DE102008043166B4 - Method for controlling an injection system of an internal combustion engine - Google Patents

Abstract

Method for controlling an injection system of an internal combustion engine having at least one, in particular equipped with a low-pressure stage, electrically controllable injector, wherein a fuel metering in a first partial injection (PI) and at least a second partial injection (MI) is divided, one to be injected by means of at least one injector Fuel quantity determining control signal is corrected in response to at least one signal characterizing the electrical spray distance and a signal characterizing the rail pressure, characterized in that the correction of the control signal is a shortening of the driving time, wherein the shortening of the driving time is selected so that one by a time-delayed provision of an injector needle of the injector caused additional amount of at least one second injection (MI) is compensated.

Description

The invention relates to a method for controlling an injection system of an internal combustion engine having at least one, in particular equipped with a low-pressure stage, an electrically controllable injector according to the preamble of claim 1.

The subject matter of the present invention is also a computer program and a computer program product with a program code which is stored on a machine-readable carrier for carrying out the method.

State of the art

Compliance with emission limits is one of the priority objectives in the development of internal combustion engines. Progress has been made here with common-rail injection systems, which make a decisive contribution to the reduction of pollutants. The advantage of such common rail systems is that they can be operated independently of the injection pressure, the speed and the load. In order to achieve future exhaust emission limits, an improvement in the mixture formation is required in diesel engines. This is achieved on the one hand through the use of so-called exhaust gas recirculation (EGR), which successfully counteracts nitric oxide formation. On the other hand, a high exhaust gas recirculation rate favors the formation of soot. To prevent this, a rapid mixing of the fuel-air mixture must take place, which is realized in current common-rail systems, inter alia by increasing the injection pressures.

On the other hand, high injection pressures require rate shaping, which is usually realized today via multiple injections. In this case, variable designs of different injections are provided, these injections following each other up to a hydraulic distance of zero.

Such close accumulated multiple injections, for example, one or more pilot injections, a main injection and one or more post-injections require on the one hand very fast valve switching times, which can be realized with today's pressure-balanced solenoid valves or with piezo valves. On the other hand, a "hard" injector mechanism is required, which allows a quick reset. Such injector mechanics, however, are not equipped with electrically operated injectors with solenoid valves, especially with a low-pressure stage. Rather, the nozzle needle is strongly biased at high Raildrücken by the low pressure stage. This bias causes some time delays both when opening and when returning to the end of the injection. If, in this case, a second injection follows in time so close to the previous injection that the injector has not yet been reset, this results in an undesired additional quantity that is disadvantageous for the emission behavior of the engine. Such a solenoid valve injector, for example, from the book publication BOSCH Automotive Paperback, 25th edition, October 2003, Friedrich Viehweg & Sohn Verlag / GWV Fachverlage GmbH, page 706, out.

Disclosure of the invention

The inventive method with the features of claim 1 has the advantage that even such, in particular equipped with a low pressure stage injector for very closely attached multiple injections can be used.

The basic idea of the invention is to compensate for the negative effects caused by the compression of the nozzle needle by a change in the sense of a correction of the drive signal. The invention is based on the basic idea and the recognition that a mechanical or hydraulic reset follows a certain law, which can be described mathematically simple. This is because in the dominating effects, for example, the compression of the nozzle needle, a far-reaching linearity is observed.

From the DE 10 2007 043 879 A1 For example, a method of operating an injector is known in which a subsequent injection time for a subsequent injection following a previous injection with a previous injection time after a pause time is shortened by a correction time derived from the pause time. The correction described here is a pressure wave correction.

In contrast to known from the prior art method in which the injection quantity is stored as a function of driving time and pressure in a map, in the present invention, a quantity of a subsequent injection determining control signal is corrected as a function of the electrical spray distance and the rail pressure in that the correction of the control signal is a shortening of the activation duration.

This correction goes beyond those known from the prior art and for example from the DE 10 2006 043 326 A1 emanating Pressure wave correction addition. With the pressure wave correction, a surplus caused by pressure waves is corrected for multiple injections. However, this method is unsuitable for a non-periodic, linear increase. In contrast to the pressure wave correction, the method according to the invention provides a correction function which depends only on a few input variables and has a nearly linear behavior. It is an essential point of the present invention that here not a pressure wave is corrected, which depends on the temperature and the rail pressure, but a conditional by the pressure-induced compression of the components, in particular the nozzle needle and the nozzle seat provision. An incomplete provision results in an increase in quantity in a multiple injection. In this case, the invention exploits the knowledge that the additional amount caused by the mechanical reset time does not depend on the pilot injection quantity, but rather depends only on the electrical spray distance and the rail pressure. This considerably simplifies the determination of the correction. The fact that the mechanical reset time is determined by typical injector sizes, such as axial stiffness of the nozzle needle, size of the inlet throttle, stiffness of the nozzle seat, the correction does not have to be performed for each Injektorexemplar, but can be determined for an entire series once, so injector specific , It is also essential that this effect is temperature-independent and behaves approximately linearly. The reset time is a function of the rail pressure. Within the component tolerances, the correction function can thus be used in a very advantageous manner for all injectors of one type.

In this case, the shortening of the activation duration is selected such that an excess quantity of the at least one second injection caused by a time-delayed reset of the injector needle of the injector is compensated.

The measures listed in the dependent claims advantageous refinements and improvements of the method specified in the independent claim are possible.

An advantageous embodiment provides for storing the dependencies of the shortening of the activation duration of the signal characterizing the electrical spray distance and of the signal characterizing the rail pressure in an injector-specific characteristic field. "Injector-specific" again means assigned to a series of injectors which show the same behavior within the tolerances for each injector.

The characteristic diagram is preferably determined by application to a test stand of the internal combustion engine or to the internal combustion engine itself.

list of figures

Embodiments of the invention are illustrated in the drawings and explained in more detail in the following description.

• 1 schematisch einen Magnetventil-Injektor, bei dem das erfindungsgemäße Verfahren bevorzugt zum Einsatz kommt;
• 2 schematisch den Einspritzverlauf einer Vor- und einer Haupteinspritzung bei einem solchen Magnetventil-Injektor;
• 3a die Einspritzmenge über der Differenz der zeitlichen Ansteuerdauer zweier aufeinanderfolgender Einspritzungen und eine Korrektur der Ansteuerdauer nach dem erfindungsgemäßen Verfahren;
• 3b schematisch ein Kennfeld zur Bestimmung der korrigierten Einspritzzeit der Haupteinspritzung;
• 4 schematisch den Einspritzverlauf anhand des Ansteuersignals und der Nadelöffnung nach einer Korrektur mit Hilfe des erfindungsgemäßen Verfahrens und
• 5 die eingespritzte Menge über der Differenz der zeitlichen Einspritzdauern zweier aufeinanderfolgender Einspritzungen nach erfolgter Korrektur mit Hilfe des erfindungsgemäßen Verfahrens.

Show it:

• 1 schematically a solenoid valve injector, in which the inventive method is preferably used;
• 2 schematically the injection curve of a pre-injection and a main injection in such a solenoid valve injector;
• 3a the injection quantity over the difference of the time control duration of two successive injections and a correction of the activation duration according to the method according to the invention;
• 3b schematically a map for determining the corrected injection time of the main injection;
• 4 schematically the injection curve based on the drive signal and the needle opening after a correction using the method according to the invention and
• 5 the injected amount over the difference of the time periods of injection of two successive injections after correction by means of the method according to the invention.

Embodiments of the invention

In 1 schematically a solenoid valve injector is shown, which is electrically controlled. The injector has a fuel return 101 , a magnetic coil 102 , a magnet armature 103 , a valve ball 104 , a valve control room 105 , a pressure shoulder of the nozzle needle 106 , a spray hole 107 , an outlet throttle 108 as well as a high pressure connection 109 , an inlet throttle 110 and a valve piston 111 on. The valve piston 111 lies on the actual nozzle needle 112 on. The operation of such a solenoid valve injector will be described below.

The fuel is from the high pressure port 109 via an inlet channel to the nozzle needle 106 as well as via the inlet throttle 110 in the valve control room 105 guided. The valve control room 105 is about the outlet throttle 108 passing through a solenoid valve, formed from magnetic coil 102 , Magnet armature 103 and valve ball 104 , can be opened, with the fuel return 101 connected.

In the closed state of the outlet throttle 108 the hydraulic force predominates on the valve piston 111 against the force on the pressure shoulder 106 the nozzle needle 112 , As a result, the nozzle needle becomes 112 pressed into its seat and closes the high pressure passage close to the engine compartment. When the engine is not running and there is no pressure in the rail (not shown), the nozzle spring closes the injector. When driving the solenoid valve, the outlet throttle 108 open. The inlet throttle 110 prevents complete pressure equalization, so that the pressure in the valve control room 105 and thus the hydraulic force on the valve spool 111 sinks. Once the hydraulic force those on the pressure shoulder 106 the nozzle needle 112 falls below, opens the nozzle needle 112 , The fuel now passes through the spray holes 107 in the combustion chamber of the internal combustion engine (not shown).

When the solenoid valve is no longer actuated, the armature becomes 103 pressed down by the force of the valve spring. The valve ball 104 closes the outlet throttle 108 , This builds up in the valve control chamber via the inflow of the inlet throttle 110 again a pressure as in the rail. This increased pressure exerts an increased force on the control piston 111 out, leaving the nozzle needle 112 closes again. The flow of the inlet throttle 110 determines the closing speed of the nozzle needle 112 ,

This indirect control of the nozzle needle 112 A hydraulic booster system is used because of the rapid opening of the nozzle needle 112 required forces can not be generated with the solenoid valve. The time required in addition to the injected amount of fuel control passes through the throttles of the control room 105 in the fuel return.

Today's common-rail systems are operated with a pressure of up to 2500 bar. In this case, a plurality of pilot injections, a main injection and optionally a plurality of post-injections are sold in order to achieve optimum exhaust emission values. It is necessary to make even very close accumulated multiple injections, which in turn require fast valve switching times, which, however, can not be readily realized with the above-described pressure balanced solenoid valve or with piezo valves. The mechanical return time of such valves is by injector-specific variables, such as axial stiffness of the nozzle needle 112 , Size of the inlet throttle 110 , Stiffness of the nozzle seat and the like determined. In other words, this reset time is injector-specific and therefore does not have to be determined for each injector. Rather, injectors of one type behave the same within predetermined tolerances. Such an injector now has the decisive disadvantage that the nozzle needle 112 is strongly biased at high rail pressures. This bias causes a certain time delay both when opening and when returning to the end of the injection. If an injection follows in time so close to a previous injection that the injector has not yet been reset, this results in an undesired additional quantity. This will be described below in connection with 2 explained.

In the upper graph of the 2 is the drive "e" of such an injector over the time "t" to achieve a pilot injection PI based on the curve 210 and to achieve a main injection MI based on the curve 220 shown. In the lower half of the 2 is the needle opening "n" over the driving time " ET "Represented by a curve 211 for the pilot injection PI and a curve 221 for the main injection MI , When the temporal spray distance TDIFF * the main injection MI from the pilot injection PI is chosen as large as in 2 shown, time delays occur when opening or resetting the valve needle 112 not in appearance. Again 2 it can be seen, a reset of the injector is carried out in the time interval that in 2 With 212 is designated. Then follows the main injection. However, if on the pre-injection within a much shorter time difference TDIFF before the main injection MI takes place as indicated by the curves 230 and 231 in 2 is shown, a needle opening already takes place when a provision of the nozzle needle 112 not yet done. An earlier opening causes the nozzle needle 112 opens wider, and thus opens longer, so that a larger injection quantity is injected, as it is based on the curve 231 in 2 becomes clear.

In 3 schematically a "TDIFF quantity wave" is shown. The curve 310 describes the total injection quantity of a pre / main injection pattern at a given rail pressure. The electric injection distance tdiff is varied here, while the activation time for the pilot injection PI and the main injection MI are constant. This feature basically shows three areas. In a first area I the injections are hydraulically "grown together" into a single injection. The hydraulic spraying distance of zero was undercut.

In one area III the injector is mechanically reset. An increased or reduced quantity is triggered by hydraulic pressure oscillations. The low frequency oscillations can be corrected with the pressure wave correction known per se and mentioned above.

In one area II the injector is mechanically not yet fully reset. This results in the pre-injection PI following main injection MI an extra amount in the manner described above. The correction is now carried out in a special, based on a linear approximation embodiment of the inventive method, to which this is not limited, however, characterized in that a quantity correction by shortening the injection time is achieved in the following manner: $${\text{Q}}_{\text{MI}\text{, corr}}=\mathrm{\Delta }\text{Q}/\mathrm{\Delta }\text{TDIFF}·\left(\text{MIN}\left(\text{TDIFF}\right)-\text{TDIFF}\_0\right),$$

The injection timing of the main injection is then calculated according to a linear approximation as follows: $${\text{ET}}_{\text{MI}\text{, corr}}=\mathrm{\Delta }\text{ET /}\mathrm{\Delta }\text{Q}·{\text{Q}}_{\text{MI}\text{, corr}},$$

It should be pointed out here, however, that the invention is not limited to a linear approximation, but that, purely in principle, a polynomial of higher order can also be used for the approximation.

The correction of the main injection is: $${\text{ET}}_{\text{MI}}={\text{ET}}_{0\text{MI}}+{\text{ET}}_{\text{MI}\text{, corr}},$$

It will be in the field II By linear approximation it is determined how the injection quantity "Q" increases as a function of a shortening of the spray distance TDIFF. The found quantity Q _MI , _corr is used to determine the correction of the drive duration ET _{MI, corr} in a map in which the relationship between the drive amount "Q" over the drive duration "ET" is stored. This map is schematic in 3b shown. This correction term is used to determine the injection duration " ET "The main injection by taking a splash distance TDIFF_0 is assumed, which represents the spray distance from which the provision of the injector is completed and a correction is no longer required.

Thus, there is a correction of the injection quantity determining control signal as a function of at least one of the electrical spray distance characterizing signal and a signal characterizing the rail pressure, because the correction takes place at a specific predetermined and prevailing rail pressure.

The result of an injection sequence corrected in this way is in 4 shown schematically, in the upper half of the control "e" of the solenoid valve injector over time and in the lower half of the needle opening "n" over the electrical drive time are shown. On a pre-injection PI passing through the bend 410 is represented, follows at a time interval TDIFF * a main injection MI * passing through the bend 420 or alternatively at a reduced time interval TDIFF another injection MI passing through the bend 430 is represented. The curves correspond 410 . 420 . 411 . 421 the curves 210 . 220 . 211 . 221 in 2 , Unlike the one in 2 However, due to the above-described correction method, no additional injection quantity is produced because, due to the correction of the control signal, the injection time of the main injection is shortened MI as it is based on the curve 430 is shown. In this case, therefore, despite the fact that the valve needle 112 not yet fully reset, while the main injection MI is already done, no Mehrehrspritzmenge made, as shown by the curve 431 is clearly visible.

In 5 is schematically the injection quantity " Q "On the time interval between injection of the two injections" TDIFF "Shown, with the curve 530 the injection amount over the time difference without the above-described correction and the curve 520 represent the injection quantity over the time difference of the electrical signals of the injection times with the above-described correction. As can be seen from these two curves is in the with II designated area a significant reduction in the injected amount by shortening the electrical drive time feasible, the area II - As already mentioned above - is the area in which the injector is mechanically not yet fully reset, so the area in which an extra amount of this not yet complete retraction of the injector occurs.

1. 1. Die Korrektur muss nur einmal für die entsprechende Baureihe durchgeführt werden und ist dann für sämtliche Exemplare gültig. Baureihe ist gleichzusetzen mit Injektortyp. Die Korrektur erfolgt also injektorspezifisch.
2. 2. Die Korrektur ist eine Funktion, die lediglich von dem elektrischen Spritzabstand und dem Raildruck abhängt.
3. 3. Die Korrekturfunktion ist nicht von der Kraftstofftemperatur und der Voreinspritzmenge abhängig.
4. 4. Der Mengenanstieg ist näherungsweise linear. Daher müssen nur wenige Prüfpunkte zur Erstellung der Korrekturfunktion bekannt sein.

The crucial idea of the invention is, in other words, in the field II an extra amount of main injection MI to correct over an electrical drive time reduction. This correction is therefore easy to carry out, because the extra amount in the area II mainly mechanically founded. This results in the following advantages over, for example, pressure wave corrections:

1. 1. The correction must be made only once for the corresponding series and is then valid for all copies. Series is equivalent to injector type. The correction is thus injector-specific.
2. 2. The correction is a function that depends only on the electrical spray distance and the rail pressure.
3. 3. The correction function does not depend on the fuel temperature and the pilot injection quantity.
4. 4. The increase in volume is approximately linear. Therefore, only a few checkpoints for creating the correction function need to be known.

The method described above can advantageously be implemented, for example, as a computer program on a computing device, in particular the control unit of the internal combustion engine, and run there. The program code may be stored on a machine-readable medium that the controller may read. In this way, updates and minor changes in the procedure are later "playable".

Claims (6)

Method for controlling an injection system of an internal combustion engine having at least one, in particular equipped with a low-pressure stage, electrically controllable injector, wherein a fuel metering in a first partial injection (PI) and at least a second partial injection (MI) is divided, one to be injected by means of at least one injector fuel amount-determining control signal in dependence is corrected at least from an electrical spraying distance characterizing signal and by a rail pressure characterizing signal, characterized in that the correction of the control signal to shorten the control period, the shortening of the control period is chosen such that by a time-delayed provision of an injector needle of the injector caused additional amount of at least one second injection (MI) is compensated. Method according to Claim 1 , characterized in that the dependencies of the shortening of the driving time of the signal characterizing the electrical spray distance signal and of the signal characterizing the rail pressure in a map (350) are stored. Method according to Claim 2 characterized by the following steps: - in a predeterminable range of the time difference of the injection times (TDIFF), a variable characterizing the increase of the injected quantity over the time difference of the injection times of the first partial injection (PI) and the second partial injection (MI) is approximated - depending on the increase of the injected amount, a correction term of the injection timing of the main injection (ET _MI , _corr ) is determined by the map (350) - the injection timing of the main injection is corrected by the main injection timing correction term (ET _MI , _corr ). Method according to Claim 3 , characterized in that the functional relationship between the quantity characterizing the injected quantity over the time difference of the injection times is determined by linear approximation or by an approximation by means of a polynomial of higher order. Method according to one of Claims 2 to 4 , characterized in that the characteristic field is determined by application to a test stand of the internal combustion engine or to the internal combustion engine itself. Computer program that is trained all the steps of a procedure according to one of Claims 1 to 5 execute when it runs on a computing device, in particular the control unit of an internal combustion engine.

Images