Method for exhaust gas aftertreatment of an internal combustion engine with at least one SCR catalyst

Abstract

Method for exhaust gas aftertreatment of an internal combustion engine (10) with at least one SCR catalyst (22) to which exhaust gas of an internal combustion engine (10) is supplied via an exhaust gas path (2), wherein the at least one SCR catalyst (22) is arranged virtually in n-bricks (n b ) is divided into where desired NH3 target levels (NH3 Soll,i ) for the at least one SCR catalyst (22) as a function of a desired target NOx conversion efficiency (η NOx ) is determined, where for the determination of the desired NH3 target levels (NH3 Soll,i ) a stationary state is assumed, where the desired NH3 target levels (NH3 Soll,i ) for the at least one SCR catalyst (22) by means of an inverse SCR model (F -1 ) using a current NOx concentration (NOx Us ) upstream of the SCR catalyst (22), an SCR catalyst temperature (T SCR ), especially SCR catalyst temperatures of the bricks (T SCR,i ), an exhaust gas mass flow (dmEG), an oxygen concentration (O2), an exhaust gas pressure (p) upstream of the SCR catalyst (22) and a desired target NOx conversion efficiency (η NOx ) is determined, where depending on the determined NH3 target level (NH3 Soll ) a control of the SCR catalyst (22) is carried out.

Description

Prior ArtSelective catalytic reduction (SCR) by means of ammonia (NH3) or ammonia-releasing reagents constitutes a promising method for the reduction of nitrogen oxides in oxygen-rich exhaust gases. The operating window of an SCR catalytic converter 22 or its efficiency is defined by the physical variables of the temperature and the space velocity. The degree of coverage of the catalyst with adsorbed NH3is decisive for the efficiency. As the temperature increases, the ability of the SCR catalyst 22 to store ammonia decreases. In order to maintain high efficiencies, the ammonia quantity to be converted must be dosed instantaneously to the measured nitrogen oxide quantity upstream of the SCR catalyst. Ammonia which does not react with NOx, desorbs from the catalyst as ammonia slip or oxidizes it due to high temperature on the catalyst must likewise be added in a postdosed fashion. In order to achieve the highest possible nitrogen oxide conversions, it is essential to operate the SCR system at a high ammonia fill level. If the temperature of the filled SCR catalytic converter 22 rises as a result of a load jump of the internal combustion engine, its ammonia storage capacity of the SCR catalytic converter 22 falls, which can lead to ammonia slip. SCR catalytic converters which are installed close to the engine in order to convert nitrogen oxides early after engine start are subject in particular to dynamic temperature gradients. Depending on the NH3 fill level loading or the gradients thereof, this fact can lead to increased NH3 desorption. A second SCR catalyst downstream of the first SCR catalyst 22 may therefore be provided in the exhaust line to adsorb and subsequently react ammonia from ammonia slip of the first catalyst. The guidelines for onboard diagnostics (OBD) require that both SCR catalysts must be monitored.Disclosure of the InventionIn a first aspect, the invention relates to a method for exhaust gas aftertreatment of an internal combustion engine having at least one SCR catalyst to which exhaust gas of an internal combustion engine is supplied via an exhaust gas path, wherein the at least one SCR catalyst is virtually divided into i-brick in the direction of flow of the exhaust gas, wherein desired NH3 soll levels for the at least one SCR catalyst are determined as a function of a desired desired nominal NOxconversion efficiency, wherein a steady state is assumed for the determination of the desired NH3 soll levels, wherein the desired NH3 soll levels for the at least one SCR catalyst are determined by means of an inverse SCR model by means of a current NOxconcentration upstream of the SCR catalyst, an SCR catalyst temperature, In particular, the SCR catalyst temperatures of the brick, an exhaust gas mass flow, an oxygen concentration, an exhaust gas pressure upstream of the SCR catalyst and a desired setpoint NOxconversion efficiency are determined, wherein a regulation of the SCR catalyst is carried out as a function of the determined NH3 soll fill level.The advantage of this method is that, in the event of a motor load jump (operating point change), which leads to an increase in the temperature in the SCR catalyst or to dynamic temperature gradients, a high NOx / NH3conversion can nevertheless be ensured. In contrast to a fill level-based control system which compares only the actual NH3with the NH3target fill level, the method presented here optionally supplies deviating target variables which are calculated on an efficiency basis.Therefore, the efficiency-based method presented results in a higher total fill level which does not lead to a dosing pause. By recursive calculation of the efficiency-based target fill level, current motor operating points can also be included in the calculation.Thus, the method can be used to perform improved emission cleaning.In a further embodiment, for the assumed steady state, a balancing of the input and output concentrations of the at least one SCR catalyst on the basis of the NH3 füllstands is described as follows: with η Nox,i the NOxconversion efficiency of the ith brick, and ξ a stoichiometric factor which corresponds to the NH3 äquivalenten from the NOxconversion and is in particular in the range of [1:1.3]. The condition of stationarity has the advantage that it provides a clear solution to the fill level distribution. Without this condition, an infinite number of solutions would be possible, which may also be partially non-physical.In a particular embodiment, an NH3 dosier amount is determined as a function of a difference between the desired NH3 soll fill level and a current NH3 füllstand level and is metered by means of the SCR injection system. By this regulation, the SCR system is quickly guided to its desired operating point. Thus, harmful emissions can be reduced.In an alternative embodiment, the desired NOxconversion efficiency is determined or permanently predefined from a characteristic diagram as a function of the SCR catalyst temperature and / or the exhaust gas mass flow.In a particular embodiment, the inverse SCR model is determined by means of a reaction kinetic or a spatially resolved model. The use of an inverse model has the particular advantage that all physical quantities used in the model are taken into account as far as the inverse model is concerned. Consequently, this approach enables an operating and cat-state-dependent NH3 soll fill level specification, which ensures a higher NOxconversion in certain situations.In an advantageous embodiment, the desired NH3 soll fill level is composed of the sum of the desired NH3 soll fill levels of the n virtual brick.By dividing into the virtual brick, it is thus possible to resort to the internal states of the SCR catalytic converter or to those of the SCR model as setpoint variables. Thus, more precise information is available for the SCR catalyst, so that improved emission cleaning can be performed.In a further embodiment, the regulation is carried out on any virtual brick, in particular on the first brick.In a particular configuration, the total NOxconversion efficiency of the SCR catalyst is determined as a function of the NOxconversion efficiencies of the individual virtual brick according to the following formula:In an advantageous embodiment, two SCR catalytic converters are arranged one after the other in the flow direction of the exhaust gas, wherein the desired setpoint NOxconversion efficiency is determined as a function of the state of the second catalytic converter, wherein an actual NOxconversion efficiency and / or actual NH3 einzel levels and / or actual NH3 gesamt level are used for the calculation. This has the particular advantage that the states from the entire exhaust system are efficiently processed into the desired NOxconversion efficiency and are indirectly converted into the inverse model via this variable. Thus, the inverse model does not need to be continuously adjusted to a new target NOx conversion efficiency. This represents an intuitive approach.In further aspects, the invention relates to a device, in particular a control unit and a computer program, which are set up, in particular programmed, for executing one of the methods. In yet another aspect, the invention relates to a machine-readable storage medium on which the computer program is stored.Brief Description of the DrawingsThe invention is explained in more detail below with reference to an exemplary embodiment shown in the figures. The following are shown: FIG. 1 shows a schematic illustration of an internal combustion engine with an exhaust gas aftertreatment system according to the invention, and FIG. 2 shows a flow chart for graphically illustrating the sequence of an exemplary embodiment of the method.DETAILED DESCRIPTION OF THE DRAWINGSAn internal combustion engine 10 has in its exhaust line 11 an SCR exhaust gas aftertreatment system 25 with at least one SCR catalytic converter 22, which is illustrated in FIG. 1. This has a reducing agent metering unit 21, with which a urea water solution (AdBlue) can be injected into the exhaust tract 11. Ammonia is liberated therefrom at the high temperatures of the exhaust gas. Downstream of the reducing agent metering unit 21, a first SCR catalytic converter 22 is arranged. Further SCR catalysts may be disposed downstream of the first SCR catalyst 22. A first NOx sensor f and a temperature sensor 12 are arranged upstream of the reducing agent metering unit 21 and downstream of an internal combustion engine 10 in the exhaust line 11. The first NOx sensor 31 measures a first NOx concentration sensor value NOx 1, preferably as a NOx concentration or as a NOx mass flow. A second NOx sensor 32 is arranged downstream of the first SCR catalytic converter 22 and measures a second NOx concentration sensor value NOx 2, preferably as a NOx concentration or as a NOx mass flow. This also applies in particular to the NH3 concentration or an NH3 mass flow.In an optional embodiment, an NH3sensor 33 can furthermore be installed downstream of the SCR catalytic converter 22. The NH3sensor can determine an NH3mass flow.All NOx sensors 31, 32 transmit their signals to an electronic control unit 100. Since the NOx sensors 31, 32 react to ammonia in a cross-sensitive manner in addition to nitrogen oxides, their signals are sum signals of nitrogen oxides and ammonia. The first NOx sensor 31 is arranged upstream of the reducing agent dosing unit 21 in such a way that it reliably measures only the amount of nitrogen oxide in the exhaust gas. The reducing agent metering unit 21 also forwards the ammonia quantity metered into the exhaust system 11 to the control unit 100.Furthermore, an SCR model F for the SCR catalytic converter 22 is stored on the control unit 100, said model dividing the SCR catalytic converter 22 into n virtual brick, with n=1... n b, i ∈ N. In this case, the SCR catalytic converter 22 is modeled by means of a multi-slice model, wherein the current fill level quantity m NH3,i is determined for each virtual brick. Preferably, the fill level amounts correspond to NH3 fill level amounts. Furthermore, an exhaust gas mass flow dm EG, an exhaust gas pressure p, an exhaust gas temperature T, an oxygen concentration O 2 downstream of the internal combustion engine 10 and upstream of the SCR catalytic converter 22, and a metered quantity information item, can be determined. In particular, a brick temperature T exh,i is determined for each brick n of the SCR catalytic converter 22 as a function of the determined exhaust gas temperature T by means of a temperature model stored on the control unit 100 and is used as an input variable for the SCR model F.The invention relates to a method for determining a setpoint NH3 füllstand m NH3Soll for operating the SCR catalytic converter 22 depending on a desired nitrogen oxide (NOx) conversion rate η NOx. The actual NH3 füllstand level m NH3Ist is usually approximated by a model and cannot be measured directly. Depending on the application and complexity, the physical models can take account of a fill level distribution in the longitudinal direction of the catalytic converter.Under certain boundary conditions, this type of determination of the working point can avoid the interruption of the adblue metering, which can arise when the current NH3 füllstand level falls down to a lower NH3 soll level, and thus ensure higher nitrogen oxide conversions.The operating point of an SCR catalytic converter is determined decisively by the adsorbed ammonia quantity (NH3 füllstand). The ability of the SCR catalyst 22 to store ammonia is critically affected by the temperature of the catalyst T.The invention relates to a functionality which assigns a desired NOxconversion efficiency η NOx to an NH3 füllstand of the catalytic converter under the assumption of a steady state in a steady state, and thus determines the NH3 ziel fill level m NH3 of the SCR catalytic converter 22 for this NOxconversion efficiency η NOx wherein, when considering a multibrick model, n b defines the number of brick.The target NH3 distribution in the longitudinal direction of the SCR catalyst 22 is described by the target NH3 individual levels m NH3,i. Here, F -1 corresponds to an inversion of a physical SCR model F.An illustrative overview of the inputs and outputs of the SCR model F and of the inverted SCR model functionality is given in FIG. 2.Formally, the SCR model F can be defined as a function of physical variables such as, for example, the exhaust gas temperature T, the exhaust gas mass flow dm EG, the NH3 dosier quantity (us: upstream), the NOxconcentration, the oxygen concentration x O2, the exhaust gas pressure p and the NH3 füll levels.Reaction kinetic models can be used, for example, to calculate the SCR model. For a particular NH3 füllstand level of the SCR catalyst 22 and an AdBlueDosing (AdBlueDosing), the concentration of NOxand NH3downstream of the SCR catalyst 22 (ds: downstream) can be calculated or estimated with (k corresponds to NOxor NH3), r NOx function for calculating the NOxconcentrations, r NH3 function for calculating the NH3concentrations, m NH3,i ith NH3mass, T SCR,i temperature of the ith brick, NH3concentration of the previous brick, NOxconcentration of the previous brick, x is O2 the oxygen concentration and p is the exhaust pressure.Consequently, the NOx conversion efficiency can be determined depending on the above quantities as follows. The total conversion of the SCR catalyst 22 is given by.The condition of stationarity can be described by balancing the input and output concentrations (NOxand NH3) of the SCR catalyst 22 with reference to the NH3 füllstands: where ξ is the stoichiometric factor corresponding to the NH3 äquivalenten from the NOxconversion and is in the range of [1:1.3]. The factor φ can stand for further NH3 sources, such as, for example, an NH3 oxidation. Further,In the case of n b= 1 considering Equation (8), the right side of Equation (7) becomes zero. The inversion of the SCR guide model then relates exclusively to equation (3). Depending on the complexity of the functional relationship with the function of the NOx concentration r NOx an analytical solution of the inversion can be specified. As an example, for r NOx and i=1 starting from a linear NOxconversion reaction, the functional relationship is given as follows: wherein describes the maximum possible NH3 füllstand. The residence time τ can be determined from physical variables T, p, dm EG and the catalyst-specific volume. A reaction rate of the NOx conversion is given by k NOx and can be calculated from the reaction kinetic approach of Arrhenius law, which depends on the temperature.If, finally, equation (5) is readjusted and combined with equation (9), the inverted function F -1 for the 1-brick case can be specified:This equation can be adapted for the corresponding desired variables and the desired desired NH3 füllstand NH3 füllstand results from equation (1). In the case of more complex models and / or for n b >1, a recursive method is suitable for solving equation (1).The iterative solution would correspond to a minimization problem according to equation (7) and can be implemented using numerical optimization methods, such as, for example. Newton's method, pseudo-Newton, etc.The aim in numerical methods is not to determine the absolute target value, but rather to preset or fall below a suitable tolerance threshold ε tol in order to save computing time. That is, according to equation (7), the condition must be fulfilled.For example, it is sufficient within the scope of this invention to define a steady state with a ∂ t m NH3,i ≤ 1 mg / s.It should be mentioned that from a physical point of view, the filling levels in the stationary state in a multi-pane model are decreasing monotonically in the longitudinal direction. The fill level gradient that results is determined decisively by the NH3 ox (if taken into account). The total fill level is calculated as given in equation (2).In the numerical minimization, the left-hand side of equation (11) is recursively calculated starting from a starting value in order to obtain further estimated values for, where l corresponds to the iterative running number.When the condition of equation (11) is satisfied, the calculation ends after the l-th iteration.Depending on the method, the way in which the currently estimated setpoint NH3 füllstands level is incremented differs.For example, in Newton-based methods, recourse is made to the derivative matrix (Jacobi matrix) of the function to be minimized, which results for equation (11). Furthermore, it is necessary in an iterative process to start with a suitable starting value. For this purpose, an approximated analytical solution, as given in equation (10), can be used. It is to be considered to what extent the model to be inverted can be approximated into a 1-brick solution in order to estimate an initial value.The solution of the iterative approach yields the same result for the NOx conversion efficiency η Nox. by inserting it into the SCR model F.In order to ensure a minimum total setpoint fill level, it is necessary to make a maximum selection between the fill level calculated from the method and a minimum specification.Furthermore, it may be necessary for the NOx concentration downstream of the SCR catalytic converter 22 and the exhaust gas mass flow dm EG likewise to have to exceed a minimum value.It can be used especially for the latter variable for small values (for example. No load) results in very low fill levels.Furthermore, the method could be combined with the secondary condition of limiting the expected ammonia slip at high conversion rates.FIG. 2 shows, by way of example, the sequence of the method for an exhaust gas aftertreatment system having only one SCR catalytic converter 22, wherein the SCR catalytic converter 22 is virtually divided into n-brick.In a step 200, an exhaust gas mass flow dm EG, an exhaust gas pressure p, an SCR catalyst temperature T SCR, an oxygen concentration O 2, a NOx concentration NOx Us downstream of the internal combustion engine 10 and upstream of the SCR catalyst 22, and a desired setpoint NOx conversion efficiency η NOx are ascertained continuously by the control unit 100.In particular, a brick temperature T SCR,i is determined for each brick of the SCR catalytic converter 22 as a function of the determined exhaust gas temperature T exh by means of a temperature model stored on the control unit 100.The desired setpoint NOxconversion efficiency η NOx can be determined or permanently predefined from a characteristic map as a function of the SCR catalyst temperature T SCR and / or the exhaust gas mass flow dm EG. The characteristic map is determined in an application phase and stored in the control unit 100.In an alternative embodiment, two SCR catalytic converters (cannings) are positioned one behind the other. In this configuration, the desired target NOx conversion efficiency η NOx can also be determined depending on the state of the second catalyst. Here, the magnitudes of the actual NOxconversion efficiency η NOx,Ist and / or actual NH3 einzel levels m NH3,Ist,i and / or actual NH3 gesamt level m Nox,ges can be used. The method is then continued in a step 210.In a first step 210, a release condition for the method is checked. For this purpose, a dosing readiness for the SCR injection system is determined by the control unit 100. This is done primarily by means of the feedback of the pressure p fed back by the pump unit. If the pressure p of the SCR injection system determined by the control unit 100 exceeds a predefinable pressure threshold, the SCR injection system is thus ready for metering. In addition, the SCR catalyst temperature T SCR of the SCR catalyst 22 can be determined by the control device 100. If the SCR catalyst temperature T SCR exceeds a predeterminable temperature T Kat,min, in particular 180° C., the operating temperature for the SCR catalyst 22 is thus reached and the release condition is issued. The method is then continued in a step 220.In a step 220, a starting value for the method is determined. This is preferably an estimated value.In a first embodiment, the starting value can be determined on the basis of a solution of the inverse model F -1 at a previous time t-1, this approach being based on the fact that no abrupt changes in the NOxand NH3 konzentrationen occur within time intervals of less than 100 ms.In a second specific embodiment, the starting value is ascertained on the basis of an actual NH3 füllstands level distribution NH3 Ist that is current for SCR catalytic converter 22. The current actual NH3 füllstands level distribution NH3 Ist is calculated from step 200 by means of the SCR model F and the continuously determined variables.In a third embodiment, a total NH3 soll fill level NH3 ges from a substitute model, preferably a characteristic curve stored on the control device 100, is used, wherein the total NH3 soll fill level NH3 ges is distributed uniformly over the SCR catalyst 22.In a fourth embodiment, the starting value can be determined from a calculation for an SCR catalytic converter 22 with only one brick. For this purpose, the inverse model F -1 is solved analytically for the one brick case and a total NH3 soll fill level NH3is obtained ges for the SCR catalyst 22.Thus, a starting value for the method can be determined and the method can be continued in a step 230.In a step 230, the NOx concentrations and the NH3 concentrations are then iteratively solved according to the formulae (3) and (4): with the function for calculating NOx concentrations, r NH3 the function for calculating NH3 concentrations, mNH3,i ith NH3 mass, T SCR,i the temperature of the ith brick, the NH3 concentration of the previous brick, the NOx concentration of the previous brick, x O2 the oxygen concentration and p the exhaust gas pressure.The method is then continued in a step 240.In a step 240, a balancing of the NOxand NH3 konzentrationen upstream and downstream of each brick i is determined according to formula (7): with η Nox,i the NOxconversion efficiency of the ith brick, the NOxconcentration of the previous brick, the NH3concentration of the previous brick and ξ a stoichiometric factor corresponding to the NH3 äquivalenten from the NOxconversion and in particular in the range of [1:1.3].The method is then continued in a step 250.In a step 250, it is checked whether equation (11) is fulfilled for all brick i of the SCR catalytic converter 22: wherein ∈ tol is a suitable, predefinable tolerance threshold. This tolerance threshold can preferably be interpreted as a rate of change of the setpoint NH3individual fill levels NH3 Soll i.e. it indicates how strongly the fill levels are allowed to change in the defined stationary point. This can be determined in an application phase for the SCR catalytic converter 22.If the tolerance threshold is not fulfilled for all brick i, the method is continued again in step 220, wherein there an adaptation of the NH3 füll for the current estimated value (at the start of the method, it is the starting value). This can be effected, for example, by means of a Newton-based method. For this purpose, the derivative matrix (Jacobi matrix) of the function to be minimized for formula (11) must be calculated:The quotient of the function to be minimized and the Jacobi matrix determines the step size and direction of the adaptation of the current estimated value according to the following formula:In the event that the condition of formula (11) is fulfilled for all brick i, the method is continued in a step 260.In a step 260, the desired NH3 einzel levels are then combined to form a desired NH3 füllstand level NH3 Soll and the method can be continued in a step 270.In a step 270, the determined desired setpoint NH3 füllstand level is added to the dosing strategy and set between setpoint and actual values by means of a dosing strategy, preferably by means of a P-control. The proportion of the P regulator is added to the pilot quantity. Subsequently, the method can be started or ended from the front in step 200.

Claims

Method for exhaust gas aftertreatment of an internal combustion engine (10) having at least one SCR catalyst (22), to which exhaust gas of an internal combustion engine (10) is fed via an exhaust gas path (2), wherein the at least one SCR catalyst (22) is virtually divided into n-brick (n b) in the direction of flow of the exhaust gas, wherein desired NH3 desired filling levels (NH3 Soll,i) for the at least one SCR catalyst (22) are determined as a function of a desired desired desired NOx conversion efficiency (η NOx) wherein a steady state is assumed for the determination of the desired NH3 desired filling levels (NH3 Soll,i), wherein the desired NH3 soll levels (NH3 Soll,i) for the at least one SCR catalyst (22) are determined by means of an inverse SCR model (F -1) by means of a current NOxconcentration (NOx Us) upstream of the SCR catalyst (22), an SCR catalyst temperature (T SCR), in particular of SCR catalyst temperatures of the brick (T SCR,i), an exhaust gas mass flow (dmEG), an oxygen concentration (O 2), an exhaust gas pressure (p) upstream of the SCR catalytic converter (22) and a desired desired NOx conversion efficiency (η NOx) are determined, wherein a regulation of the SCR catalytic converter (22) is carried out as a function of the determined NH3 desired fill level (NH3 Soll).Method according to Claim 1, characterized in that, for the assumed steady state, a balancing of the input and output concentrations of the SCR catalyst (22) on the basis of the NH3 fill level is described as follows: ∂ m N H 3, i ∂ t = 0 = - ξ x NOx, i - 1 d x η NOx, i + x NH3,i - 1 ds - x NH3,i ds, with η Nox,i of the NOx conversion efficiency of the i-th brick, and ξ a stoichiometric factor which corresponds to the NH3 equivalents from the NOx conversion and is in particular in the range of [1:1.3].Method according to Claim 1, characterized in that an NH3 metering quantity is determined as a function of a difference between the desired NH3 nominal fill level (NH3 Soll) and a current NH3 fill level (NH3 Ist) and is metered by means of the SCR injection system (25).Method according to Claim 1, characterized in that the desired NOx conversion efficiency (η NOx) is determined or fixedly predefined from a characteristic diagram as a function of the SCR catalyst temperature and / or the exhaust gas mass flow (dm EG).Method according to Claim 1, characterized in that the inverse SCR model (F -1) is determined by means of a reaction kinetic or a spatially resolved model.Method according to Claim 1, characterized in that the desired NH3 desired fill level (NH3 Soll) is composed of the sum of the desired NH3 desired fill levels (NH3 Soll,nb) of the n virtual brick.Method according to one of the preceding claims, characterized in that the regulation is carried out on any desired virtual brick (n b), in particular on the first brick (n 1).Method according to one of the preceding claims, characterized in that the total NOx conversion efficiency (η Nox,ges) of the SCR catalyst (22) is determined as a function of the NOx conversion efficiencies (η Nox,i) of the individual virtual brick according to the following formula: η NOx,ges = 1 - ∏ 1 n b (1 - η NOx,i ) Method according to Claim 4, characterized in that two SCR catalytic converters (22) are arranged one after the other in the direction of flow of the exhaust gas, wherein the desired desired setpoint NOx conversion efficiency (η NOx,Soll) is determined as a function of the state of the second catalytic converter, wherein an actual NOx conversion efficiency (η NOx,Ist) and / or actual NH3 individual fill levels (m NH3,Ist,i) and / or actual NH3 total fill level (m NOx,ges) are used for the calculation.Computer program which is configured to carry out a method according to one of Claims 1 to 9.Electronic storage medium comprising a computer program according to claim 10.Device, in particular control device (100), which is configured to execute a method according to one of Claims 1 to 9.

Images