Method for controlling an injector, and controller

Abstract

The invention relates to a method for actuating an injector (1), in particular a fuel injector, in which a reciprocating movement of a nozzle needle (2) for opening and closing at least one injection opening (3) is controlled by a pressure in a control chamber (4), the pressure in the control chamber (4) is measured by means of a pressure sensor (5) integrated in the injector (1), information characterizing the reciprocating movement of the nozzle needle (3), in particular time data and flow data, being derived from a sensor signal of the pressure sensor (5). And using the information to correct an injection rate model based on previously measured time data and flow rate data.

Description

Technical Field

[0001] The invention relates to a method for actuating an injector, in particular a fuel injector for injecting fuel under high pressure into a combustion chamber of an internal combustion engine.

[0002] Furthermore, the invention relates to a control unit for carrying out the method. Background Art

[0003] From the prior art, fuel injectors are known that have a nozzle needle that is arranged longitudinally displaceably in a pressure chamber for opening and closing at least one injection opening, wherein the movement of the nozzle needle is controlled servohydraulically, i.e., by the pressure in a control chamber that applies a hydraulic closing force to the nozzle needle. The pressure in the control chamber is controlled by a control valve that adjusts the pressure in the control chamber electromagnetically or by means of a piezoelectric actuator. The control valve can be actuated very precisely by a controller. However, there is a time delay between the control current and the actual movement of the nozzle needle. Therefore, for precise control, it is important to know the exact point in time of the nozzle needle movement and the start of injection in order to readjust the current for actuating the control valve.

[0004] For example, the movement of the nozzle needle can be detected via the pressure curve in the control chamber. For this purpose, DE 10 2015 207307 A1 proposes a fuel injector with a servohydraulically controlled nozzle needle and a pressure sensor. With the help of the pressure sensor, the pressure change in the control chamber or in a pressure chamber hydraulically connected to the control chamber can be detected. By evaluating the pressure signal of the pressure sensor, the exact time of the opening and closing of the nozzle needle can be determined. As a result, the control signal for actuating the control valve can be readjusted if necessary.

[0005] The time curve of the pressure in the control chamber of a servohydraulic injector provides precise information about the movement of the nozzle needle. By using algorithms for signal analysis, characteristic times of the injection, such as the start of injection, the end of injection or the needle opening time, can be determined. In these algorithms, a combination of digital filtering, threshold methods and numerical derivations is usually used.

[0006] In the operation of internal combustion engines, injection rate models are often used to regulate the injected fuel quantity with greater accuracy and to optimize combustion parameters. In this context, 0D models (zero-dimensional models) are preferred over 1D models (one-dimensional models) based on flow simulations due to the lower computing power. These models often use greatly simplified diagrams of the injection rate curve, e.g. Fig.11 However, there is a deviation between the measured actual curve (solid line) and the model (dashed line), so that the prediction of the injection quantity by the model is inaccurate.

[0007] Therefore, it has been proposed to estimate the injection quantity using the absolute pressure or the pressure difference of the sensor signal of the pressure sensor integrated into the injector in order to increase the accuracy. However, due to wear processes, the sensor signal amplitude of the pressure sensor changes during the service life of the injector, thereby affecting the accuracy of the estimation due to this change. Furthermore, if the maximum injection rate is calculated with the aid of a fixed linear correlation with the pressure drop, this correlation changes over time due to the drift / wear of the pressure sensor, which leads to systematic errors in the injection quantity estimation. Summary of the invention

[0008] The present invention is therefore directed to the object of increasing the accuracy of a model-based injection quantity estimation.

[0009] In order to achieve this object, a method is proposed having the features of claim 1. Advantageous developments of the invention are revealed in the dependent claims. Furthermore, a control unit is specified for carrying out the method.

[0010] A method for controlling an injector, in particular a fuel injector, is proposed, wherein a reciprocating movement of a nozzle needle is controlled by means of a pressure in a control chamber for opening and closing at least one injection opening, and the pressure in the control chamber is measured by means of a pressure sensor integrated in the injector. Information characteristic of the reciprocating movement of the nozzle needle, in particular time data and flow data, is derived from a sensor signal of the pressure sensor and is used to correct an injection rate model, which is based on previously measured time data and flow data.

[0011] The proposed method uses a measured and therefore "real" injection rate curve as a model for the correction function instead of a simplified curve, for example in the form of a trapezoid. This has the advantage that model errors are reduced to a minimum. Preferably, measurement data of a nominal injector are used for modeling the injection rate. These measurement data preferably consist of time data and flow data with a defined sampling frequency.

[0012] Modeling mainly consists in changing or adapting the time data and flow data with the help of information derived from the sensor signal of the pressure sensor. These data are used to parameterize and correct the injection rate model.

[0013] This correction is carried out during the operation of the injector, preferably over the entire service life of the injector. In this way, an individualized control of the injector can be achieved, which takes into account the actual wear state of the injector.

[0014] Preferably, the following information, in particular time data and flow data, is derived from the sensor signal of the pressure sensor and is used to correct the injection rate model:

[0015] Start of the opening movement of the nozzle needle (NOS),

[0016] End of the opening movement of the nozzle needle (NOE),

[0017] Start of closing movement of the nozzle needle (NCS),

[0018] End of closing movement of the nozzle needle (NCE) and

[0019] Maximum injection rate (Q max ).

[0020] It is further preferred that only this information is derived from the sensor signal of the pressure sensor for parameterizing and correcting the injection rate model. In particular, the use of the sensor signal amplitude is omitted, since, as already mentioned at the outset, the sensor signal amplitude changes over time due to aging and wear of the pressure sensor. In particular, the independence from the absolute sensor signal amplitude increases the accuracy of the model compared to the prior art mentioned at the outset.

[0021] The time data NOS, NOE, NCS and NCE can be derived from the sensor signal of the pressure sensor by means of a detection algorithm. Here, the detection algorithm can use the derivative and / or average value of the sensor signal. In addition, the sensor signal can be preprocessed beforehand, in particular filtered, for example preprocessed by means of a low-pass filter.

[0022] Maximum injection rate Q max Preferably, it is determined from the time data NOS, NOE, NCS and NCE. In this way, the maximum injection rate Q can be detected. max These changes over time depend on the age / wear of the injectors.

[0023] It has been shown that the time data NOS, NOE, NCS, NCE, maximum injection rate Q max There are dependencies with the nozzle needle stroke. These can be identified using conventional methods with linear formulas or by using artificial intelligence with neural networks and can be used to determine Q max Furthermore, it has been shown that these correlations are independent of other wear parameters, such as wear of the nozzle needle seat and / or changes in flow conditions in the region of the throttle plate delimiting the control chamber. These correlations are preferably determined and stored beforehand in an experimental manner, preferably in the control unit.

[0024] Then, the parameter Q obtained from the time data NOS, NOE, NCS, NCE max It is also preferably used as an input variable in the injection rate model.

[0025] Furthermore, preferably, the corrected profile of the injection rate is then calculated by mathematically transforming the time data and the flow data derived from the sensor signal of the pressure sensor. In this case, in particular, a linear transformation can be applied. In order to further increase the accuracy, the transformation can be performed using different mathematical models, for example in a polynomial, exponential or logarithmic manner, and / or the transformation can be applied only over a certain period of the opening phase or the closing phase.

[0026] When mathematically transforming the time data, preferably only the time data NOS, NOE, NCS, NCE derived from the sensor signal of the pressure sensor are used. This means that only these time data are used in order to avoid errors or inaccuracies.

[0027] The modified injection rate model is then preferably also used to calculate the injection quantity, which is preferably calculated by calculating the maximum injection rate Q max The calculated injection quantity can then be converted into a change in the electrical control duration. Here, the deviation obtained from the actual injection quantity can be used as an influencing factor for the injection duration.

[0028] Furthermore, a control unit is proposed which is designed to carry out the method according to the invention. For this purpose, an injection rate model is preferably stored in the control unit, which is based on measured data of a nominal injector. In order to correct the injection rate model, the control unit evaluates the sensor signal of the pressure sensor. For this purpose, a detection algorithm can be stored in the control unit, by means of which the relevant time data and flow data can be derived from the sensor signal. If this evaluation detects drift or wear, the injection rate model can be corrected accordingly by means of the control unit. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The method according to the invention and its advantages are explained in detail below with reference to examples and drawings. In the drawings:

[0030] Figure 1 A flow chart is shown for illustrating the flow of the method according to the present invention,

[0031] Figure 2 , a) shows a schematic longitudinal section of an injector which can be controlled according to the method of the invention, and b) shows an enlarged detail of an injector,

[0032] Figure 3 Different graphs are shown: a) for the time curve of the current, b) for the time curve of the nozzle needle stroke, c) for the time curve of the injection rate and d) for the time curve of the sensor signal of the pressure sensor,

[0033] Figure 4Different graphs are shown for linearly operated injectors (with stops): a) for the time curve of the injection rate of a reference injector, b) for the time curve of the injection rate of a reference injector with the aid of Q max c) for representing the time curve of the injection rate adapted with the aid of information from the sensor signal of the pressure sensor,

[0034] Figure 5 Different diagrams are shown for a ballistically operated injector (without stop): a) for the time curve of the injection rate of a reference injector, b) for the time curve of the injection rate of a reference injector using Q max c) for representing the time curve of the injection rate adapted with the aid of information from the sensor signal of the pressure sensor,

[0035] Figure 6 A graphical representation showing the transformation of the time axis during the opening phase of the injector,

[0036] Figure 7 A graph is shown for representing the modeling of the injection rate in the closing phase of the injector in linear operation,

[0037] Figure 8 A graphical representation showing the transformation of the time axis during the closing phase of the injector,

[0038] Fig. 9 shows a diagram for illustrating the modeling of the injection rate in the closing phase of the injector during ballistic operation with the aid of an EOI threshold value,

[0039] Fig.10 A graph is shown for illustrating the modeling of the injection rate in a closing phase of the injector during ballistic operation with the aid of a ballistic point; and

[0040] Fig.11 A graph is shown comparing a simplified injection rate model (dashed line) with a measured curve (solid line). DETAILED DESCRIPTION

[0041] The method according to the invention serves to actuate an injector 1 which has a servohydraulically controlled nozzle needle 2 and a pressure sensor 5 for measuring the control pressure exerted on the nozzle needle 2 . Figure 2 Such an injector 1 is shown by way of example in . Figure 2 a) shows the complete injector 1, Figure 2 b) Only the area of ​​the nozzle and the area of ​​the control valve are shown.

[0042] exist Figure 2The injector 1 shown in FIG. 1 has a nozzle needle 2 which is received in a nozzle body 6 so as to be reciprocating and is used to open and close at least one injection opening 3. The end of the nozzle needle 2 facing away from the injection opening 3 delimits a control chamber 4 which is hydraulically connected to a pressure chamber 8 via a channel 7. The pressure chamber 8 is constructed in a throttle plate 9 and is delimited by a membrane 10 which is attached to the throttle plate 9, behind which a pressure sensor 5 is arranged, so that pressure changes occurring in the pressure chamber 8 can be detected by means of the pressure sensor 5. The pressure sensor 5 is integrated into a valve plate 11, on which the valve body 12 of the control valve 13 is attached. The actuation of the control valve 13 leads to a change in the pressure in the control chamber 4, which applies a hydraulic force acting in the closing direction to the nozzle needle 2. Therefore, the pressure in the control chamber 4 controls the reciprocating movement of the nozzle needle 2. Therefore, information representative of the reciprocating movement of the nozzle needle 2 can be obtained from the sensor signal of the pressure sensor 5.

[0043] from Figure 3 The nozzle needle stroke ( Figure 3 b) and the sensor signal of the pressure sensor 5 ( Figure 3 d). The sensor signal allows the start of the nozzle needle movement during opening (time NOS) and the end of the nozzle needle movement during opening (time NOE) to be identified. Similarly, the start (NCS) and end (NCE) of the nozzle needle movement during closing can be derived from the sensor signal. The times NOS, NOE, NCS and NCE are given in Figure 3 c) with a maximum injection rate Q max However, in Figure 3 The current curve shown in a) has no direct correlation with the nozzle needle stroke, since there is a time delay between the control current and the movement of the nozzle needle 2 .

[0044] Thus, time data and flow data can be derived from the sensor signal of the pressure sensor 5, which data allow the injection rate model to be corrected according to the current event and the control current to be adapted accordingly. In this way, the aging / wear of the injector 1 can also be taken into account during the control.

[0045] exist Figure 1 The main steps of such a correction function are shown in . The correction function comprises four main steps A to D, which are implemented by means of the control unit 14 . For this purpose, the sensor signal of the pressure sensor 5 integrated into the injector 1 is provided to the control unit 14 via a signal line 15 .

[0046] Step A is used for feature recognition. The sensor signal is first processed using a low-pass filter 17. The time data NOS, NOE, NCS, NCE are then derived from the signal using a detection algorithm 18 stored in the control unit 14.

[0047] Step B is used for drift detection. In order to detect changes in the injection characteristics of the injector 1 over the operating time (this is also called drift), the time data obtained from the sensor signals NOS, NOE, NCS, NCE in step A are stored, specifically at different system pressures and / or current durations, so that the maximum injection rate Q can be calculated. max Using time data NOS, NOE, NCS, NCE, maximum injection rate Q max The correlations with the nozzle needle travel are used for the calculation. These correlations were previously determined experimentally.

[0048] For this purpose, the needle closing time, t closing =NCE-NCS. The test is preferably designed so that the relevant changes in the needle stroke and the nozzle flow rate are measured. According to formula (1), the relationship for the two system pressures can be determined. The greater the difference between the two system pressures, the more accurately the nozzle needle stroke and the maximum injection rate Q can be determined. max This is because, when the system pressures differ greatly from each other, the coefficients a and b also differ greatly from each other, and therefore the difference in the slopes of the two straight lines according to formula (1) is large.

[0049] (1)

[0050] In formula (1), a, b and c are the sys The specific constants determined experimentally under the conditions of It can be determined according to formula (2). Preferably, the time data stored in the controller in the 0-km situation and the currently measured time data are used for the determination according to formula (3).

[0051] (2)

[0052] is the regulated injector at nominal pressure p nom The maximum injection rate under .

[0053] The 0-km measurement with the regulated injector is calculated.

[0054] (3)

[0055] These correlations can be used to determine That is, determine the change in the maximum injection rate of the injector at the nominal pressure. Then, the result can be used in step C (see Figure 1 ) is used as an input value for the injection rate model.

[0056] As mentioned above, in contrast to the prior art, the injection rate model is based on a measured injection rate curve. Preferably, the injection rate curve of the nominal injector is measured at different system pressures and long energization durations and stored in the controller as a reference. Here, the rate information consists of time data "t" and corresponding flow data "Q".

[0057] In a first calculation step of the injection rate model, preferably corresponding to the existing system pressure p ist The nominal flow data are scaled by a factor k1 which is defined by formula (4).

[0058] (4)

[0059] is the maximum injection rate of the reference injector at nominal pressure. According to formula (2), another input value is calculated, which specifies the maximum injection rate of the injector to be regulated. These steps are respectively shown graphically in Figure 4 a) to Figure 4 c) shows for linear operation and Figure 5 a) to Figure 5 c) is shown for a ballistic run.

[0060] At the operating point in which the opening nozzle needle reaches the stop (linear operation), the scaled injection rate curve is adapted with the aid of the time data NOS, NOE, NCS, NCE (see Figure 4 c). At the same time, the time axis of the reference velocity curve is adapted by means of the actual time. The difference between the time data NOS of the reference injector and the regulated injector is taken into account in the following way: Preferably, the time axis is corrected with this value according to formula (5), where p ist is the current system pressure.

[0061] (5)

[0062] The modeling of the opening phase is preferably achieved by transforming the time axis from the nominal opening time (NOE-NOS) to the opening time of the regulated injector. In this case, in particular a linear transformation can be applied, such as for example in Figure 6 The transformation can be performed using different mathematical models, for example in a polynomial, exponential or logarithmic manner and / or only over a certain period of the start-up phase.

[0063] The difference between the time data NOE of the reference injector and the injector being controlled is taken into account by correcting the time axis by this value, preferably according to formula (6).

[0064] c6)

[0065] With the onset of needle closing of the regulated injector, the injection rate curve is modified and the closing is modeled. For this purpose, the closing process of the reference measurement is used. Thus, all linear points can be calculated with the reference injection rate of the long injection (see Figure 7 ).

[0066] If the NOE of the regulated injector is greater than the NOE of the reference injector, the injection rate curve can be approximated in this region by means of linear extrapolation. Preferably, however, a measurement with the maximum possible current duration is used as a reference to avoid such an additional calculation step.

[0067] Preferably, an additional time point is defined for modeling the closure, specifically defining the time point EOI threshold (EOI Schwelle ), from which the modeled closing process is identical to the reference closing process. Between the NCS and EOI thresholds, a similar transformation of the time axis as before in the start phase is applied. The EOI threshold is defined according to formula (7) as the fraction of the flow at the NCS time point, where the model parameters are k2·∈·]0;·1[.

[0068] (7)

[0069] The entire modeling of the timeline when closed is exemplified in Figure 8 Shown in.

[0070] Furthermore, the scaled injection rate curve of the reference injector is preferably transformed during the closing phase in such a way that continuity of the modeled rate curve results. This is exemplified in Fig. 9 Shown in.

[0071] The transformation process described above generates a modeled injection rate, by integrating which the injection quantity can be calculated (see Figure 4 c)). The slopes of the rising and falling edges can also be used as characteristic values ​​for describing the state of the injector function by selective averaging of the derivatives.

[0072] In operating points where the nozzle needle does not reach its stop during opening (ballistic operation), there is no time NOE. However, the transformation in the opening phase can be applied similarly to the transformation in linear operation. Since the NOE is not available, this information is obtained from the actual learned values ​​of the linear operation. In principle, the modeled injection rate follows the opening edge calculated above. Until the closing phase begins. In addition, the model parameters Δt and ΔQ are introduced and according to Fig.10 Explain them. According to this, Δt is defined as the maximum injection rate Q on the scaled reference rate at the time point NCS+Δt max Therefore, at time point NCS calculates Q at point B max ΔQ is defined as follows: At the injection rate Q max -ΔQ reaches the point A at which the modeled injection rate no longer follows the reference rate.

[0073] It has been shown that Δt and ΔQ can be defined using the time constants k3 and k4 given in equations (8) and (9), which lead to good model results.

[0074] e8)

[0075] (9)

[0076] In formulas (8) and (9):

[0077] is the needle opening time of the modeled curve NCS-NOS, and

[0078] is the needle opening time of the regulated injector for the linear operating point or a learning parameter.

[0079] The curve course between point A and point B is preferably modeled using a parabola.

[0080] Then, between points B and C, the same transformation as for the linear operating region is performed (see Fig.10 to the right of the diagram).

[0081] In step D (see Figure 1 ), the estimated injection quantity can be converted into a change in the electrical control duration. The determined deviation of the actual injection quantity is preferably used as an influencing factor for the injection duration. The correspondingly matched control current is conducted from the controller 14 to the injector 1 via the control line 16.

Claims

1. A method for controlling an injector (1), in particular a fuel injector, in which the reciprocating movement of a nozzle needle (2) is controlled by the pressure in a control chamber (4) for opening and closing at least one injection opening (3), and the pressure in the control chamber (4) is measured by means of a pressure sensor (5) integrated in the injector (1), wherein: Information characteristic of the reciprocating movement of the nozzle needle (3), in particular time data and flow data, is derived from the sensor signal of the pressure sensor (5) and is used to correct an injection rate model based on previously measured time data and flow data.

2. The method according to claim 1, characterized in that: The following information, in particular time data and flow data, is derived from the sensor signal of the pressure sensor (5) and is used to correct the injection rate model: The start of the opening movement of the nozzle needle (NOS), end of the opening movement of the nozzle needle (NOE), The start of the closing movement of the nozzle needle (NCS), The end of the closing movement of the nozzle needle (NCE) and Maximum injection rate (Q max ).

3. The method according to claim 2, characterized in that The time data (NOS, NOE, NCS, NCE) are derived from the sensor signal of the pressure sensor (5) by means of a detection algorithm which utilizes a derivative and / or an average value of the sensor signal.

4. The method according to claim 2 or 3, characterized in that: The maximum injection rate (Q max ) is obtained from the time data (NOS, NOE, NCS, NCE).

5. The method according to claim 4, characterized in that The following dependence previously determined experimentally and stored is used to determine the maximum injection rate (Q max ), the correlation exists between the time data (NOS, NOE, NCS, NCE), the maximum injection rate Q max and between the strokes of the nozzle needle (2).

6. The method according to any one of the preceding claims, characterized in that The maximum injection rate (Q) is calculated by mathematically transforming the information obtained from the sensor signal of the pressure sensor (5), in particular the time data and the flow data. max ) of the corrected curve.

7. The method according to claim 6, characterized in that When mathematically transforming the time data, only the time data (NOS, NOE, NCS, NCE) derived from the sensor signal of the pressure sensor (5) are used.

8. The method according to any one of the preceding claims, characterized in that The modified injection rate model is used to calculate the injection quantity, which is preferably calculated by calculating the maximum injection rate (Q max ) is performed by integrating the corrected curve.

9. The method according to claim 8, characterized in that The calculated injection quantity is converted into a change in the electrical actuation duration, wherein preferably the ascertained deviation of the actual injection quantity is used as an influencing factor for the injection duration. 10 . A control unit which is designed to carry out the method according to one of the preceding claims.

Images