DE102019007085B4 - Method for operating a delivery metering system for a fluid, delivery metering system and motor vehicle with such a delivery metering system - Google Patents

Abstract

Method for operating a delivery metering system (10) for a fluid (1), preferably a delivery metering system (10) for a reducing agent for exhaust gas aftertreatment, the delivery metering system (10) having a pump (2) for delivering the fluid (1 ) and a dosing device (3), which is fluidically connected to the pump (2) via a pressure line (4); wherein via the pump (2), preferably via a speed variation of the pump (2), a fluid pressure pfluid of the fluid ( 1) in the pressure line (4) is regulated to a target fluid pressure value psoll; and wherein the setpoint fluid pressure value psollin is changed as a function of an actuation of the dosing device (3); characterized by i. an adaptation of a first fluid pressure gradient estimator G1 to a drop in the fluid pressure pfluid, preferably due to extraction, when the dosing device (3) is open and changing the target fluid pressure value psoll on the basis of the first fluid pressure gradient estimator G1; and/orii. an adaptation of a second fluid pressure gradient estimator G2 to a preferably pump-related increase in the fluid pressure pfluid when the dosing device (3) is closed and changing the setpoint fluid pressure value psoll on the basis of the second fluid pressure gradient estimator G2,

Description

The invention relates to a method for operating a delivery metering system for a fluid, preferably a delivery metering system for a reducing agent for exhaust gas aftertreatment. Furthermore, the invention relates to a corresponding delivery metering system and a motor vehicle with such a delivery metering system.

In order to reduce the NOx emissions from internal combustion engines, in particular from diesel internal combustion engines, SCR catalytic converters are mostly used today for exhaust gas aftertreatment. A reducing agent (e.g. ammonia and/or aqueous urea solution) is added to the exhaust gas flow via suitable metering devices, which then reduces the nitrogen oxides contained in the exhaust gas to nitrogen in the downstream SCR catalytic converter. To meter the reducing agent, it is usually conveyed via a pump from a storage tank to one or more metering valves and is also brought to a predetermined metering pressure. This can be defined, for example, as a function of an operating variable of the internal combustion engine (e.g. the current speed) and/or a parameter of the exhaust gas (e.g. the current exhaust gas temperature).

In order to ensure a metering pressure that is as constant as possible, it is also known in the prior art to regulate the hydraulic pressure in the metering system. B. can be done via a corresponding variation of the pump speed.

For example, the DE 11 2012 003 259 T5 a method for reducing nitrogen oxides in an exhaust gas flow of an internal combustion engine by a reducing agent, the method comprising the steps of: determining a first temperature of the exhaust gas flow; setting a target pressure of the reductant to a first pressure or a lower second pressure based on a comparison between the first temperature and a predetermined injector activation temperature threshold; determining an operating condition of the internal combustion engine, including an "on" or an "off"condition; operating a reductant pump to deliver the reductant at the first pressure or at the second pressure to an injector when the operating state is the “on” state; setting the target pressure of the reducing agent to the second pressure when the operating state of the internal combustion engine is the “off” state; determining a second temperature of the exhaust system; determining a cool down time based on the second temperature; and operating the reductant pump for the cool down time based on the second temperature being greater than a threshold and the operating state being the “off” state.

Furthermore, from the DE 10 2014 220 150 A1 a method for controlling the pressure of an SCR denoxing system of a motor vehicle is known. It is intended to use a self-optimizing, flatness-based pilot control model for the controlled system, which adapts to the respective vehicle configuration and manages with fewer than 5 application parameters, with the flatness-based pilot control being based on a system identification. The pre-control model makes it possible to use a very simple PI controller that only corrects the deviation between a model and a controlled system as well as measured and disturbance variables. This results in a high degree of control that protects engine components and minimizes energy consumption. In addition, the pilot control is decoupled from the regulation.

The DE 11 2017 002 054 T5 discloses a diesel exhaust fluid (DEF) delivery system and methods of operation thereof. The method includes determining an operating pressure by an electronic processor and receiving a system pressure from a pressure sensor of the DEF delivery system. The method further includes determining, by the processor, a dosing request and a pressure disturbance based on the dosing request. The method further includes determining, by the processor, a control request based on the system pressure and the operating pressure and a feedforward control value based on the pressure disturbance. The method further includes generating, by the processor, an adjusted control request based on the control request and the feedforward control value. The method further includes controlling, by the processor, a dosing valve of the DEF delivery system based on the dosing request and a pressure adjustment component of the DEF delivery system based on the adjusted control request.

Furthermore, from the DE 11 2018 001 905 T5 a reducing agent dosing system with the following components is known: a doser, a fixed displacement pump in fluid communication with the doser; a source of reductant in fluid communication with the fixed displacement pump; and a controller communicatively coupled to the fixed displacement pump for controlling operation of the fixed displacement pump. The controller is programmed to operate the fixed displacement pump using a pressure control system in response to data indicating that the doser is not dosing reductant. Furthermore, the controller for operating the Fixed displacement pump programmed using a flow control system in response to data indicating that the doser is dosing reductant.

In addition, the DE 10 2004 018 221 A1 a method for introducing a reagent into an exhaust gas duct of an internal combustion engine and a device for carrying out the method. At least one catalytic converter is arranged in an exhaust gas duct of the internal combustion engine, in front of which a pressurized reagent is sprayed into the exhaust gas. The reagent pressure setpoint (pReaSw) is defined depending on a parameter (N, mL, Md, mK, vabg, pabg, TabgR, TRea).

However, such a control of the hydraulic pressure in the metering system to a predetermined, usually constant, desired pressure value is usually subject to a number of dynamic interferences, e.g. B. bounce effects when the metering valve is actuated and/or pump-induced pressure oscillations. However, the dosing processes, which usually take place regularly, have a particularly strong influence on the pressure in the dosing system. So causes a reducing agent removal by opening the metering valve z. B. a drop in pressure in the system, which the controller tries to compensate by activating the pump accordingly (e.g. by varying the pump speed). In the case of the constant setpoint pressure values of the controller mostly used in the previous systems, this counteracting results in considerable fluctuations in the pump speed, which is sometimes audible as a "pulsing" of the pump. In addition to a possible acoustic impairment due to this noise behavior of the pump, these speed fluctuations also increase its wear.

Accordingly, it is therefore an object of the invention to provide a solution for the operation of a conveying/dosing system which is improved compared to the known methods and with which the disadvantages mentioned above can be avoided. In particular, it is an object of the invention to provide a solution by means of which a low-wear operation of a conveyor-metering system can be achieved.

These objects are solved by a method, a delivery metering system and a motor vehicle with the features of the independent claims. Advantageous embodiments and applications of the invention are the subject matter of the dependent claims and are explained in more detail in the following description with partial reference to the figures.

The solution described here takes the approach of using a dynamic, i. H. time-varying to use target pressure value that is changed depending on an actuation of a metering device. In particular, a targeted adjustment of the target pressure value to pressure changes in the system and the resulting smaller deviation between the actual and target pressure value should advantageously avoid major interventions in the controlled system. This procedure is particularly practicable since the malfunctions (dosing events) that primarily occur during the operation of a conveyor dosing system are usually limited in time or occur regularly, i. H. predictable, occur.

Accordingly, a method for operating a delivery metering system for a fluid is provided. The fluid is preferably a reducing agent for exhaust gas aftertreatment, such as ammonia and/or an ammonia-releasing substance, e.g. B. aqueous urea solution. In principle, however, the method according to the invention is suitable for any fluids, i. H. Gases, liquids, gas-liquid mixtures and/or liquids with a low solid content. For example, the process can B. can also be used in connection with the operation of a delivery metering system for oil, fuel or cooling water.

The delivery metering system, which can also be referred to as a delivery and metering system, includes a pump (e.g. an impeller pump) for delivering the fluid and a metering device (e.g. a metering valve) that is connected to the pump via a Pressure line is fluidly connected. A "pressure line" can be a line, e.g. B. a pipe and / or hose line, to be understood for the transport of the fluid, which is designed for an overpressure inside the line compared to an atmospheric pressure outside the line.

Furthermore, in the above-mentioned delivery-metering system, a fluid pressure p _fluid of the fluid in the pressure line _should be regulated to a target fluid pressure value p setpoint via the pump, preferably via a speed variation of the pump. For this purpose, the regulation of the fluid pressure p _fluid can - in a manner known per se - a detection of the controlled variable (fluid pressure p _fluid or fluid pressure value), for example via a sensor device (e.g. a pressure sensor) arranged on the pressure line, a determination of the deviation of the Include controlled variable from the reference variable (nominal fluid pressure value p _desired ) and setting a manipulated variable (e.g. pump speed) on the basis of the determined deviation between the controlled variable and the reference variable. Here, classic cal control procedures, e.g. B. P control, PI control or PID control can be used. Instead of a regulation to a time-constant setpoint fluid pressure value p setpoint , which is mostly used in such systems, the setpoint fluid pressure value p setpoint _is changed, preferably dynamically _, as a function of an actuation (e.g. an opening and/or closing) the dosing device. In other words, the setpoint fluid pressure value p setpoint _should be a dynamic setpoint value that is varied on the basis of an actuation of the dosing device. Opening and/or closing the dosing device can _thus “trigger” a change in the setpoint fluid pressure value p setpoint , with this change also being able to take place with a time delay. The change in the setpoint fluid pressure value P _setpoint preferably reduces the control deviation between p _fluid and p _setpoint , as a result of which a corresponding counter-control of the pump and thus fluctuations in the operating mode of the pump can be avoided in an advantageous manner.

According to a first aspect of the invention, the change can include lowering the target fluid pressure value p _setpoint after opening the metering device and/or raising the target fluid pressure value P _setpoint after closing the metering device. In this case, the lowering is intended to lower the setpoint fluid pressure value _p setpoint to a setpoint fluid pressure value p setpoint that is smaller by comparison, and the raising of the setpoint fluid pressure value p setpoint _{is intended} to increase the setpoint fluid _pressure value p setpoint _to a setpoint fluid pressure value p that is larger by comparison _shall denote. Furthermore, the lowering and/or raising should take place after the opening or closing, and this can take place immediately afterwards or with a time delay. The change thus preferably includes a temporary lowering of the target fluid pressure value p _set after the metering device has been opened, e.g. B. until the next closing of the metering device, and / or a temporary increase in the target fluid pressure value p _is after closing the metering device, z. B. for a predetermined time interval. Since the aforementioned actuations (opening/closing) of the dosing device usually induce the above-mentioned behavior of the fluid pressure p _fluid in normal operation of a delivery dosing system and a certain inertia is also often present in the controlled system, a corresponding lowering or raising of p _{The aim is to} reduce the system deviation in an advantageous manner and thus avoid a strong variation in the manipulated variable (e.g. pump speed). In comparison to the procedure that has been the norm up to now, in which an occurring error is primarily reduced by appropriate countermeasures by varying the manipulated variable, the reduction of the error here should also include a corresponding adjustment of the setpoint to the controlled variable.

According to a further aspect of the invention, the change can take place on the basis of a rate of change of the fluid pressure p _fluid in the pressure line, which is preferably caused by extraction. In this case, the “rate of change” variable, which can also be referred to as the rate of change in the fluid pressure p _fluid , is intended to denote a change in the fluid pressure p _fluid over time. In other words, the method or the step of changing p setpoint _can react to the speed at which the fluid pressure p _fluid is currently changing. This is advantageous since a current change in the fluid pressure p _fluid can be used to extrapolate its future progression—quasi anticipatingly. For example, a momentarily high rate of change in the fluid pressure P _fluid will most likely also lead to a significant change in P _fluid in the subsequent time segment, so that by taking this variable into account, a corresponding adjustment of p setpoint _can advantageously take place. For this purpose, the method can determine a rate of change of the fluid pressure p _fluid - z. B. in the case of a withdrawal-related drop in the fluid pressure p _fluid with the metering device open - and changing the target fluid pressure value P _should include on the basis of the rate of change. The latter can z. B. proportional to the rate of change of the fluid pressure p _fluid .

In order to be more robust against short-term fluctuations or noise in the fluid pressure p _fluid , the method according to the invention can include adapting a first fluid pressure gradient estimator G ₁ and changing the setpoint fluid pressure value p _set on the basis of this first fluid pressure gradient estimator G ₁ . According to the usual understanding, the term “adapt” can denote an adaptation of an existing starting value for the fluid pressure gradient estimator G ₁ to changing circumstances. In contrast to a pure "determination", a previous or previous value of the variable to be adapted is also included in the adaptation. For example, the first fluid pressure gradient estimator G ₁ can be adapted via the relationship G _1,new = G _1,old + η (G _1,old -G _tat ), where G _1,new is the newly adapted value of the previous value (G _{1 ,old} ) of G ₁ which is determined taking into account a unitless learning factor η and the deviation between G _1,old and a preferably currently determined actual fluid pressure gradient (G _tat ) of the fluid pressure p _fluid . The latter can z. B. on the basis of an opening of the metering device resulting pressure drop of p _fluid can be determined. In other words, the method can include adapting the first fluid pressure gradient estimator G ₁ to a, preferably extraction-related tes, drop in the fluid pressure p _fluid when the dosing device is open.

The aforementioned adaptation can also take place several times during a preferably predetermined opening period of the dosing device. In other words, when the dosing device is open, the first fluid pressure gradient estimator G ₁ should be adapted at least twice to a fluid pressure p _fluid that changes over time as a result. In this way, better adaptation to changing pressure conditions can be achieved in an advantageous manner. The multiple adaptation during the opening period z. B. at fixed time intervals (e.g. every 0.1 s). Since the opening duration of the dosing device (valve opening duration) to achieve a specific injection quantity of reducing agent is usually predetermined, ie previously specified, the time interval or the number of adaptation processes per opening duration can be specified in advance or adapted to the opening duration.

In addition or as an alternative, the adaptation can also include a standardization of the removal-related decrease with respect to a preferably predetermined dosing period T. The term "metering period" is intended to denote the time interval between two consecutive metering events or reducing agent injections, with the metering period being divided into a "first" time period t ₁ with the metering device open and a "second" period t ₂ with the metering device closed. By normalizing the drop caused by extraction and/or the actual fluid pressure gradient G _tat determined therefrom, for example using the normalization factor T/(Tt ₁ ), a standardized change in p _fluid over time can be provided in an advantageous manner, which in turn can then be described more precisely as follows executed - _can also be used to calculate an increase in the setpoint fluid pressure value p setpoint .

In addition or as an alternative, the first fluid pressure gradient estimator G ₁ can also be checked for plausibility. It can be checked whether G ₁ a predetermined plausibility criterion, z. B. in the form of a predetermined threshold value and / or a specific range of values, is sufficient to be able to preferably determine whether there is a meaningful or suitable estimated value for the fluid pressure gradient. Correspondingly, the plausibility check can also be referred to as a validation and/or a check. In addition or as an alternative, the plausibility criterion can also characterize the functionality of the delivery and metering system. In other words, a plausibility check of the first fluid pressure gradient estimator G ₁ can be carried out to check the functionality of the delivery and metering system. For example, on the basis of a fluid pressure gradient estimator G ₁ that is too low, it can be concluded that a metering device is clogged. If predetermined plausibility criteria are not met, errors in the delivery/dosing system and/or a basic suitability of G ₁ for adjusting the setpoint fluid pressure value p setpoint _can advantageously be quickly detected.

According to a further aspect of the invention, changing the target fluid pressure value P target _can include both a reduction in the target fluid pressure value p _target based on the first fluid pressure gradient estimator G ₁ and an increase in the target fluid pressure value p _target based on the first fluid pressure gradient estimator G ₁ . In other words, the fluid pressure gradient estimator G ₁ previously determined or adapted on the basis of a drop in the fluid pressure P _{fluid can} be used both to reduce the setpoint fluid pressure value p setpoint , preferably when the dosing device is open, and to increase the setpoint fluid pressure value p _setpoint , preferably when closed dosing device. The latter, at first glance not intuitive, using a gradient adapted when p _{fluid falls} to calculate an increase in the setpoint fluid pressure value p setpoint with the metering valve closed results from the fact that, as a rule, after the metering device has been closed, the original output fluid pressure is Further delivery of the pump can be achieved again, if necessary, however, compared to the pressure drop with a changed slope. Consequently, the first fluid pressure gradient estimator G _{1 can} in principle also be used to calculate an increase in the setpoint fluid pressure value p set , with a correction factor c ₁ sometimes also being included here to take account of different opening and closing times (t ₁ and t ₂ ). For example, this can be done using the factor c ₁ = -t ₂ /T. The advantage of this variant is that the target fluid pressure value P _soll can be calculated for the entire metering period T on the basis of just “one” adaptation during the drop in the fluid pressure p _fluid in a metering period T.

However, in order to avoid changing or adjusting the target fluid pressure value p _if the pressure rises, e.g. B. after closing the dosing device, the method according to the invention can also include an adaptation of a second fluid pressure gradient estimator G ₂ to a preferably pump-related increase in the fluid pressure p _fluid when the dosing device is closed. As mentioned above in connection with the adaptation of the first fluid pressure gradient estimator G ₁ , the adaptation of G ₂ z. B. via the relationship G _2,new = G _2,old + η · (G _2,old -G _tat ), where G _2,new represents the newly adapted value of the previous value (G _2,old ) of G ₂ , considering a units loose learning factor η and the deviation between G _2,old and a preferably currently determined actual fluid pressure gradient (G _tat ) of the fluid pressure p _fluid is determined. In contrast to G _{1 ,} however, G ₂ is to be determined on the basis of a pressure rise in the fluid pressure p _fluid that results after the metering device has been closed as a result of the delivery of the pump or as a result of a pressure rise caused by this. The adaptation preferably takes place several times during a preferably predetermined closing period of the dosing device. Furthermore, the method can in turn also include a plausibility check of the second fluid pressure gradient estimator G ₂ , preferably for checking the functionality of the delivery and metering system, with the methods mentioned above in connection with the plausibility check of G ₁ being able to be used again here.

Irrespective of whether the second fluid pressure gradient estimator G ₂ is checked for plausibility or not, the setpoint fluid pressure value p _soll can therefore be changed on the basis of the second fluid pressure gradient estimator G ₂ . Mirroring the case of the first fluid pressure gradient estimator G ₁ , the method can have the variant that changing the setpoint fluid pressure value P _setpoint both _increases the setpoint fluid pressure value p setpoint , preferably with the dosing device closed, on the basis of the second fluid pressure gradient estimator G ₂ and also a Lowering of the setpoint fluid pressure value p _setpoint , preferably with the dosing device open, on the basis of the second fluid pressure gradient estimator G ₂ . In other words, the fluid pressure gradient estimator G ₂ previously determined or adapted on the basis of an increase in the fluid pressure p _fluid can thus be used both for increasing and decreasing the setpoint fluid pressure value P _soll , with a correction factor c ₂ , e.g. B. in the form of c ₂ = -t ₁ / T, to take into account different opening and closing times.

However, if both G ₁ and G ₂ are present, it is particularly advantageous that - according to a further aspect of the invention - changing the setpoint fluid pressure value p _setpoint lowers the setpoint fluid pressure value p _setpoint on the basis of the first fluid pressure gradient estimator G ₁ and a Raising the target fluid pressure value p target based on the second fluid pressure gradient estimator G _{1 may} include. Advantageously, the fluid pressure gradient estimator G ₁ previously determined or adapted on the basis of a drop in p _{fluid for} calculating a drop in the setpoint fluid pressure value _p setpoint when the metering valve is open and the fluid pressure gradient estimator G ₂ to calculate an increase in the setpoint fluid pressure value p _{is to} be used when the dosing valve is closed. Overall, this advantageously enables the best possible adaptation of the setpoint fluid pressure value p _setpoint in the various operating phases.

According to a further aspect of the invention, the adaptation can take place with a regulated fluid pressure p _fluid and/or during operation of the pump. In other words, the adaptation can take place during normal operation of the conveyor-dosing system and not in a special calibration mode in which z. B. certain functions such as the pump, the control and / or the metering device are disabled. The adaptation can take place more or less continuously, and to avoid possible instabilities, a period of the immediate opening and/or closing process of the dosing device and/or particularly short or long opening durations of the dosing device can be excluded from the adaptation.

Additionally or alternatively, the adaptation can include a first adaptation during a first dosing process of the dosing device and at least a second adaptation during a second dosing process of the dosing device. In other words, the method can include an adaptation for different dosing processes, with the adaptation preferably taking place for each dosing process. In this case, the adaptation should be able to refer both to the adaptation of G ₁ and/or G ₂ . In addition to the multiple adaptation already described above during a dosing process, e.g. B. multiple adaptation during the opening time of a (first) dosing process, the method can thus also or alternatively multiple adaptation in several dosing processes, z. B. an adaptation for each opening duration of a first, second, third, etc. dosing process. In this way, a fluid pressure gradient estimator that is as up-to-date as possible can be provided over the entire operation in an advantageous manner, and it is thus also possible to react to changing operating conditions.

In addition or as an alternative, the adaptation, ie the adaptation of G ₁ and/or G ₂ , can be influenced via a changeable learning factor η. The strength of the adaptation or the degree of adaptation can be varied in an advantageous manner by means of this parameter, which is preferably unitless. Merely as an example, the learning factor η can also flow into the adaptation via the relationship G _1/2,new =G _1/2,old +η·(G _1/2,old −G _tat ), with G _1/2,new the represents the newly adapted value of the previous value (G _1/2,old ) of G _1/2 , which is determined taking into account the deviation between G _1/2,old and an actual fluid pressure gradient (G _tat ) of the fluid pressure p _fluid . η can be fixed or variable and z. B. assume values between 0 < η ≤ 1. In this way, the adaptation process can be influenced in an advantageous manner and, overall, a more robust operating behavior of the delivery and metering system can be achieved.

According to a further aspect of the invention, the setpoint fluid pressure value p _setpoint can fluctuate around a constant base setpoint fluid pressure value p ₀ . In other words, the resulting setpoint fluid pressure value p _setpoint can change from p ₀ to both higher and lower setpoint fluid pressure values, preferably symmetrically, ie to the same extent to higher and lower setpoint fluid pressure values, and/or alternately temporarily to higher and lower desired fluid pressure values. In this way, the most balanced possible adaptation of p _desired can be achieved in an advantageous manner.

In order to enable an efficient and rapid calculation of the setpoint fluid pressure value p setpoint in an advantageous manner, the setpoint fluid pressure value p setpoint _can be changed on the basis of a model that is preferably _linear over time, according to a further aspect of the invention. For this purpose, the method _can include a preferably continuous calculation of p set , with the time t _being included in no higher powers than 1 in the calculation of p set. In addition or as an alternative, the setpoint fluid pressure value p _setpoint can describe a function that is linear in sections over time. In other words, the setpoint fluid pressure value p _{setpoint can} be described in sections over time by functions of the form m·t+b, where m and b are real numbers.

According to a further aspect of the invention, the fluid pressure p _fluid of the fluid _can be regulated to a setpoint fluid pressure value p setpoint by varying the speed of the pump. In this context, a “speed variation” is to be understood as changing a size of the pump that indicates the number of pump revolutions in a specific time. In this context, the term "speed" can generally be understood both for rotary movements (e.g. in the case of a rotary vane pump) and other periodic processes (e.g. in the case of a diaphragm pump or reciprocating pump). In other words, the term rotational speed can also designate a frequency. In addition or as an alternative, the fluid pressure p _fluid of the fluid _can also be regulated to the setpoint fluid pressure value p setpoint by means of PI regulation. Alternatively, other control methods known in the prior art, including e.g. B. P control or PID control are used and / or other manipulated variables are used for control, including z. B. a ladder vane position and / or a throttle setting.

Furthermore, a delivery-dosing system for a fluid is provided. The fluid is preferably a reducing agent for exhaust gas aftertreatment, such as ammonia and/or an ammonia-releasing substance, e.g. B. aqueous urea solution. In principle, however, the corresponding conveying and dosing system is suitable for any fluids, ie gases, liquids, gas-liquid mixtures and/or liquids with a low solids content. The delivery metering system includes a pump (e.g. an impeller pump) for delivering the fluid, a metering device that is fluidically connected to the pump via a pressure line, and a control device that is designed to control the pump, preferably to regulate a fluid pressure p _fluid of the fluid in the pressure line by controlling a speed of the pump according to a method as described in this document. For this purpose, the delivery-dosing system can also comprise a sensor device (e.g. a pressure sensor) preferably arranged on the pressure line, which is designed to detect the fluid pressure p _fluid of the fluid in the pressure line (fluid pressure value). Furthermore, the control device can be designed to determine a deviation of the detected fluid pressure value from the setpoint fluid pressure value p _set and to define a manipulated variable (e.g. pump speed) on the basis of the determined deviation between the controlled variable and the reference variable. In this way, a delivery-metering system is advantageously provided overall, by means of which fluctuations in the operation of the pump, which in particular increase wear, can be avoided.

Furthermore, a motor vehicle is provided which has a delivery metering system as described in this document. The motor vehicle is preferably a commercial vehicle. In other words, the motor vehicle can be a vehicle that, due to its design and equipment, can be designed for transporting people, for transporting goods, or for towing trailer vehicles. For example, the motor vehicle can be a truck, a bus and/or a trailer truck.

• 1: Eine schematische Darstellung eines Förder-Dosiersystems für ein Fluid gemäß einer Ausführungsform der Erfindung;
• 2: Eine exemplarische Messung des Verlaufs des Fluiddrucks p_fluid in einem Förder-Dosiersystem als Funktion der Zeit bei verschiedenen Regelungen; und
• 3: Eine schematische Darstellung eines Kraftfahrzeugs mit einem Förder-Dosiersystem gemäß einer Ausführungsform der Erfindung.

The aspects and features of the invention described above can be combined with one another as desired. Further details and advantages of the invention are described below with reference to the accompanying drawings. Show it:

• 1 FIG. 1: A schematic representation of a delivery-metering system for a fluid according to an embodiment of the invention;
• 2 : An exemplary measurement of the progression of the fluid pressure p _fluid in a conveyor Dosing system as a function of time with different regulations; and
• 3 : A schematic representation of a motor vehicle with a delivery metering system according to an embodiment of the invention.

Identical or functionally equivalent elements are denoted by the same reference symbols in all figures and some of them are not described separately.

1 shows a schematic representation of a delivery-metering system 10 for a fluid 1, ie a system for delivering and metering a fluid 1, according to an embodiment of the invention. In the present case, the fluid 1 is a reducing agent for exhaust gas aftertreatment, such as, for. B. aqueous urea solution. The delivery/dosing system 10 has a pump 2 for delivering the reducing agent stored in a fluid container 7 (eg a tank). Furthermore, the delivery-metering system 10 comprises a metering device 3 in the form of a metering valve which is fluidly connected to the pump 2 via a pressure line 4 and via which the reducing agent can be introduced or sprayed into an exhaust tract 9 in the present case. The two arrows should 1 mark the flow direction of the exhaust gas in the exhaust tract 9. To set the corresponding dosing parameters, such as e.g. B. amount, spray jet shape, etc., is - in addition to the design of the metering device 3 itself - especially the applied to the metering device 3 fluid pressure p _fluid of the reducing agent is crucial. Correspondingly, in the present delivery and metering system 10, regulation of p _fluid to a setpoint fluid pressure value p setpoint _is provided, with this z. B. a corresponding variation of a current speed of the pump 2 can be done.

For this control of p _fluid, the delivery-dosing system 10 can also include a control device 5, which a current fluid pressure p _fluid or a corresponding fluid pressure signal (cf. 2 ) is provided, wherein the fluid pressure signal can be both an actual fluid pressure and/or a variable from which the fluid pressure of the reducing agent can be derived. In order to provide this variable, the delivery/dosing system 10 can comprise a sensor device 6 which is arranged on the output side of the pump 2 and which is designed to detect and provide the fluid pressure p _fluid of the reducing agent in the pressure line 4 . In this case, the control device 5 can also be designed to determine a deviation between p _fluid and the setpoint fluid pressure value p _setpoint and to define a manipulated variable (e.g. pump speed) on the basis of the determined deviation between the controlled variable and the reference variable, in order to thereby regulate the reducing agent pressure . In this context, control methods known per se in the prior art, including e.g. B. PI control can be used. In addition, the reducing agent pressure can also be regulated via an optional—shown here—return line to the tank 6 with a throttle 8 arranged there.

In connection with the regulation of the fluid pressure p _fluid of the reducing agent, it is provided that instead of a time-constant setpoint fluid pressure value p setpoint , which is mostly used in systems of this type, a dynamic setpoint fluid pressure p _{setpoint is} introduced, which depends on an actuation (e.g. B. an opening and / or closing) of the metering device 3 is to be changed. For this purpose, p _set is calculated, preferably continuously, with actuations of the dosing device 3 _triggering a corresponding adjustment of the set fluid pressure value p set , to which reference is made below to FIG 2 should be dealt with in more detail.

2 shows an exemplary measurement of the course of the fluid pressure p _fluid in a delivery-metering system 10 as a function of time t with different regulations. In the first approximately 16 seconds, PI control takes place using a dynamic setpoint fluid pressure value p _setpoint , while the dynamic variation of p _setpoint is then deactivated and a constant setpoint fluid pressure value p _setpoint is set. Before the specific changing or calculation of the setpoint fluid pressure value p setpoint in the dynamic case is _to be discussed in detail, fundamental differences between the two types of regulation should first be highlighted.

Looking at the progression of the fluid pressure P _fluid with the pump 2 running, after an initially essentially constant progression around p ₀ (up to approx. 1 s), it exhibits a plurality of jagged modulations which are associated with metering processes D that occur regularly here. The dosing processes D are determined by a dosing period T, which is divided into a first time period t ₁ with the dosing device 3 open and a second time period t ₂ with the dosing device 3 closed, with in 2 only the opening of the dosing device 3 is highlighted by the dashes “D”. If the dosing device 3 is opened, i.e. fluid 1 is removed, this induces a removal-related drop in the fluid pressure P _fluid , whereas the subsequent closing of the dosing device 3 due to the pump 2 “conveying” causes an increase in the fluid pressure p _fluid .

In the case of in the right image area of 2 (from about 16 s) shown PI control over a constant setpoint fluid pressure value p _set , the PI Control to compensate for the aforementioned fluctuations or modulations by a corresponding control of the pump speed N _p , which z. B. on the rashes of the in 2 shown curve of the pump speed N _p can be seen. Despite this considerable counter-steering of the control, it is not able - at least in the case of the settings used here - to completely compensate for the fluctuations induced by the actuation of the dosing device 3, so that for a "more constant" course of p _fluid closer to the setpoint p _should actually be even stronger control interventions or .a faster regulation would be necessary. However, since this would be associated with correspondingly even stronger fluctuations (“pulsing”) in the pump speed _Np , which are sometimes perceived as acoustically disturbing or can shorten the service life of pump 2, this is shown in the left image area of 2 illustrated, control procedures take a different approach.

As can be seen from the range between 0 and 16 seconds, the setpoint fluid pressure value p _setpoint does not assume a value that is constant over time, but varies dynamically depending on an actuation of the metering device 3. Except for a short adaptation phase when the metering device 3 is opened for the first time (about 1 s), after the metering device 3 is opened, the target fluid pressure value p setpoint _is lowered and after the metering device 3 is closed, the setpoint fluid pressure value p setpoint _is increased. In this way, the control deviation between p _fluid and p _setpoint can be reduced in an advantageous manner and a large variation in the pump speed N _p can thus be avoided. This is also reflected in the significantly lower fluctuations in N _p compared to the constant setpoint fluid pressure value p _setpoint . Correspondingly, in the case of the dynamic setpoint fluid pressure value p setpoint , the pump 2 _runs more smoothly, which in turn counteracts wear on the pump 2 .

The continuous calculation or adjustment of p set as described above _{can be carried out} as a function of actuation of the dosing device 3 e.g. B. on the basis of temporally linear changes in p _should take place. Thus, the target fluid pressure value p _set when the dosing device 3 is open, e.g. B. via the relationship p (t)=p _start +G ₁ · t _should be calculated, where p _start should designate a fluid pressure start value estimator, G ₁ a (negative) “first” fluid pressure gradient estimator and t should designate the time from the opening of the dosing device 3. The fluid pressure gradient estimator is intended to designate a fixed value for the fluid pressure gradient, with the fluid pressure gradient intended to describe the change in fluid pressure Δp _fluid in a preferably predetermined time interval Δt and/or the derivation of fluid pressure P _fluid after time t.

In principle, p _start and G ₁ can be fixed, particularly if the system behavior is known or the fluctuations to be expected are foreseeable. In this way, the first fluid pressure gradient estimator G ₁ can be adapted to a drop in the fluid pressure p _fluid , preferably due to removal, when the dosing device 3 is open. For this purpose, G ₁ z. B. can be determined via the relationship G _1,new = G _1,old + η (G _1,old -G _tat ), where G _1,new is the newly adapted value of the previous value (G _1,old ) of G ₁ which is determined taking into account a unitless learning factor η and the deviation between G _1,old and an actual fluid pressure gradient (G _tat ) of the fluid pressure p _fluid that is preferably currently determined.

In contrast to a pure determination of the fluid pressure gradient estimator G ₁ , its previous value also flows into its adaptation, it being possible for a basic estimator for G ₁ to be specified at the beginning of the adaptation process. In order to advantageously obtain the best possible adaptation to changing pressure conditions during the opening time t ₁ of the metering device 3, the adaptation can also take place several times (e.g. every 0.1 s) when the metering device 3 is open. Furthermore, in a preferred embodiment, the actual fluid pressure gradient G _tat , ie the preferably current change over time in p _fluid , can be a normalized actual fluid pressure gradient G _tat based on the known time period t ₁ with the metering device 3 open and the known dosing period T is normalized by the factor T/(Tt ₁ ). Accordingly, in this case the first fluid pressure gradient estimator G ₁ can also be referred to as the normalized first fluid pressure gradient estimator G ₁ , in which case a corresponding denormalization _should then be carried out to calculate the setpoint fluid pressure value p setpoint with the dosing device 3 open, ie P _setpoint (t)=p _start +G ₁ ·t ·(Tt ₁ )/T.

After the change over time in p des while the metering device 3 _is open has been described via the above-mentioned relationship, the metering device 3 is closed after the time period t ₁ , as a result of which the fluid pressure p _fluid in the pressure line 4 rises again. As mentioned above, in this period of time t ₂ until the metering device 3 opens again, the target fluid pressure value p _setpoint - starting from the target final fluid pressure value p _end when the metering device 3 is closed - should be increased again. This can z. B. the relationship p _{is to} be used to calculate (t)=p _end +G ₂ ·(Tt), where G ₂ has one (Positive) “second” fluid pressure gradient estimator, which is adapted analogously to G ₁ , but to a preferably pump-related increase in the fluid pressure p _fluid when the dosing device 3 is closed. As an alternative to this, the setpoint fluid pressure value p _soll can also be increased on the basis of the first fluid pressure gradient estimator G ₁ .

In other words, the fluid pressure gradient estimator G ₁ previously determined or adapted on the basis of a drop in the fluid pressure p fluid _can also be used to increase the setpoint fluid pressure value p _setpoint . This is possible because the original outlet fluid pressure can usually be reached again by the further delivery of the pump 2 after the metering device 3 has been closed. In the case of different opening and closing times (t ₁ or t ₂ ), however, it is advantageous to take this into account via a corresponding correction factor c ₁ =-(Tt ₁ )/T. After t ₂ has elapsed, the dosing device 3 opens again and another calculation cycle for p _setpoint for the next dosing period T starts. For an adjustment of p set that _is as balanced as possible, it can fluctuate, preferably symmetrically, around a constant base set fluid pressure value p ₀ , for which purpose p _start based on the pressure drop to be expected in interval t ₁ as p _start =p ₀ +G ₁ . t _1/2 can be fixed.

Overall, the control deviation between p _fluid and p _setpoint can be reduced in an advantageous manner, as a result of which strong counter-control of the pump 2 and thus fluctuations in the operating mode of the pump 2 can be avoided.

3 shows a schematic representation of a motor vehicle 20 with a delivery metering system 10 according to an embodiment of the invention. In the present case, the motor vehicle 20 is an articulated vehicle, ie a combination of a tractor unit and a semi-trailer. The delivery and metering system 10 arranged in the motor vehicle 20 comprises a pump 2 for delivering a fluid 1 and a metering device 3 which is fluidically connected to the pump 2 via a pressure line 4 . Furthermore, the delivery-dosing system 10 comprises a control device 5, which is designed to control the pump 2, preferably by controlling the speed of the pump 2, a fluid pressure p _fluid of the fluid 1 in the pressure line 4 _to a target fluid pressure value p setpoint to be regulated as described above. In other words, the setpoint fluid pressure value p setpoint _should be changed as a function of an actuation of the metering device 3 .

Although the invention has been described with reference to specific embodiments, it will be apparent to those skilled in the art that various changes may be made and equivalents substituted without departing from the scope of the invention. Accordingly, the invention should not be limited to the disclosed embodiments, but should include all embodiments falling within the scope of the appended claims. In particular, the invention also claims protection for the subject matter and the features of the subclaims independently of the claims referred to.

Reference List

1

Fluid

2

pump

3

dosing device

4

pressure line

5

control device

6

sensor device

7

fluid container

8th

throttle

9

exhaust tract

10

Conveying dosing system for a fluid

20

motor vehicle

D

dosing process

np

pump speed

p0

Base target fluid pressure value

pfluid

fluid pressure

psoll

target fluid pressure value

T

dosing period

t1

First period

t2

second period

Claims (13)

Method for operating a delivery metering system (10) for a fluid (1), preferably a delivery metering system (10) for a reducing agent for exhaust gas aftertreatment, the delivery metering system (10) having a pump (2) for delivering the fluid (1 ) and a dosing device (3) which is fluidically connected to the pump (2) via a pressure line (4); wherein via the pump (2), preferably via a speed variation of the pump (2), a fluid pressure p _fluid of the fluid (1) in the pressure line (4) _is regulated to a setpoint fluid pressure value p setpoint; and wherein the setpoint fluid pressure value p setpoint _is changed as a function of an actuation of the dosing device (3); characterized by i. adapting a first fluid pressure gradient estimator G ₁ to a drop in fluid pressure p _fluid , preferably due to extraction, when the dosing device (3) is open and changing the setpoint fluid pressure value p _set on the basis of the first fluid pressure gradient estimator G ₁ ; and/or ii. an adaptation of a second fluid pressure gradient estimator G ₂ to a preferably pump-related increase in the fluid pressure p _fluid when the dosing device (3) is closed and changing the setpoint fluid pressure value p _setpoint on the basis of the second fluid pressure gradient estimator G ₂ , procedure after claim 1 , characterized in that the change _includes lowering the target fluid pressure value p setpoint after the metering device (3) has opened and/or raising the target fluid pressure value p setpoint _after the metering device (3) has closed. Method according to one of the preceding claims, characterized in that the change takes place on the basis of a rate of change of the fluid pressure p _fluid in the pressure line (4), which is preferably caused by extraction. Method according to one of the preceding claims, if related back to variant i., characterized in that a) the adaptation takes place several times during a predetermined opening time of the dosing device (3); and/or b) that the adaptation comprises a normalization of the removal-related decrease with respect to a predetermined dosing period T; and/or c) that a plausibility check of the first fluid pressure gradient estimator G ₁ takes place, preferably for checking the functionality of the delivery-dosing system (10). Method according to one of the preceding claims, if related back to variant i., characterized in that changing the target fluid pressure value p _target both lowering the target fluid pressure value p _target on the basis of the first fluid pressure gradient estimator G ₁ and raising the target fluid pressure value p _{is to} include based on the first fluid pressure gradient estimator G ₁ . Method according to one of the preceding claims, if related back to variant ii., characterized by a plausibility check of the second fluid pressure gradient estimator G ₂ , preferably for checking the functionality of the delivery-dosing system (10). Method according to one of the preceding claims, if related back to variant i. and ii. if referring back to claim 4 or 5 , characterized in that changing the target fluid pressure value p _target lowering the target fluid pressure value p _target based on the first fluid pressure gradient estimator G ₁ and raising the target fluid pressure value p _target based on the second fluid pressure gradient estimator G ₁ . Method according to one of the preceding claims, characterized in that a) the adaptation takes place with a regulated fluid pressure p _fluid and/or during operation of the pump (2); and/or b) that the adaptation comprises a first adaptation during a first dosing operation of the dosing device (3) and at least a second adaptation during a second dosing operation of the dosing device (3); and/or c) that the adaptation can be influenced via a changeable learning factor η. Method according to one of the preceding claims, characterized in that the _desired fluid pressure value psoll fluctuates around a constant basic desired fluid pressure value po, preferably symmetrically and/or alternately. Method according to one of the preceding claims, characterized in that the change in the setpoint fluid pressure value p _setpoint takes place on the basis of a time-linear model; and/or the setpoint fluid pressure value p _should describe a function that is linear in sections over time. Method according to one of the preceding claims, characterized in that a) the fluid pressure p _fluid of the fluid (1) is regulated to a setpoint fluid pressure value p _setpoint by varying the speed of the pump (2); and/or b) that the fluid pressure p _fluid of the fluid (1) _is regulated to the setpoint fluid pressure value p setpoint by means of PI regulation. Delivery metering system (10) for a fluid (1), preferably a delivery metering system (10) for a reducing agent for exhaust gas aftertreatment, comprising a) a pump (2) for delivering the fluid (1); b) a dosing device (3) which is fluidically connected to the pump (2) via a pressure line (4); and c) a control device (5) which is designed to control the pump (2), preferably by controlling a speed of the pump (2), according to a fluid pressure p _fluid of the fluid (1) in the pressure line (4). a procedure of Claims 1 until 11 to settle. Motor vehicle (20), preferably a commercial vehicle, having a delivery metering system (10). claim 12 .

Images