamber pressure signal and a third threshold value, above which local maxima of the first derivative of the second filtered control chamber pressure signal are taken into account to determine the needle closing end.

[0044] Figure 7 shows schematically the current curve of the fuel injector and the control chamber pressure signal as well as a time interval in which a local maximum of the control chamber pressure signal is determined as the beginning of needle closing.

[0045] Figure 8 shows schematically the control chamber pressure signal and its first and second derivatives for determining the needle opening end.

[0046] Embodiment(s) of the invention

[0047] Figure 1a shows a schematic longitudinal section of a fuel injector in which the method according to the invention can be used. Only the essential areas of the fuel injector are shown. The fuel injector has a housing 1 comprising a valve body 2, a valve plate 3, a throttle plate 4, and a nozzle body 5, which are adjacent to one another in this order. A pressure chamber 8 is formed in the nozzle body 5, which can be filled with fuel under high pressure. The fuel is guided via a high-pressure bore (not shown) through the valve body 2, the valve plate 3, and the throttle plate 4 into the pressure chamber 8. A nozzle needle 10 is arranged in the pressure chamber 8 for longitudinal displacement and interacts with a nozzle needle seat 11 formed at the lower end of the pressure chamber 8.

[0048] At the lower end of the nozzle body 5, a plurality of injection openings 12 are formed, which, when the fuel injector is installed, open into a combustion chamber of an internal combustion engine. When the nozzle needle 10 rests against the nozzle needle seat 11, it closes the injection openings 12 from the pressure chamber 8. The nozzle needle 10 also defines a control chamber 18 with its end face facing away from the nozzle needle seat 11. The control chamber 18 is connected to the pressure chamber 8 via an inlet throttle (not shown in detail), so that when the nozzle needle 10 is closed, the same pressure prevails in the control chamber 18 as in the pressure chamber 8.

[0049] A switching valve 20 is arranged in a low-pressure chamber 21 formed in the valve body 2. The switching valve 20 has a magnet armature 23, an electromagnet 24, an armature spring 25, and a switching valve seat 26. The magnet armature 23 is arranged longitudinally movable in the low-pressure chamber 21 and cooperates with the switching valve seat 26 to open and close a drain bore 28 that connects the control chamber 18 to the low-pressure chamber 21. The low-pressure chamber is connected via a line to a low-pressure return line (not shown).

[0050] The pressure in the control chamber 18 can be determined by means of a pressure sensor 36, which in the example shown is arranged in the valve plate 3, but can also be located elsewhere, for example in the valve body 2. The pressure sensor 36 can, for example, be designed as a piezoelectric sensor and connected via an electrical connection cable to a control unit (not shown), which can preferably be the engine control unit.

[0051] The opening and closing of the nozzle needle 10 of the fuel injector, and thus the fuel injection, is controlled by the switching valve 20. When the electromagnet 24 of the switching valve 20 is energized, it exerts an attractive force on the magnet armature 23 and moves it out of the switching valve seat 26 against the force of the armature spring 25. When the magnet armature 23 opens the drain hole 28, fuel flows from the control chamber 18 into the low-pressure chamber 21, thereby reducing the pressure in the control chamber 18. Due to the now higher pressure in the pressure chamber 8, the nozzle needle 10 lifts off the nozzle needle seat 11 and opens the injection openings 12. When the energization of the electromagnet 24 is discontinued, the armature spring 25 pushes the magnet armature 23 back into its closed position. The fuel pressure in the control chamber 18 then rises again to the level of the pressure chamber 8 and pushes the nozzle needle 10 back into its closed position.

[0052] During the process described above within the scope of a fuel injection, the pressure in the control chamber 18 has a characteristic curve, which is illustrated by a control chamber pressure signal S of the pressure sensor 36, shown schematically in Figure 1b. The characteristic times of a needle opening start NOS, a needle opening end NOE, a needle closing start NCS, and a needle closing end NCE are marked in the signal curve S.

[0053] Figure 1 b also shows a schematic current profile l c by the electromagnet 24 with a control start SOE and control end EOE, a nozzle needle stroke curve NL of the nozzle needle 10 and an injection rate Q of the fuel injector according to Figure 1a.

[0054] When the switching valve 20 is de-energized or the electromagnet 24 is de-energized, the pressure in the control chamber 18 corresponds to the pressure in the pressure chamber 8. When the switching valve 20 is actuated or the electromagnet 24 is energized, the magnet armature 23 opens the drain bore 28 and the pressure in the control chamber 18 drops because more fuel flows out of the control chamber 18 through the drain bore 28 than flows in through the inlet throttle (cf. the signals current profile lc, nozzle needle lift NL, and control chamber pressure S after the start of control SOE in Figure 1 b). The nozzle needle 10 then moves out of the nozzle needle seat 11 in the opening direction. As long as the nozzle needle 10 is in motion, the pressure in the control chamber 18 results from a balance of forces on the nozzle needle 10. This means that the pressure in the control chamber 18 increases as long as the nozzle needle 10 moves upwards (cf. the signals nozzle needle stroke NL and control chamber pressure S during the time interval from NOS to NOE in Figure 1 b).When the nozzle needle 10 reaches its maximum lift and rests against an upper nozzle needle stop, a pressure drop occurs in the control chamber 18 corresponding to the flow through the drain bore 28 and the inlet throttle (cf. the control chamber pressure S signal directly after the time NOE in Figure 1b). When the control of the switching valve 20 is switched off, the pressure in the control chamber 18 increases until a force equilibrium prevails on the nozzle needle 10 and it moves back towards the nozzle needle seat 11 (cf. the control chamber pressure S signal after EOE shortly before NOS in Figure 1b). Due to the closing movement of the nozzle needle 10, the pressure in the control chamber 18 subsequently drops again until the nozzle needle 10 reaches the nozzle needle seat 11 (cf. the nozzle needle lift NL and control chamber pressure S signals during the time interval between NOS and NOE in Figure 1b).When the nozzle needle 10 then hits the nozzle needle seat 11, the pressure in the control chamber 18 finally rises again to the pressure in the pressure chamber 8 (cf. the signals nozzle needle stroke NL and control chamber pressure S at the time NOE).

[0055] The described relationships between the pressure in the control chamber 18 and the movements of the switching valve 20 and the nozzle needle 10 also apply in the ballistic operation of the nozzle needle 10, ie when the current supply lc of the electromagnet 24 is so short that the nozzle needle 11 does not reach the nozzle needle stop.

[0056] Figure 2 schematically shows a section of the control chamber pressure signal S from Figure 1 b in a time interval tfüt, which extends from a start of control SOE of the fuel injector to a predetermined time t xand in which the filtering of the control chamber pressure signal S is carried out with a first and second filter. The filtered control chamber pressure signals S resulting from the filtering with the first and second filters mi and S m 2 (first and second filtered (control room pressure) signals) are also shown in Figure 2. The filters used can be moving average filters, whose filter width is preferably dependent on a sampling frequency of the received control room pressure signal, and / or other filters. Based on the enlarged section of the signal curves S, S mi and S m2 in the range of a first negative signal amplitude of the control chamber pressure signal S after the start of control SOE, it becomes clear that the second filter has a larger filter width than the first filter. In other words, the second filtered control chamber pressure signal Sm2 is more heavily filtered / averaged than the first filtered control chamber pressure signal Smi, whereby its first negative signal amplitude has a lower level than that of the first control chamber pressure signal S m i. The first negative signal amplitude of the first filtered control chamber pressure signal S midiffers only insignificantly from that of the received control chamber pressure signal S. Based on the difference in the signal curve of the two filtered signals Smi and Sm2 in the area of ​​the first negative signal amplitude of the control chamber pressure signal S after the start of control SOE, a first characteristic point in time, in particular a needle opening start NOS, can be determined, as will be explained in more detail in connection with Figures 3 and 4.

[0057] Figure 3 schematically shows a difference signal Sm2-Smi plotted over a time axis t, which is formed from the two filtered control chamber pressure signals Smi and Sm2 shown in Figure 2. In contrast to Figure 2, the difference signal S shown in Figure 3 is based on m 2-S miHowever, based on a control pressure signal curve that extends over the entire fuel injection cycle. To determine the first characteristic time, in particular to determine the beginning of needle opening (NOS), all signal values ​​of the difference signal Sm2-Smi that occur before a first predetermined time h after the start of control (SOE) are first removed.

[0058] The first predetermined time h can advantageously be a time from which a needle opening start (NOS) can generally occur. With a preferred servo injector, this can only occur after a time delay after the start of control (SOE). This time delay can, for example, be measured on one or more fuel injectors on the test bench, and the predetermined time can be determined based on the measured time delay.

[0059] Figure 4 shows schematically the first filtered control chamber pressure signal S mi, the difference signal S m 2-S mi and a first threshold Sthreshoidi , above which local maxima of the difference signal S m 2-S mi to determine the beginning of needle opening (NOS). All signals are plotted against a time axis t.

[0060] To determine the needle opening start NOS, all local maxima of the difference signal S m 2-S m i, which exceed both a first threshold value and the second threshold value Sthreshoid2, are determined from the first predetermined time h. The first threshold value preferably represents a minimum height and the second threshold value Sthreshoid2 a minimum prominence that represents a local maximum of the difference signal S m 2-S mimust exceed to be determined as relevant. For the sake of clarity, only the first threshold Sthreshoidi is shown in Figure 4, and all local maxima of the difference signal S m 2-S mi from time h are marked with a circle. Based on the local maxima thus determined, the needle opening start NOS can be determined as the time of the first local maximum after time h.

[0061] Figure 5 shows schematically the current curve lc of the fuel injector and the first filtered control chamber pressure signal S mi, which has a reference amplitude Samplitude, which can be used, for example, to determine the second threshold value Sthreshoid2 shown in Figure 4. In particular, the shown reference amplitude Samplitude can be used as a basis for all threshold values ​​to be determined in the context of determining the one or more characteristic points in time. The signals shown are each shown over a time axis t. According to Figure 5, the reference amplitude Samplitude is based on a signal amplitude of the first filtered signal S mi which occurs as the first negative signal amplitude after the start of SOE of the fuel injector. The start of SOE is indicated in Figure 5 by the current waveform lc of the fuel injector. In order to determine the reference amplitude Samplitude as accurately as possible, an output level Li of the first filtered signal (e.g., a zero line of the signal S mi) for example, by averaging a predetermined number of signal values ​​before the start of control SOE of the fuel injector. In addition, a level of the first negative signal amplitude can be determined based on a first minimum l_2 of the first filtered signal S mi For example, by averaging a predetermined number of signal values ​​in a predetermined range around a minimum value of the first filtered signal Smi, which occurs as the first minimum value after the start of control SOE of the fuel injector. Preferably, all signal values ​​for determining the minimum l_2 of the first filtered signal S mi in a time interval between the start of control SOE of the fuel injector and a second predetermined time t2. The reference amplitude Samplitude can then be determined by forming a difference between the determined output level Li and the determined first minimum l_2 of the first filtered signal S mibe determined.

[0062] Figure 6 shows schematically the current curve Ic of the fuel injector and a first derivative S' m 2 of the second filtered control chamber pressure signal S m 2 and a third threshold Sthreshoid3, above which local maxima of the first derivative S' m 2 to determine the needle closing NOE. The signals shown are each plotted against a time axis t.

[0063] According to Figure 6, to determine the needle closing NOE, all local maxima of the first derivative S' m2 of the second filtered control chamber pressure signal Sm2 that exceed a third threshold value Sthreshoid3 are determined. These are each marked with a circle. Since the needle closing NOE must occur after the activation end EOE of the fuel injector, only the local maxima determined after the activation end EOE are relevant for the desired time. The activation end EOE is marked in Figure 6 using the current waveform lc of the fuel injector. The needle closing NCE is then determined as the last local maximum of the first derivative S'. m 2 of the second filtered control chamber pressure signal S m 2. In other words, the needle closing NCE can be determined based on a local maximum of the first derivative S'm2 of the second filtered control chamber pressure signal S m2, which occurs as the last maximum in the signal in question and exceeds a third threshold Sthreshoid3. The third threshold preferably represents a minimum prominence that represents a local maximum of the first derivative S' m 2 of the second filtered signal S m 2 to be determined as relevant.

[0064] Figure 7 schematically shows the current profile lc of the fuel injector and the control chamber pressure signal S as well as a time interval in which a local maximum of the control chamber pressure signal S is determined as the needle closing start NCS. The time interval shown (hatched) extends from the control end EOE of the fuel injector, which is again predetermined by the current profile lc, to the needle closing end NCE, which was already determined according to Figure 6. In this time interval, the needle closing start NCS can be easily determined based on the global maximum of the received control chamber pressure signal S shown. In other words, the needle closing start NCS can be determined based on a global maximum of the received control chamber pressure signal S, which lies in time between the control end EOE of the fuel injector and the needle closing end NCE.

[0065] Figure 8 schematically shows the received control chamber pressure signal S and its first and second derivatives S', S" for determining the needle opening end NOE. According to Figure 8, an absolute minimum of the second derivative S" of the received control chamber pressure signal S can be determined as the needle opening end NOE if an absolute minimum of the first derivative S' of the received control chamber pressure signal lies before the needle closing start NCS. In this case, the nozzle needle 10 is opened so far that it reaches the nozzle needle stop. The absolute minimum of the second derivative S" then corresponds to the maximum negative curve curvature of the received control chamber pressure signal S and thus to the desired needle opening end NOE. However, if the absolute minimum of the first derivative S' of the received control chamber pressure signal S lies after the needle closing start NCS (not shown), it can be determined that there is no needle opening end NOE.In this case, the nozzle needle 10 is operated in the ballistic range, ie the nozzle needle 10 is not opened so far that it reaches the upper nozzle needle stop.

[0066] In summary, the method according to the invention enables the determination of characteristic points in time during a fuel injection, based on which the actual injected fuel mass can be determined. This can be compared with a pre-controlled fuel quantity, and the difference between the two quantities can be used to control the fuel injection quantity. Thus, by determining the characteristic points in time, the accuracy of fuel metering can be increased at all operating points of the internal combustion engine and under different external conditions.

[0067] Furthermore, knowledge of the characteristic points in time enables stable positioning of the start of injection and / or the injection duration over the service life of the injector and an extended possibility of diagnosis in the event of implausible changes in one or more characteristic points in time during operation.

Claims

Claims 1 . A method for determining one or more characteristic times (NOS, NOE, NOS, NOE) of a fuel injection carried out by means of a fuel injector of an internal combustion engine, wherein the fuel injector receives a control signal with a control start (SOE) and a control end (EOE) for carrying out the fuel injection, and a sensor (36) for detecting fuel pressure changes is arranged in the fuel injector, comprising the steps: Receiving a signal (S) from the sensor (36) for detecting fuel pressure changes for an injection of the fuel injector; Filtering the received signal (S) with a first filter and a second filter, the second filter having a different filter characteristic than the first filter; Forming a difference signal (S m 2-S m i) by subtracting the first filtered signal (S mi) from the second filtered signal (S m 2); Determining at least one characteristic time (NOS, NOE, NOS, NOE) from the received signal (S) and / or the second filtered signal (S m 2) and / or the difference signal (S m 2-S m i).

2. The method according to claim 1, wherein determining at least one characteristic time (NOS, NOE, NOS, NOE) from the received signal (S) and / or the second filtered signal (S m 2) and / or the difference signal (Sm2-Smi) comprises: a) determining a first characteristic time (NOS) based on a local maximum of the difference signal (S m 2-S m i); Forming a first derivative (S' m 2) of the second filtered signal (S m2 ); and / or b) determining a second characteristic time (NOE) based on a local maximum of the first derivative (S' m2) of the second filtered signal (S m 2); and / or c) determining a third characteristic time (NCS) based on a local maximum of the received signal (S); and / or d) forming a second derivative (S") of the received signal (S); determining a fourth characteristic time (NOC) based on a minimum of the second derivative (S") of the received signal (S). Method according to claim 2, wherein, to determine the first characteristic time (NOS), all local maxima of the difference signal (S m 2-S m i) which exceed both a first threshold value and a second threshold value (Sthreshoid2) are determined from a first predetermined time (h), and a first determined local maximum of the difference signal (S m 2-S mi) is determined as the first characteristic time (NOS). Method according to claim 3, wherein it is determined that no characteristic first time (NOS) exists if the first determined local maximum of the difference signal (S m 2-S m i) has the highest value of all determined local maxima. Method according to claim 3 or 4, wherein the first threshold value is formed from a predetermined first value and a reference amplitude (Samplitude), and the second threshold value (Sthreshoid2) is formed from a predetermined second value and the reference amplitude (Samplitude). Method according to one of claims 2 to 5, wherein to determine the second characteristic time (NOE) all local maxima of the first derivative (S' m 2) of the second filtered signal (S m 2) that exceed a third threshold (Sthreshoids) are determined from the end of operation (EOE) of the fuel injector, and a last determined local maximum of the first derivative (S' m 2) of the second filtered signal (S m 2) is determined as the second characteristic time point (NCE).

7. The method according to claim 6, wherein the third threshold value (Sthreshoids) is formed from a predetermined third value and the reference amplitude (Samplitude).

8. The method according to any one of claims 2 to 7, wherein a global maximum of the received signal (S) from the activation end (EOE) of the fuel injector up to the second characteristic time (NCE) is determined as the third characteristic time (NCS).

9. The method according to any one of claims 2 to 8, wherein an absolute minimum of the second derivative (S") of the received signal (S) is determined as the fourth characteristic time (NOE) if an absolute minimum of the first derivative (S') of the received signal (S) lies before the third characteristic time (NCS), and it is determined that no fourth characteristic time (NOE) is present if the absolute minimum of the first derivative (S') of the received signal lies after the third characteristic time (NCS).

10. Method according to one of the preceding claims, wherein the fuel injector is controlled, in particular in subsequent fuel injections, taking into account the at least one determined characteristic time (NOS, NOC, NCS, NCE). 11 . Computing unit configured to carry out a method according to one of the preceding claims.

12. A computer program which causes a computing unit to carry out all method steps of a method according to one of claims 1 to 10 when executed on the computing unit.

13. A machine-readable storage medium having a computer program according to claim 12 stored thereon.

Images