DE102015202727A1 - Method for controlling the injection quantity of a reducing agent into the exhaust gas tract of an internal combustion engine - Google Patents

Abstract

The invention relates to a method and a corresponding metering system for controlling the output of a metering injection quantity of reducing agent, which is metered into the exhaust gas of an internal combustion engine upstream of a NOx storage catalytic converter to the exhaust gas via the metering system, via a control device a required target metered amount of the reducing agent and an actually metered actual metered amount of the reducing agent is determined and wherein the actual metered amount is compared with the desired metered amount of the reducing agent. A high accuracy of the injection quantity is obtained by determining a correction variable from the comparison of the actual and the set metering quantity and that the injection quantity is controlled taking into account the correction variable.

Description

State of the art

The invention relates to a method and a corresponding metering system for controlling the output of a metering injection quantity of reducing agent, which is metered into the exhaust gas of an internal combustion engine upstream of a NO _x storage catalytic converter to the exhaust gas via the metering system, via a control device, a required nominal metering of the Reducing agent and an actually metered actual metered amount of the reducing agent is determined and wherein the actual metered amount is compared with the target metered amount of the reducing agent.

In exhaust after-treatment systems of lean-burn engines, nitrogen oxide (NO _x ) reduction components are provided to meet increasingly stringent NO _x limits. One possibility is the use of a so-called NO _x storage catalytic converter, in which nitrogen oxides are stored continuously during lean operation and thus removed from the exhaust gas. For regeneration, a reducing environment is set in the exhaust gas, in which the stored nitrogen oxides are reduced and the memory is released for a new storage cycle. A reducing environment can be created either by adjusting a fuel-rich operation of the internal combustion engine or by adding fuel as a reducing agent in the exhaust gas. The latter has the advantage that the regeneration of the NO _x storage catalytic converter can proceed largely independently of engine operation. Such systems require a separate metering system with which the injection quantity of fuel injected into the exhaust gas can be set as accurately as possible.

In order to ensure a trouble-free reduction agent addition in the exhaust tract, the DE 601 05 917 T2 a diagnostic device for detecting various defects in a reducing agent supply device, which is used for injecting a reducing agent into the exhaust gas system arranged in the NO _x catalyst. In the method, the pressure of the reducing agent is detected when supplied to an injection unit, which together with the injection time results in the injected amount of the fuel. In addition, the actual state of the exhaust gas is detected, for example, via the oxygen concentration in the exhaust gas. The amount of reducing agent determined by means of the pressure can be assigned to specific setpoint values of the exhaust gas state via known relationships. These target values are compared with the actual values of the exhaust state and, in the event of a deviation, are concluded of a shortage in the reductant supply device. Detailed measures to eliminate the deviations are not specified.

The DE 60 2004 000 285 T2 discloses a method for detecting an error in the reductant detection based on an air / fuel ratio. Here, an actual amount of supplied fuel as a reducing agent is estimated from a change amount of an air-fuel ratio. The estimated amount is compared with a desired amount of reducing agent. If the two values deviate, a feed error is detected. The compensation of the supply error is not the subject of the disclosure.

The present invention is based on the object of providing a method and a corresponding system for metering the injection quantity of a reducing agent into the exhaust gas tract with a high accuracy of the injection quantity.

Disclosure of the invention

The object is achieved for the method with the features of claim 1 and for the metering system with the features of claim 12. In the method, it is provided that a correction variable is determined from the comparison of the actual and the set metering quantity, and that the injection quantity is controlled taking into account the correction variable.

In the metering system it is provided that a comparison unit is provided, by means of which a correction variable can be determined from the comparison of the actual and the desired metering quantity, and that a control device is formed, by means of which the injection quantity can be regulated taking into account the correction variable.

By means of the method according to the invention and the metering system according to the invention, different influencing factors on the injection or metering quantity can be compensated by means of a control, so that the accuracy of the metered addition can be increased. The function of the control can be advantageously realized at least partially, for example in the control device, which can also be designed for other control and / or regulating functions. Influences on the injection quantity can be, for example, environmental influences such as ambient temperature, fuel type (viscosity / density and, for example, their change over the temperature) and / or the inlet pressure. In addition, influences on individual components of the dosing system and thus on the injection quantity can be affected. These can, for example, manufacturing / accuracy tolerances, Aging effects, pollution etc. According to the invention, it is not absolutely necessary to know the sources of interference, for example if it is the sum of deviations of several components due to production-related and / or age-related measurement inaccuracies. However, knowledge of the source (s) may be beneficial. The correction variable can be determined, for example, from a deviation of a controlled variable resulting from a setpoint / actual value comparison via a mathematical relationship or a link via a table. The controlled variable is eg the injection quantity or a related variable. With the correction value, a manipulated variable for controlling the injection quantity is adjusted. The calculation rule can vary, eg depending on the control strategy and / or the manipulated variable. It would also be conceivable to extrapolate correction quantities under certain influences. By means of the regulation according to the invention, an adequate reduction of the emissions over the entire service life of a motor vehicle can be ensured.

According to the invention, it is preferably provided that the actual metered amount of the reducing agent is determined as a function of a storage volume change in a pressure accumulator of a pump storage unit of the metering system. Such a determination of the actual metered amount is therefore possible in particular in the case of a metering system having a pressure accumulator of a pump storage unit. Thus, the actual metered amount can be determined independently of the exhaust state. Alternatively or additionally, the actual metered quantity can be determined as a function of a conversion rate determined via an NO _x sensor arranged downstream of the NO _x storage catalytic converter. In particular, an underdosage of reducing agent can be detected by means of the control device via the NO _x sensor, since in this case there is an inadequate conversion rate of NO _x in the exhaust gas, so that NO _{x is} detected. A combination of the conversion rate determined via the NO _x sensor with one or more other methods for determining the actual metered amount would also be conceivable.

If the actual metered amount of the reducing agent is determined as a function of the storage volume change, an advantageous variant of the invention is that the storage volume change in the pressure accumulator is determined via a pressure difference formed from an initial pressure and a final pressure of the withdrawal period. The withdrawal period corresponds e.g. the duration of, or possibly even more, injection process and is preferably determined and specified via the control device. For pressure determination comes a, preferably attached to the accumulator of the pump storage unit mounted pressure sensor used. For example, when using a spring accumulator as pressure accumulator on the known accumulator volume and the spring rate of a pressure accumulator spring, taking into account the frictional forces in the piston guide, be closed to the injected volume. It is advantageous here that due to the formation of differential pressure, an offset of the pressure sensor is eliminated and thus a more accurate pressure measurement than when using an absolute pressure can be obtained.

Alternatively or additionally, e.g. for control purposes, it may be provided that the storage volume change is determined via the stroke position of a storage piston pressure-independent. Here, with the attachment of an inductive or capacitive sensor to the accumulator contactless piston stroke and its change during an injection process, and thus also the injected volume can be determined. This makes it possible to dispense with measured values of a pressure sensor, so that this procedure can be carried out, for example. also allows a check of a determined using a pressure sensor injection quantity.

A desired dosing quantity can be obtained particularly simply if it is determined speed-dependently and torque-dependent from a map stored in the control device. Advantageously, this map has been previously described e.g. determined experimentally on a test bench and stored in the control device. It would also be conceivable that the inventive method with the metering additionally or alternatively added fuel for the regeneration of a diesel particulate filter (DPF) in the exhaust system. For this purpose, a separate map should be determined and deposited in the control device.

Preferably, the metered addition of the injection quantity can take place by specifying a pressure difference in the pressure reservoir of the metering system. In this case, the required target dosing is converted into a change in volume of the storage volume and this in turn, taking into account the spring constant, friction, etc., in a corresponding pressure difference. Now, depending on a given initial pressure, a withdrawal period can be chosen so long, ie the valve remain open until a certain final pressure has been set according to the predetermined pressure difference in the memory. As a further possibility, an initial pressure in the pressure accumulator can be determined (for example, calculated or taken from a stored characteristic map) with which the required pressure difference is established over a constant withdrawal period. This specific initial pressure is then set in the pressure accumulator and, starting therefrom, with the aid of the pressure sensor and the control device, reducing agent is metered into the exhaust gas tract over the constant withdrawal period. Alternatively to the specification of the pressure difference, the Metering of the injection quantity advantageously via specification of the opening duration of an injection device of the metering done. The injection device preferably has a solenoid valve injector. In this case, it is particularly simple if activation periods are stored as a function of the accumulator pressure in an empirically determined characteristic map which is necessary in order to provide a certain injection quantity. Under knowledge and / or assumption of specific (substance) data of the reducing agent, the activation duration of the injection device could also be calculated.

For a sufficiently rapid pressure build-up in the pressure accumulator in order to provide a required target metering, it is advantageous if the control of a pump of the pump storage unit is performed so that a pump piston of the pump reaches the rear end position during a power stroke. This is especially true when a solenoid pump is advantageous for use. The rear end position forms a so-called Mechanical Stop Point (MSP). By reaching the rear end position, the pump can deliver the maximum possible amount per working stroke into the pressure accumulator. Another way to provide a faster pressure build-up in the accumulator is to increase the drive frequency of the pump. The pump thus performs more strokes per unit of time and can fill the reservoir faster.

To ensure that the rear end position of the pump piston is reached, this is preferably monitored by means of a current signal analysis and / or a pressure signal analysis. In this case, an induced by the movement of the pump piston current waveform is recorded for the current signal analysis, from the course of which conclusions can be drawn on the piston speed and in particular on the MSP. A corresponding method is in the not yet published patent application DE 10 2014 223 066 the applicant described. Alternatively or additionally, the pressure signal can be recorded. In this case, the MSP can be detected, for example, via an oscillation in the course of the pressure.

In order to avoid an overload of the control device used for the metering, it can be provided that pumping operations of the pump are controlled by means of the control device such that they do not overlap with metering operations in terms of time. Otherwise, pumping operations can be controlled independently of dosing operations. In particular, it may be necessary in a non-leak-free system to pump periodically even without metering event in order to maintain a certain storage pressure.

An advantageous control of the injection device is achieved in that a valve of the injection device is energized during a valve drive with an opening current and then with a relation to the opening current by at least 10% reduced holding current. The valve is in particular a solenoid valve and the valve actuation duration is correlated or corresponds to the opening duration via a factor. The high opening current leads to a quick opening of the valve. In addition, if the valve does not fully open due to prevailing ambient conditions, gain may be integrated, for example, via a separate power supply (e.g., a condenser in the valve control unit). This provides for a short period of time, e.g. 2-4 ms, a higher voltage available, which results in a faster current rise and thus in a faster opening of the valve. Depending on the configuration of the gain, higher opening currents may occur depending on the amplification voltage. The holding current keeps the valve open. Following the valve actuation period, a negative voltage, for example the negative battery voltage, is preferably applied in order to ensure rapid demagnetization of a solenoid coil of the valve and thus fast closing.

For a regulation of the injection quantity with a desired actual value deviation, the injection quantity is preferably adjusted by varying the opening duration of the injection device. Alternatively, the injection quantity is adapted via a change in the pressure, in particular the initial pressure, in the pressure accumulator. The latter is particularly advantageous in a metering system with a pump storage unit applicable because here the pressure upstream of the injector regardless of a system pressure of another, e.g. superior system in the vehicle via the native pump is controllable. It would also be conceivable to combine the two measures, but the regulation would be correspondingly expensive.

Advantageous design options for the control result when the correction variable is stored in the control device and / or formed depending on the environmental conditions different correction variables and stored in the control device. Thus, following the determination according to the invention of a correction variable according to the invention, these can simply be resorted to and a set metering quantity can be corrected by means of the correction variable. A particularly precise control can be obtained if, depending on different operating states or operating state regions of the internal combustion engine, for example speed and torque-dependent, different correction values are stored. The determination of a correction variable according to the invention can also be carried out, for example, for checking during each dosing process, in which already determined correction values can be adjusted and overwritten in the case of a change, for example due to aging effects of components of the dosing system or of the NOx storage catalytic converter. In this way, a learning control can be obtained. It can also be provided to carry out checks of the correction quantities only at certain time intervals or as a function of events, for example maintenance. If, during an injection cycle, no setpoint-actual value comparison with subsequent correction-variable formation is carried out, the regulation can nevertheless resort to the already stored correction variables and carry out a correction of the setpoint dosing quantity. Irrespective of the deposit of one or more correction variables, it is conceivable that an evaluation of the deviation is made in the setpoint actual value adjustment. If, for example, the deviation is outside a certain tolerance range, an error message could be output that indicates a defect in the system. Such could arise, for example, when the valve of the injector is blocked and no regulation of the injection quantity is possible.

The method according to the invention can be used particularly advantageously with a metering system which has a pump storage unit with a pump and a pressure accumulator. Preferably, the dosing system additionally comprises other components, such as a check valve, one or more pressure sensors and the like.

The invention will be explained in more detail by means of embodiments with reference to the drawings. Show it:

1 1 is a schematic illustration of a metering system according to the invention for carrying out a method according to the invention, arranged in an overall system,

2 a block diagram for providing an injection quantity without control,

3a , b is a block diagram of a control according to the invention ( 3a ) and exemplarily simplifies the pressure curve in a pressure-variable control ( 3b )

4 Pressure and current waveforms of a pump of a pressure storage unit for determining a rear end position of the pump piston,

5 Control voltage and current waveforms of a valve of an injector in an exemplary driving strategy and

6 a current waveform of a valve of the injector during a drive time.

1 schematically shows a metering system according to the invention 3 for carrying out the method according to the invention and suggestively its embedding in the overall system of a motor vehicle. For this purpose, as further subsystems, a low-pressure circuit 1 of fuel with a fuel tank 10 , as well as an internal combustion engine 2 shown, wherein the internal combustion engine 2 is formed in the present example of a diesel engine. The low pressure circuit 1 is located upstream of the dosing system 3 , Downstream of the internal combustion engine 2 is an exhaust tract 4 with an exhaust pipe 40 arranged, by means of the internal combustion engine 2 formed exhaust gas through an exhaust aftertreatment system 44 to be led. The exhaust aftertreatment system 44 includes a NO _x storage catalyst 44.1 and a diesel particulate filter disposed downstream therefrom 44.2 (DPF). In addition, other, not shown components in the exhaust aftertreatment system 44 be arranged, such as an oxidation catalyst. It would also be an overall system with a lean-burn gasoline engine and a corresponding exhaust aftertreatment system 44 conceivable, which is also a NO _x storage catalytic converter 44.1 can have. In the exhaust pipe 40 In addition, there are various sensors, such as temperature sensors 42 , a NO _x sensor 43 for the detection of nitrogen oxides in the exhaust gas downstream of the NO _x storage catalytic converter 44.1 and one to the DPF 44.2 assigned differential pressure sensor 41 arranged. Other sensors may be present, eg at least one lambda probe. The overall system is furthermore a control unit 8th assigned to which the electrical components are connected via signal lines.

The structure of the dosing system 3 corresponds in its essential components to such, as for example in the DE 10 2012 204 385 A1 the applicant is indicated. The dosing system 3 The metered addition of fuel (diesel in the present example) serves as a reducing agent in exhaust gas upstream of the NO _x storage catalytic converter 44.1 by means of an injection device 35 , Also a use of the reducing agent for the regeneration of the DPF 44.2 could be possible. For metering points the injector 35 of the dosing system 3 for example, one in the exhaust pipe 40 pointing solenoid valve or similar injector (eg piezo) with a valve on, in this case, for example, a solenoid valve 35.1 is. The solenoid valve injector allows a fine atomization of the reducing agent and thus a good distribution in the exhaust. For injection of the reducing agent via the injection device 35 a high pressure is required, which is at least higher than the exhaust gas pressure, and is suitable for metering a sufficient amount of reducing agent with a corresponding atomization at the valve in the exhaust line. For example, pressures of more than 10 bar, eg 30 bar, suitable. The appropriate pressure is from a pump storage unit 30 provided via a reducing agent line 34 with the injector 35 connected is.

The pump storage unit 30 is in a housing 31 arranged and includes a pressure accumulator 33 and an upstream of the pressure accumulator 33 in the reducing agent line 34 arranged pump 32 , The pump 32 , which may be in particular a solenoid drive, brings the from the low pressure circuit 1 taken fuel to the high pressure required for its use as a reducing agent. Thus, the pressure of the reducing agent for the control is independent of a total pressure controllable. Via a check valve 34.1 the reducing agent enters the pressure accumulator 33 , which is provided here as a spring store. Other pressure accumulator systems would be possible, for example hydraulic accumulators. In the accumulator 33 becomes the pressure level of the reducing agent via a storage piston 33.1 that with a storage piston spring 33.2 is maintained. The accumulator 33 is also a pressure sensor 33.3 assigned, which measures the accumulator pressure and to a control device 36 of the dosing system 3 transfers. If the pressure level is below a certain setpoint, the accumulator can 33 over the pump 32 be charged, ie the pressure can be increased. The pump 32 , the accumulator 33 as well as the injection device 35 and optionally other components of the dosing system 3 stand with the control device 36 in data transmission connection. The control device 36 is preferably in the higher-level control unit 8th integrated and also has a control device 58 with a comparison unit 37 on.

2 schematically shows a control of the required injection quantity initially without regulation, as preferred via the control device 36 running. As input variables are, for example, that of the pressure sensor 33.3 transmitted current initial pressure 53 in memory before, as well as the exhaust gas to be supplied to the target dosing 51 at reducing agent. The nominal dosage 51 For example, it was speed and torque dependent from one in the control device 36 determined map determined. These and / or possibly other variables are a manipulated variable determination 54 fed. The manipulated variable determination 54 may contain, for example, empirical maps and / or calculation rules, via which a corresponding manipulated variable to be set 55 is determined to the requested target dosing 51 adjust. Accordingly, the injector becomes 35 controlled and an actual dosing 52 injected into the exhaust pipe.

Conceivable are different, for example, control strategies with time or pressure control. In the strategy with time control, the injection quantity is controlled over a control period 76 (see. 5 and 6 ) as a manipulated variable 55 the injector 35 for opening the solenoid valve 35.1 depending on the accumulator pressure, in particular the initial pressure 53 , The determination of the activation period 76 , which is necessary to provide a certain amount of injection, depending on the initial pressure 53 take place via an empirically determined map. Also a calculation would be conceivable.

In a pressure control strategy, control is preferably via one of the initial pressures 53 and the final pressure in the accumulator 33 (after a withdrawal period) formed pressure difference 61 (see. 3b ) as a manipulated variable 55 , In this case, the required target dosing 51 in a corresponding pressure difference 61 converted. The valve is opened and as soon as the final pressure to form the desired pressure difference 61 is reached, it is closed again. This requires monitoring of the pressure curve during the injection period, preferably via the control device 36 , This strategy may vary depending on the initial pressure 53 and / or influenced by influencing variables, different opening times of the valve. It is conceivable that in the event that the desired pressure difference 61 does not set over a maximum (determinable) period, an error is issued. For example, in this case there could be a defect, for example jamming, of the valve. If the opening time should be kept constant, the injection quantity control would also be above the initial pressure 53 as a manipulated variable 55 conceivable. By a variation of the initial pressure 53 in accumulator 33 Injection quantities can be varied at constant valve opening time, so that in this way different, depending on the engine operating condition, desired metered quantities 51 and related pressure differences 61 can be reached at a constant opening time. The change of the pressure difference 61 by increasing the initial pressure 53 is in 3b illustrated. The possibility of control with reference to the pressure difference 61 arises in particular in the metering system 3 with the accumulator 33 and the native pump 32 , which allows a decoupled from a higher-level system pressure control.

3a provides a block diagram with a control of the injection quantity according to the invention by means of the control device 36 The regulation runs via a regulating device 58 preferably in the control device 36 contained or integrated. It is first as in 2 a manipulated variable 55 determined. The determined manipulated variable 55 can now contact a rule member 58.1 are passed, where it is determined whether a correction factor K, which may also depend on the operating conditions, has already been determined. In this case, preferably, the corresponding correction quantity K from a correction amount memory 57 taken, for example, can contain a corresponding table or a map. The manipulated variable 55 is now corrected by the correction quantity K and the corrected manipulated variable 55 the injector 35 specified. Subsequently, the actual injected metered dose is 52 determined by a procedure A, B or C, whereby also several procedures eg for the examination can be combined. Other approaches would be conceivable.

In the procedure A, the storage volume change in the accumulator 33 from the pressure difference 61 (formed between the start and end pressure of the sampling period) determines the actual dosage 52 equivalent. In the procedure B, the storage volume change over the stroke position of the accumulator piston 33.1 independent of pressure. For this purpose, an inductive and / or capacitive sensor come to the pressure accumulator 33 For use, via the contactless piston stroke and thus also the injected volume can be determined. In the procedure C, the actual metered amount 52 depending on a downstream of the NO _x storage catalytic converter 44.1 arranged NO _x sensor 43 determined turnover rate. Because with the different procedures the actual metering quantity 52 is calculated from different sizes, it may be useful to perform them at least partially or temporarily redundant to make them mutually plausible. A direct flow measurement would also be conceivable, for example by means of MFC, which would, however, mean an additional component outlay.

The determined in this way actual dosage 52 becomes the comparison unit 37 for a setpoint-actual value comparison 50 in fed. If a deviation is detected, is via a control element 58 .1 the correction value K adjusted or redetermined if there was no correction size K present. The originally determined, uncorrected manipulated variable can also be included in the determination 55 be included, which is also the rule member 58.1 can be supplied. The correction quantity K becomes in the correction amount memory 57 deposited, where it can overwrite an already stored correction size K. This allows a learning and constantly adaptive regulation. The actual manipulated variable can also be checked for checking 56 the regulator 58.1 be supplied.

Also at the in 3a shown different strategies come into question, eg based on the respect 2 described tax strategies. For example, in a time-controlled control, the activation duration 76 the corrected manipulated variable 55 be and according to the correction quantity K on the drive time 76 Respectively. In the case of pressure-controlled control with a constant opening duration, the actuating variable is 55 by means of the correction quantity K the initial pressure can be corrected, with variable opening duration the pressure difference 61 , 3b illustrates in a representation of a printing axis 59 over a timeline 60 the effect of changing the initial pressure 53 , which by means of a correction quantity K to an initial pressure 53 ' is corrected. If this is set, there is one opposite the pressure difference 61 increased pressure difference 61 ' over the same period. In addition, a plausibility check of the pressure value may make sense in a pressure-controlled control. For example, is the value of the pressure sensor 33.3 outside its measuring range, an error can be reported via a diagnostic error check.

For the regulation of the invention, the pump should 32 the accumulator 33 fill so fast that the pressure level required for the next injection process, or at least the required volume is available on demand. The fastest pressure build-up takes place when the pump 32 at each stroke a rear end position 65 of the pump piston, a so-called Mechanical Stop Point, MSP achieved. Here then promotes the pump 32 the maximum possible amount per stroke in the accumulator 33 , In addition, a faster pressure build-up can be achieved by increasing the pumping frequency. Information in this regard may be from a high frequency sampled pressure and current signal from the pump 32 to be taken as they are in 4 are shown. This leaves the pressure signal of the pump 32 represented in the upper diagram with the pressure axis 59 over time 60 , on the one hand, that of the pump 32 detected pressure level. The lower diagram shows the signal of the pump current 66 over the timeline 60 , In this case, the current signal course over the pump injection cycle 62 a characteristic course. In particular, the branches differ during the pump drive period 63 and the pump piston return path 64 , From the pressure and current signal conclusions can be drawn on the piston speed and its position. A method, as can be seen from the current signal of the MSP is, for example, in the aforementioned patent application not yet published DE 10 2014 223 066 the applicant described. The MSP can also be controlled by an oscillation in the pressure signal, as in 4 , upper diagram is recognizable, can be detected. For the detection of the oscillation, a bit mask can be used, wherein for the behavior of the pressure, different bits are set, each corresponding to a condition (eg at inflection points of the pressure curve). If all conditions are met, the bit mask is compared to a previously defined MSP mask. At agreement is the MSP, more exactly the rear end position 65 , detected. Both methods can be used for a mutual plausibility check.

If it is determined during the current and / or pressure signal analysis that the MSP is not reached, the control time of the pump and / or the control voltage can be increased on the control side. An increase in the activation time of the pump causes the pump 32 even at higher pressures can perform the full stroke (the higher the accumulator pressure, the slower the piston speed). Increasing the voltage can speed up the current and allow the MSP to be reached faster. An additional measure for a faster pressure build-up is the increase of the pumping frequency.

To control the injection device, a so-called peak and hold control is advantageously used in the control according to the invention. A corresponding exemplary voltage and current profile is in 5 shown. The upper diagram shows a valve voltage 70 , the lower one corresponding valve current 71 over the timeline 60 , In each case a valve activation duration is shown 76 , which is divided into an opening current duration 74 with provision of an opening current 72 and a holding current duration 75 , where a holding current 73 provided. The high opening current 75 leads to a quick opening of the valve. The holding current reduced by more than 50% in this example is sufficient to keep the valve open. Obtained is such a course of the valve current 71 by an appropriate timing of the voltage, eg by pulse width modulation (see upper diagram). After completion of the valve control, the negative battery voltage is applied to ensure rapid demagnetization of the coil and thus a fast closing of the valve.

When controlling, it should also be noted that the valve activation duration 76 does not exactly correspond to its opening time. This illustrates a in 6 illustrated valve current signal 77 over the timeline 60 , The figure illustrates that the valve opening duration 78 , within which the valve is at least almost fully open, is shorter than the valve drive duration 76 , This influence can be influenced by factors such as in a map in the control device 36 are deposited, corrected.

The method according to the invention with the control involving a correction variable K is particularly suitable for use with the metering system 3 suitable, the described pump storage unit 30 having. With such a dosing system 3 arise manifold control strategies, in particular on the inclusion and / or adaptability of the pressure and / or pressure differences 61 , This refers both to the manipulated variables and to the possibility of the actual metered quantity 52 to investigate. The individual possible methods can be executed redundantly for a mutual plausibility check or verification. In this way, a high accuracy of the injection quantity of reducing agent can be obtained.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 60105917 T2 [0003]
• DE 602004000285 T2 [0004]
• DE 102014223066 [0015, 0038]
• DE 102012204385 A1 [0029]

Claims (15)

Method for controlling the dosing system ( 3 ) delivered injection quantity of reducing agent into the exhaust tract ( 4 ) an internal combustion engine ( 2 ) upstream of a NOx storage catalyst ( 44.1 ) the exhaust gas via the metering system ( 3 ) is metered in, wherein a control device ( 36 ) a required nominal dosage ( 51 ) of the reducing agent and an actually metered actual metered amount ( 52 ) of the reducing agent is determined and wherein the actual metered amount ( 52 ) with the nominal dosage ( 51 ) of the reducing agent, characterized in that a correction variable (K) is determined from the comparison of the actual and the nominal metering quantity and that the injection quantity is controlled taking into account the correction variable (K). Method according to claim 1, characterized in that the actual metered quantity ( 52 ) of the reducing agent as a function of a storage volume change in an accumulator ( 33 ) a pump storage unit ( 30 ) of the dosing system ( 3 ) and / or in dependence on a downstream of the NOx storage catalytic converter ( 44.1 ) arranged NOx sensor ( 43 ) determined turnover rate. A method according to claim 2, characterized in that the storage volume change in the accumulator ( 33 ) via an initial pressure ( 53 ) and a final pressure of the withdrawal period formed pressure difference ( 61 ) and / or the stroke position of a storage piston ( 33.1 ) is determined independently of pressure. Method according to one of the preceding claims, characterized in that the desired dosing amount ( 51 ) from one in the control device ( 36 ) stored map speed-dependent and torque-dependent is determined or determined. Method according to one of the preceding claims, characterized in that the metered addition of the injection quantity by presetting a pressure difference ( 61 ) in the accumulator ( 33 ) of the dosing system ( 3 ) or by specifying an opening duration of an injection device ( 35 ) of the dosing system ( 3 ) he follows. Method according to one of claims 2 to 5, characterized in that the control of a pump ( 32 ) the pump storage unit ( 30 ) is guided so that a pump piston of the pump ( 32 ) at a working stroke, the rear end position ( 65 ) reached. A method according to claim 6, characterized in that reaching the rear end position ( 65 ) of the pump piston is monitored by means of a current signal analysis and / or a pressure signal analysis. Method according to one of claims 6 or 7, characterized in that pumping operations of the pump ( 32 ) by means of the control device ( 36 ) are controlled so that they do not overlap with dosing over time. Method according to one of the claims 5 to 8, characterized in that a valve of the injection device ( 35 ) during a valve actuation period ( 76 ) with an opening stream ( 72 ) and then with an opposite to the opening stream ( 72 ) by at least 10% reduced holding current ( 73 ) is energized. Method according to one of claims 5 to 9, characterized in that the injection quantity by varying the opening duration ( 78 ) of the injection device ( 35 ) and / or a change in the pressure in the accumulator ( 33 ) is adjusted. Method according to one of the preceding claims, characterized in that the correction variable (K) in the control device ( 36 ) and / or depending on the environmental conditions different correction quantities (K) are formed and stored in the control device ( 36 ) are deposited. Dosing system ( 3 ) for an exhaust aftertreatment system ( 44 ) for controlling the injection quantity of a reducing agent, which in the exhaust gas tract ( 4 ) an internal combustion engine ( 2 ) upstream of a NO _x storage catalytic converter ( 44.1 ) is metered into the exhaust gas, with at least one control device ( 36 ), a reducing agent line ( 34 ) and an injection device ( 35 ), whereby a required nominal dosage ( 51 ) of the reducing agent and an actually metered actual metered amount ( 52 ) of the reducing agent can be determined and wherein the actual metered amount ( 52 ) with the nominal dosage ( 51 ) of the reducing agent, characterized in that a comparison unit ( 37 ) is present, by means of which a correction variable (K) can be determined from the comparison of the actual and the nominal metering quantity, and in that a regulating device ( 58 ) is formed, by means of which the injection quantity is controllable taking into account the correction variable (K). Dosing system ( 3 ) according to claim 12, characterized in that the dosing system ( 3 ) a pump storage unit ( 30 ) with a pump ( 32 ) and a pressure accumulator ( 33 ) having. Dosing system ( 3 ) according to claim 13, characterized in that the actual dosage amount ( 52 ) of the reducing agent in dependence on a storage volume change in the pressure accumulator ( 33 ) the pump storage unit ( 30 ) of the dosing system ( 3 ) via one in the control device ( 36 ) predetermined withdrawal period and / or depending on a downstream of the NO _x storage catalytic converter ( 44.1 ) arranged NO _x sensor ( 43 ) in connection with the control device ( 36 ) determined turnover rate can be determined. Dosing system ( 3 ) according to claim 14, characterized in that the storage volume change in the accumulator ( 33 ) via an initial pressure ( 53 ) and a final pressure of the withdrawal period formed pressure difference ( 61 ) or via a stroke position of a storage piston ( 33.1 ) can be determined independently of pressure.

Images