DE102006046470B4 - Method for operating an injection valve - Google Patents

Abstract

Method for operating an injection valve (11), in particular an injection system of an internal combustion engine, wherein the injection valve (11) comprises a piezoelectric actuator (12) and to the piezoelectric actuator (12) an electrical voltage (U) is applied, resulting in a longitudinal expansion or Shortening (x) of the piezoelectric actuator (12), wherein with a closed injection valve (11) a holding voltage (± U1) is applied, which is changed to initiate an injection with a gradient with an approximately linear voltage change, that with a change in the gradient (gE, gL) the time (tE, tL) of the beginning of the voltage change is shifted (tE ', tL'), characterized in that the displacement is carried out so that at the loading or unloading edge (L, L ', E, E ') a linear voltage curve over time passes through a characteristic point voltage / time (S1, S2), and that the displacement of the charging or discharging start (tE, tL, tE ', tL') is determined on the basis of a straight line equation, the straight line passing through the characteristic point (S1, S2).

Description

State of the art

The present invention relates to a method for operating an injection valve, in particular an injection system of an internal combustion engine, wherein the injection valve comprises a piezoelectric actuator and the piezoelectric actuator, an electrical voltage is applied, which leads to a lengthening or shortening of the piezoelectric actuator, wherein when closed Injector a holding voltage is applied, which is changed to initiate injection with a gradient with an approximately linear voltage change.

A generic injection valve (injector) with piezoelectric actuator as a controller is for example from the DE 10 2004 046 080 A1 known.

From the DE 101 13 670 A1 a method and a device for controlling a piezoelectric actuator is known. In this control, the operating situation of the internal combustion engine is detected and selected the time derivative of the tapped off the piezoelectric voltage as a function of the operating situation. This is done in order to achieve a compromise between noise emissions and wear on the one hand and the required switching speed at high speeds. In this case, the times for control start or drive end are adjusted so that the periods of complete opening are kept constant.

The DE 101 48 217 C1 also describes a method for operating an internal combustion engine. In an internal combustion engine, the fuel passes through a fuel injection device, which comprises a piezoelectric actuator, in the combustion chamber of the internal combustion engine. In order to be able to inject the fuel optimally, it is proposed that the setpoint level of the drive energy and / or the setpoint gradient of the drive energy, with which the piezoelectric actuator is controlled, depend on a plurality of influencing variables.

If the charge or discharge gradient of such an actuator is changed when an injection is discontinued, then the start of injection or the end of injection and the total injection quantity change. A change in the charge or Entladegradienten can be done, for example, by Exemplar scattering of the injectors, by Exemplar scattering of the associated drive circuit or other external influences. By changing the start of injection or injection end and injection quantity, the running behavior and emissions of an internal combustion engine is adversely affected.

An object of the present invention is to reduce the effects of a change in charge and discharge gradients of a piezoelectric actuator when issuing injection on injection quantity and opening and closing timing of the injection valve.

Disclosure of the invention

This problem is solved by a method, an apparatus and a computer program according to the independent claims. The dependent claims relate to advantageous embodiments of the invention.

By means of the method according to the invention or the devices according to the invention and the computer program according to the invention, the required actuation duration and the actuation start for any desired injection quantity for a conventional main injection or pre-injection can be determined and corrected with variation of the discharge gradient or charge gradient. This means that time-consuming map measurements with different charge or discharge gradients are not necessary. It is only necessary to measure a single characteristic diagram with a constant discharge gradient or charging gradient and to record a further injection profile with different charge gradients or discharge gradients per pressure. Due to the discharge gradient, the opening behavior of the injection valve and, correspondingly, by the charging gradient, the closing behavior of the injection valve and thus overall the course of the injection as well as the start and end of injection can be influenced. The charge gradient as well as the discharge gradient can be variable depending on operating point, for example to accomplish an attached pre-injection. By varying the Entladegradienten there is a shift in the start of injection. This shift can be corrected by means of the method according to the invention or the device according to the invention and the computer program according to the invention. The correction is necessary because a shift in the start of injection brings about a deterioration of the emission values, for example with regard to exhaust gas composition and acoustics, and thus may also result in a loss in ride comfort. Furthermore, the variation of the charging or discharging gradient entails an increased or decreased quantity, which can likewise be corrected by the method according to the invention and the device according to the invention and the computer program according to the invention. It can be created in a development, a correction factor map and stored in a control unit, which can be determined all drive times for a desired amount with variation of Entladegradienten. With the aid of the determination according to the invention of Entladebeginn and discharge duration and a Vorvermessung of injectors in operation in determining the Entladegradienten at an operating point, the Ansteuerdauer and the start of control for a desired injection quantity can be calculated and corrected. For the discharge gradient as well as the charging gradient there is a characteristic point of the respective charge and discharge curve, which is to be measured at the factory and may possibly be coded into the injector in order to classify it. If the specimen scattering with respect to the characteristic point is small, coding can be dispensed with and a single value for an injector family can be assumed.

The charging gradient is an essential parameter for the seat force of the valve needle of the injection valve. The charging gradient should also be variable depending on the operating point. In order to get the required amount and thus the required activation duration at the corresponding charging gradients, a previously measured characteristic map of the injection quantity must be stored over the activation period, for example in a control unit. According to the invention, a correction factor map can be created and stored in the control unit, which can be used to determine all activation periods for a desired quantity with variations of the charging gradient. Furthermore, with the method according to the invention or the device according to the invention or the computer program and a pre-measurement, the applicators in remote operation can obtain the activation duration for the required desired spray quantity when determining the charging gradient at an operating point. The characteristic point associated with the charge gradient may also be factory-measured, as well as the characteristic point associated with the discharge gradient, and may be encoded into the injector to classify it.

The above statements relate to an injector that is discharged to stop an injection. In the case of an injector that is charged to discontinue injection, charge and discharge flanks or charge start and end of charge must be exchanged mutatis mutandis.

Brief description of the drawings

Hereinafter, an embodiment of the present invention will be explained in more detail with reference to the accompanying drawings. Showing:

1 a sketch of a fuel injection system of a motor vehicle;

2 a graph of the voltage across the piezoelectric actuator over time at a change in the voltage gradient of the falling edge;

3 a diagram of the flow over time at a voltage curve according to 2 ;

4 a diagram of the voltage across the piezoelectric actuator over time with inventive shift of the control start;

5 a diagram of the flow over time at a voltage curve according to 4 ;

6 a diagram accordingly 4 with marking of various sizes;

7 a graph of the voltage across the piezoelectric actuator over time at a change in the voltage gradient of the rising edge;

8th a diagram of the flow over time at a voltage curve according to 2 ;

9 a diagram of the voltage across the piezoelectric actuator over time in accordance with the invention displacement of the control start of the rising edge;

10 a diagram of the flow over time at a voltage curve according to 9 ,

Embodiment of the invention

In the 1 a fuel injection system of a motor vehicle is shown, which is a control unit 10 and an injection valve 11 having. The injection valve 11 is with a piezoelectric actuator 12 provided by the control unit 10 is controlled. Furthermore, the injection valve 11 a valve needle 13 on that on a valve seat 14 inside the housing of the injector 11 can sit up.

Is the valve needle 13 lifted from the valve seat, so is the injector 11 open and fuel is injected. This condition is in the 1 shown. Sits the valve needle 13 on the valve seat 14 on, so is the injector 11 closed. The transition from the closed to the open state is by means of the piezoelectric actuator 12 causes. For this purpose, an electrical voltage to the actuator 12 applied, which causes a change in length of a piezo stack, which in turn to open or close the injector 11 is exploited. The injection valve 11 has a hydraulic coupler 15 on. For this purpose is within the injector 11 a coupler housing 16 present in which two pistons 17 . 18 are guided. The piston 17 is with the actor 12 and the piston 18 is with the valve needle 13 connected. Between the two pistons 17 . 18 is a chamber 19 arranged, which is a piston / cylinder system for the transmission of the actuator 12 applied force on the valve needle 13 forms.

The coupler 15 is surrounded by pressurized fuel. The volume of the chamber 19 is also filled with fuel. About the guide gap between the two pistons 17 . 18 and the coupler housing 16 can the volume of the chamber 19 over a longer period of time to the respective existing length of the actuator 12 to adjust. For short-term changes in the length of the actuator 12 the volume of the chamber remains 19 and thus their length is almost unchanged and the change in the length of the actuator 12 gets on the valve needle 13 transfer. The guide gap between the two pistons 17 . 18 and the coupler housing 16 can form a valve in different flow directions or depending on the position of the piston 17 . 18 to the coupler housing 16 has different flow resistance or flow coefficients. For example, one or both pistons may have grooves of variable depth of the groove bottoms or the like, around the effective area through which the pistons pass 17 . 18 and the coupler housing 16 to change. The adjustment of the speed of movement of the pistons 17 . 18 to each other z. B. by the leadership games between the pistons 17 . 18 and the coupler housing 16 or by a small throttle with directional flow coefficient.

If a voltage U is applied to the piezoelectric actuator, its length changes. A very slow, quasi-stationary change in length is through the coupler 15 balanced. Is the on the piezoelectric actuator 12 applied voltage with a sufficiently large gradient changed, so that the coupler 15 the change in length of the actuator 12 can no longer compensate, then the valve needle 13 lifted off her valve seat. Usually, the at the piezoelectric actuator 12 applied voltage U held at a holding voltage U1 and for discontinuing an injection to a so-called bottom voltage UB, which is usually 0 V changed. To the injection valve 11 to close again, the bottom voltage UB is raised again to the holding voltage U1. Alternatively, this process can also be designed vice versa, that is, the injection valve 11 at a voltage of z. B. 0 V on the piezoelectric actuator 12 is closed and the voltage U to open the injector 11 is raised to a different from 0 V positive or negative voltage. Hereinafter, it is assumed in the description of the embodiment that the piezoelectric actuator 12 with closed injection valve 11 has a different from 0 V holding voltage U1 and is energized to open so that on the piezoelectric actuator 12 a bottom voltage UB of about 0V is applied.

In 2 is the voltage U across the piezoelectric actuator 12 over time t is shown as dashed curve A. With closed injection valve 11 the voltage is maintained at a holding voltage of U1, as previously explained. At a Entladebeginn t _E is the piezoelectric actuator 12 energized, so that discharges the applied voltage U with a discharge edge E with a gradient g _E to a discharge end T _eende . The gradient g _E thus corresponds to (U1 - UB) through (t _E - _E ) and is assumed to be constant here. This can be achieved by a voltage-guided energization of the piezoelectric actuator. At time t _Eenden the bottom voltage UB from 0V is applied. The bottom voltage is held for a hold time t _H until a charge start I _L. From the time charging starts t _L , the voltage U is again charged to the holding voltage U1 with a gradient g _L.

In 2 is further a voltage curve B of the voltage U across the piezoelectric actuator 12 shown over the time t, in which the Entladegradient g _{E has been changed} to a discharge gradient g _E '. The discharge start t _Eende shifts thereby up to a value t _Eende . As a result, the total opening time of the injection valve 11 changed in the in 2 illustrated embodiment, this is shortened. In 3 is the flow rate p. (p with point above the p, this corresponds to dp / dt) over time t. An integration of the flow p. through the injector 11 over time, the total injection amount p gives an injection. The dashed line A in 3 belongs to the dashed lines shown voltage waveform A in 2 , the solid line B in 3 belongs to the voltage curve B shown in solid lines with changed gradient g _E 'in 2 , Due to the lower voltage gradient g _E 'during discharge at the in 2 drawn through voltage curve B shifts the start of injection t ₁ to an injection start t ₁ '. The injection end t ₂ is identical for both voltage curves A and B. An integration via the curve A gives a higher injected fuel quantity p than an integration via the curve B. By changing the discharge gradient g _E not only the start of injection t _{1 is} shifted to t ₁ ', but also the total injection quantity p changed.

U_S1

= zum charakteristischen Punkt S₁ gehörende Spannung

UB

= 0 V = Bottomspannung

U1

= Haltespannung

AD_E

= Ansteuerdauer des gemessenen Entladegradienten, also die Zeitdifferenz t_Eende – t_E

AD_{E_korr}

= korrigierte Ansteuerdauer für den variierten Entladegradienten

y1E

= derzeitiger (gemessener) Entladegradient

sy2E

= neuer Entladegradient ist

phi_korr

= korrigierter Ansteuerbeginn.

4 shows a voltage curve on the piezoelectric actuator 12 in which the curve A corresponds to the voltage curve A in FIG 2 equivalent. In the voltage profile of the curve B 'of the start of the discharge was _E t _E as a discharge starting t' shifted so that the beginning of injection t ₁ 'to the start of injection t ₁ as it coincides belongs to the curve A. The course of injection to the in 4 shown voltage curve is in 5 shown. As can be seen fall injection start t ₁ and t ₁ 'for the curves A and B together, the injection end t ₂ falls, as well as from the 2 and 3 represented, also together. As a result, neither the beginning of injection nor the injection quantity p are displaced by the displacement of the start of discharge t _E. According to the invention, it has now been found that all unloading flanks intersect at a characteristic intersection point S ₁ . If the discharge gradient E is changed, it must go to the characteristic point S _{1 in} order to obtain the same injection quantity and the same start of injection. The calculation of the required actuation duration for a given discharge gradient g _E takes place according to the formula AD _{E_korr} = AD _E - (U1 -U _S1 ) / y1E + (U1 -U _S1 ) / y2E (Formula 1.1) With:

U _S1

= voltage belonging to the characteristic point S ₁

UB

= 0 V = bottom voltage

U1

= Holding voltage

AD _E

= _{Activation duration of} the measured _{discharge gradient} , ie the time difference t _Eende - t _E

AD _{E_korr}

= corrected drive time for the varied discharge gradient

Y1E

= current (measured) discharge gradient

sy2E

= new discharge gradient

phi _corr

= corrected start of control.

6 shows a sketch to illustrate the sizes used in the formula 1 and 2. The correction of the start of control can be determined with the following formula phi _corr = AD - AD _corr (formula 1.2)

It should be noted that linear paths in the control voltage are assumed, ie a voltage-controlled control is present. For a current-controlled control, this would have to be taken into account and possibly converted, but here too there is a characteristic intersection of all unloading flanks.

U_S2

= zum charakteristischen Punkt S₂ gehörende Spannung

UB

= Bottomspannung

AD_L

= Ansteuerdauer des alten Ladegradienten, also die Zeitdifferenz t_Le – t_L

AD_{L_korr}

= korrigierte Ansteuerdauer für den neuen Ladegradienten

y1L

= derzeitiger (gemessener) Entladegradient

y2L

= neuer Ladegradient ist

phi_korr

= korrigierter Ansteuerbeginn.

7 shows a voltage waveform of the voltage U across the piezoelectric actuator 12 over the time t with a variation of the charge gradient g _L. For clarity, the reference numerals for the Entladegradienten, as this in the 2 to 6 is shown suppressed. In 7 an example is shown in which loading flanks L and L 'start with loading gradient g _L for the loading flank L and g _L ' for the loading flank L 'in each case at the same loading time t _L. Due to the different gradients, the holding voltage U1 is reached at different times, in the case of the charging flank L this is at a time t _Le , in the case of the charging flank L 'this is the case at a time t _Le '. This results in a shift of the injection end t ₂ , for the curve A with the charging edge L, the injection end is reached up to a time t ₂ , for the curve B with the charging edge L ', the injection end is reached at a time t ₂ '. Accordingly, the flow rates integrated over time are p. different, the total injection q are so different. Similar to what has been done before, as in 9 according to the representation of 4 is sketched, a shift of the charging start t _L to a loading start t _L '. As in 10 can be seen, this causes the injection end t _E is the same for both loading flanks, a total of the same total amount of fuel was injected. As explained above on the discharge flank, it can also be observed here that all charging flanks L, which effect an equal injection quantity q, run through a characteristic point S2. The required actuation time for given charging gradients is given by the following formula AD _{L_korr} = AD _L + U _S2 / y1L - U _S2 / y2L (Formula 2.1) With

U _S2

= voltage belonging to the characteristic point S ₂

UB

= Bottom voltage

AD _L

= Control period of the old Ladegradienten, ie the time difference t _Le - t _L

AD _{L_korr}

= corrected drive time for the new charge gradient

Y1L

= current (measured) discharge gradient

Y2L

= new load gradient

phi _corr

= corrected start of control.

Again, it should be noted that is assumed by linear gradients in the drive voltage U, which is to be considered in a current-guided and thus possibly non-linear course of the voltage U.

Claims (8)

Method for operating an injection valve ( 11 ) in particular an injection system of an internal combustion engine, wherein the injection valve ( 11 ) a piezoelectric actuator ( 12 ) and to the piezoelectric actuator ( 12 ) an electrical voltage (U) is applied, which leads to a longitudinal expansion or shortening (x) of the piezoelectric actuator ( 12 ), wherein when the injection valve is closed ( 11 ) a holding voltage (± U1) is applied, which is to initiate an injection with a Gradients with an approximately linear voltage change is changed so that when a change in the gradient (g _E , g _L ) the time (t _E , t _L ) of the beginning of the voltage change shifted (t _E ', t _L '), characterized in that the displacement is carried out in such a way that, in the case of the charge or discharge edge (L, L ', E, E'), a linear voltage characteristic over time passes through a characteristic point voltage / time (S ₁ , S ₂ ), and the shift of the loading or Entladebeginns (t _E , t _L , t _E ', t _L ') is determined based on a straight line equation, the straight line through the characteristic point (S ₁ , S ₂ ) goes. A method according to claim 1, characterized in that the holding voltage (U1) for discontinuing the injection by a discharge edge (L, L ') and a charging edge (E, E') to a bottom voltage (UB) and back to the holding voltage (U1) when a loading gradient (g _L ) of the loading flank (L) or a discharge gradient (g _E ) of the unloading flank (E) changes, a shift of a loading or unloading start (t _E , t _L , t _E ', t _L ') is made A method according to claim 2, characterized in that for discontinuing an injection starting at a discharge start with a substantially constant Entladegradienten (g _L ) a discharge edge (L) to a bottom voltage (UB) and after a holding time (t _H ) starting at a charge start (t _L ) with a substantially constant charging gradient (g _L ) of a charging edge (L) is brought to the holding voltage (U1) (or vice versa), wherein in a change of the charge or Entladegradienten (g _E , g _L ) a Shift of the charging or Entladebeginns (t _E , t _L , t _E ', t _L ') is made. Method according to one of claims 1 to 3, characterized in that the characteristic point (S ₁ , S ₂ ) is determined in the experiment on the basis of at least two charging or Entladegradienten. Method according to Claim 4, characterized in that the characteristic point (S ₁ , S ₂ ) is a value pair of actuator voltage / activation duration for a characteristic charge or discharge gradient. A method according to claim 5, characterized in that the characteristic point (S ₁ , S ₂ ) is determined specifically for each injection valve and stored as a constant in a control unit. Device, in particular control device for an internal combustion engine, with means for operating an injection valve ( 11 ) in particular an injection system of an internal combustion engine, wherein the injection valve ( 11 ) a piezoelectric actuator ( 12 ) and to the piezoelectric actuator ( 12 ) an electrical voltage (U) is applied, which leads to a longitudinal expansion or shortening (x) of the piezoelectric actuator ( 12 ), wherein when the injection valve is closed ( 11 ) is applied to a holding voltage (± U1), which is changed to settle an injection with a gradient with an approximately linear voltage change, that when a change of the gradient (g _E , g _L ) the time (t _E , t _L ) of the beginning the voltage change is shifted (t _E ', t _L '), characterized in that means are provided which make the displacement so that in the loading or unloading flank (L, L ', E, E'), a linear voltage waveform over the time by a characteristic point voltage / time (S ₁ , S ₂ ) goes, and determine the shift of the charging or Entladebeginns (t _E , t _L , t _E ', t _L ') using a straight line equation, the straight line passes through the characteristic point (S ₁ , S ₂ ) Computer program with program code for performing all the steps according to one of claims 1 to 6, when the program is executed in a computer.

Images