EP3504407A1

EP3504407A1 - Method for controlling the quality of a reducing agent solution in an scr catalyst

Info

Publication number: EP3504407A1
Application number: EP17732885.3A
Authority: EP
Inventors: Christian Heimann; Tobias Pfister
Original assignee: Robert Bosch GmbH
Current assignee: Robert Bosch GmbH
Priority date: 2016-08-24
Filing date: 2017-06-23
Publication date: 2019-07-03
Also published as: DE102016215864A1; CN109642485B; KR20190038929A; CN109642485A; WO2018036686A1

Abstract

The invention relates to a method for controlling the quality of a reducing agent solution for an SCR catalyst with a nitrogen oxide regulator. At the start of the method, the SCR catalyst is set to a predetermined nitrogen oxide conversion rate. This is followed by integration of an actual ammonia mass flow to reach an actual ammonia mass, and integration of a modelled ammonia mass flow to reach a modelled ammonia mass. A metering mass factor is then calculated as a quotient of the actual ammonia mass and the modelled ammonia mass, and finally, if said metering mass factor (facDos) exceeds a set first threshold (311) (at point 312), an error state is generated which assigns the reducing agent solution a bad quality rating.

Description

Description title

Method for quality control of a reducing agent solution in an SCR catalyst

The present invention relates to a method for quality control of a reducing agent solution in an S CR catalyst with a nitrogen oxide regulator. Furthermore, the present invention relates to a computer program that executes each step of the method according to the invention, when it runs on a computing device, as well as a machine-readable storage medium, which stores the computer program. Finally, the invention relates to an electronic control device which is set up to carry out the method according to the invention.

State of the art

Today, to reduce nitrogen oxides (NOx) in the exhaust gas of

Motor vehicles including S CR catalysts (Selective Catalytic

Reduction) used. This nitrogen oxide molecules, which are located on a S CR catalyst surface, in the presence of ammonia (N H3) as a reducing agent, reduced to elemental nitrogen. The reducing agent is in the form of a urea-water solution, is eliminated in the ammonia, commercially known as Ad Blue ^®, provided and injected by means of a metering module upstream of the S CR- catalyst in an exhaust line. The determination of a desired metering rate takes place in an electronic control unit, in which strategies for operation and monitoring of the SCR system are stored.

The S CR catalysts known today store ammonia at their catalyst surface. The storage capacity is decisive for the temperature the catalyst surface and decreases with increasing temperature. The more ammonia bound to the catalyst surface and available for reduction, the higher the nitrogen oxide conversion rate. As long as the storage capacity of the S CR catalyst is not exhausted, excessively metered reducing agent is stored. On the other hand, if the metering unit makes available less reducing agent than would be necessary for the complete reduction of the nitrogen oxides present in the exhaust gas, the reduction of the CO 2, which continues to take place at the catalyst surface, becomes necessary

Nitrogen oxides, which reduces ammonia level. Today's standard dosing strategies for SCR systems include a level control, which sets an operating point in the form of a target value for the ammonia level in the S CR catalyst. This operating point is chosen so that the ammonia level is high enough to ensure both a high nitrogen oxide conversion rate, as well as a buffer against sudden excessive nitrogen oxide masses, so-called nitric oxide peaks. On the other hand, the ammonia level is selected depending on a maximum ammonia mass that can store the S CR catalyst. As a result, especially at rapid temperature rises, avoiding excessively metered ammonia unused

Catalyst surface happens; This is also referred to as ammonia slip.

A quality of the urea water solution is called quotient of a

Urea mass from which ammonia splits off and defines the total mass of the urea-water solution. DE 10 2006 055 235 A1 describes a method for detecting the quality of the reducing agent.

In this case, a nitrogen or ammonia concentration in the exhaust gas is determined by an exhaust gas sensor arranged downstream of the S CR catalyst. Subsequently, these determined values are compared with comparative values and, in the case of a deviation, a reduced quality of the reducing agent is concluded. Common nitric oxide sensors show a cross-sensitivity to ammonia, ie their sensor signal not only contains the nitric oxide concentration, but also shows a composite signal of nitrogen oxide and ammonia. In the case of a nitrogen oxide sensor arranged downstream of the S CR catalyst, an increase in the sensor signal can be attributed both to a decreasing nitrogen oxide conversion rate, ie to an increase in the nitrogen oxide concentration, and to ammonia slip and a corresponding increase in nitrogen oxide Ammonia concentration, point out. As a result, a direct distinction between nitrogen oxide and ammonia is not possible with these sensors.

From the relationship between ammonia level and nitrogen oxide conversion rate, on the one hand results in an increase in the nitrogen oxide conversion rate, when a superstoichiometric dosage of

Reducing agent lifting the stored in the S CR catalyst

Ammonia mass causes. On the other hand, the nitrogen oxide conversion rate remains the same if the S CR catalyst is already optimally operated. If the measured nitrogen oxide conversion rate decreases in this case, that is, the sensor signal described above increases, this can be due to ammonia

downstream of the S CR catalyst. In this case, it can be assumed that the maximum storage capacity of the SCR catalyst has been exhausted and thus excessively metered ammonia unused passes the catalyst surface, ie an ammonia slip takes place.

DE 10 2010 002 620 A1 describes an adjustment of the dosing mass by means of an adaptation factor, which indicates the ratio between nominal and actual dosing mass. This directly changes a precursor mass of the reducing agent and serves to regulate a nitrogen oxide concentration measured by the sensor downstream of the SCR catalyst to a modeled nitrogen oxide value. The dosing strategy adapts to the respective system and to longer-lasting environmental influences with the aid of an I-controller and can thus contribute to the number of necessary adaptation interventions

reduce systematic errors. In addition, the regulation can also take into account very large and spontaneous changes, for example when refueling with the wrong reducing agent. The I controller acts very accurately, but also correspondingly slow and can, depending on the quality of the reducing agent and its dilution, several hours to detect a fault and to control needed. The detection of the dilution of the reducing agent takes place here via the adaptation factor or via a threshold value which corresponds to the adaptation factor. Another possibility for monitoring the quality of the reducing agent is a quality sensor, as described in DE 10 2014 211 010 A1. This quality sensor is arranged directly in the reducing agent tank and analyzes the ammonia concentration by means of acoustic and / or optical methods. DE 10 2012 209 240 AI describes a method for

Plausibilisierung the quality sensor, as this has a susceptibility due to the action of the reducing agent has.

A third possibility to monitor the quality of the reducing agent is passive monitoring of the S CR catalyst, in which a nitrogen oxide concentration upstream and downstream of the S CR catalyst is detected by nitrogen oxide sensors and an efficiency is determined therefrom. Error detection occurs when a limit efficiency is exceeded for a set duration. The process is especially suitable strong

To recognize dilutions of the reducing agent, as are recognized by the relatively large tolerance of the system, in particular the nitrogen oxide sensors, only very large deviations in efficiency. To cope with these large and rapid changes in efficiency, e.g. as a result of refueling with water, was in DE 10 2012 221 574 AI a

Level observer shown. This fast P-controller permanently compensates the modeled nitrogen oxide concentration downstream of the S CR catalyst with the corresponding signal from the corresponding nitrogen oxide sensor. In case of a deviation, the P-controller can within a few seconds the

Perform level control until the nominal efficiency is reached again.

Disclosure of the invention

The method is used for quality control of a reducing agent solution for a S CR catalyst comprising a nitrogen oxide regulator, which at the beginning of the

Process controls the S CR catalyst so that it meets a predetermined nitrogen oxide conversion rate, that is, that the S CR catalyst a

correspondingly high amount of reducing agent is provided.

Preferably, a maximum nitrogen oxide conversion rate is set. In a further step, both an actual ammonia mass flow, as well as a modeled ammonia mass flow, in particular via an I regulator, integrated into an actual ammonia mass or a modeled ammonia mass. Subsequently, a metering factor is calculated by a quotient of the two

Ammonia masses is formed. The metered amount of the reducing agent is increased by the nitrogen oxide regulator to compensate for a poorer quality of the reducing agent and to achieve the high nitrogen oxide conversion rate. It follows that the increase in the dosage of the

Reducing agent leads to an increase in the metering factor. If this calculated metering factor exceeds a predetermined first threshold, an error condition is generated in the system, which ascribes a poor quality to the reducing agent solution. Finally, an error memory entry is made in an electronic control unit which controls the metering of the reducing agent.

Preferably, the calculation of the metering factor is carried out only if it was determined in a query that the actual ammonia mass and / or the modeled ammonia mass has exceeded a predetermined limit mass. This limit mass is dependent on the specified nitrogen oxide conversion rate. As a result, it is ensured that the quotient between the actual ammonia mass and the modeled

Ammonia mass is large enough so that an increase in the metering factor can be seen and this provides meaningful results.

In addition, the error state can be canceled again if the metering factor undershoots a specified second threshold. In this case, the quality of the reducing agent is again sufficiently good to again perform a conventional dosage.

In the determination of the actual ammonia mass, an opening duration and an opening frequency of a metering valve, which injects the reducing agent upstream of the S CR catalyst, may preferably be included. In addition, the modeled ammonia mass can preferably correspond to a calculated pilot mass. This pilot mass represents a calculated mass of ammonia that is injected to produce a desired ammonia mass Nitrogen oxide concentration downstream of the S CR catalyst to achieve. This is calculated by multiplying a nitrogen oxide concentration upstream of the SCR catalyst with an exhaust mass flow and a stoichiometric factor and a modeled nitrogen oxide efficiency, wherein the stoichiometric factor is a ratio between nitrogen oxide mass and

Represents ammonia mass, as a result, the reduction takes place.

According to another aspect, the integration of the actual and the modeled ammonia mass is permanently interrupted, but the calculation of the metering factor is time-discrete. As a result, the frequency of calculating the dosing factor may be dependent on a duration until a predetermined limit is reached in the integration of the actual or modeled ammonia mass. Preferably, this limit corresponds to the previously described

Limit dimensions. Then a discrete filter can be used to calculate the

Dosiermassenfaktors be used.

Preferably, a sensitivity of the filter is dependent on a time interval to a last night Nachankereignis the reducing agent.

Particularly preferred is the sensitivity of the filter over a

Sensitivity coefficient expressed and included in the calculation of the Dosiermassenfaktors by means of the filter.

The computer program is set up every step of the procedure

perform, especially when it is performed on a computing device or controller. It allows the implementation of the method in a conventional electronic control unit without having to make any structural changes. For this it is on the machine-readable

Storage medium stored.

By putting the computer program on a conventional

electronic control unit, the inventive electronic

Control unit, which is set up, the quality control of

Perform reducing agent solution.

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

FIG. 1 shows a diagram of a nitrogen oxide conversion rate depending on an ammonia level during filling and emptying of an SCR catalyst and ammonia slip, according to the prior art.

FIG. 2 shows a flow diagram of an embodiment of the invention

inventive method.

FIG. 3 shows a diagram of a time profile of an actual and a modeled ammonia mass in the S CR catalyst, according to FIG

Embodiment of the method according to the invention.

Figure 4 shows a time course of a metering factor, according to a

Embodiment of the method according to the invention.

Figure 5 shows a time course of a metering factor, according to another embodiment of the method according to the invention.

Embodiment of the invention

In conventional methods, a metering valve meters a reducing agent in a metering module upstream of an S CR catalyst, on the surface of which nitrogen oxides, by means of ammonia liberated from the urea-water solution, are converted into elemental nitrogen (not shown). A nitrogen oxide conversion rate NOxKonvert, which indicates how much nitrogen oxide is converted, is dependent on an ammonia level FN H3 of the S CR catalyst and thus on the metered ammonia mass mN H3. In FIG. 1, such a nitrogen conversion rate is NOxKonvert over the ammonia level FNhta at a filling 100 and an emptying 101 of the S CR catalyst, according to one embodiment of a conventional one

Procedure, applied. A nominal value 102 for the ammonia level FN H3 is entered in the diagram and represents the optimum ammonia filling compound FN H3, in which the S CR catalyst is optimally operated in this exemplary embodiment. In addition, a buffer region 103 with respect to nitrogen oxide peaks, which is intended to compensate for a suddenly occurring excessive nitrogen oxide mass, and a buffer region 104 with respect to an ammonia slip 105, in which ammonia unused passes through the catalyst surface, are entered around the desired value 102. An ammonia mass m N H3, which is measured at said ammonia slip 105 downstream of the S CR catalyst, is also shown above the ammonia level FN H3.

For the operation of the S CR catalyst is therefore an accurate knowledge of the ammonia FN H3 and thus the ammonia mass mN H3, by means of

Reductant solution was metered required. Since the ammonia mass mN H3 with a urea mass mChUIN O corresponds to the reducing agent, it depends on a quality Q of the reducing agent, wherein the quality Q represents a quotient of the mass of urea m CH N20 and the total mass m _{ge £} of the reducing agent:

mCH ₄ N ₂ 0

Q

(Formula 1) m tot

FIG. 2 shows a flow chart according to an embodiment of the invention

inventive method. Initially, the S CR catalyst is set to a predetermined high nitrogen oxide conversion rate NOxKonvert 200. Subsequently, a query 201 is started if there is no relevant system error 202, all nitric oxide sensors used for the measurement are ready to measure 203, the dosing module 204 is active and the metered amount of reducing agent, for example by concurrent methods, is not limited 205. If all operating parameters are met, a measured actual ammonia mass flow is integrated into an actual ammonia mass mNHslst by means of an I regulator 206, the latter being opened over an opening time t open and a

Opening frequency föff the metering valve is estimated. At the same time, a modeled ammonia mass flow is integrated 207 into a modeled ammonia mass mNHsMod. This corresponds to an ammonia pilot mass mN H3Vor, which is calculated by the following formula 2: mNH ₃ before = f _stöch (formula 2)

^' NOx ^{' m} exhaust ^' Ή NOx Here, c _{NOx is} a nitrogen oxide concentration upstream of the SCR catalyst, rh _Ab an exhaust gas mass flow in the exhaust line, η _ΝΟχ a modeled nitrogen oxide efficiency of the S CR catalyst and f _stöch a stoichiometric factor. The stoichiometric factor f _stöch gives that

Ratio of the ammonia mass mNH3 to the nitrogen oxide mass mNOx. A molar mass M _NO ^ of nitrogen dioxide corresponds approximately to an effective molar mass M _NO of the nitrogen oxides present in the exhaust gas.

Furthermore, it results from the unillustrated reaction equations of the ongoing SCR reactions that the amount of ammonia n _{m of} the

Amount of nitrogen dioxide corresponds to _N0 ^ and consequently the stoichiometric factor f _{stöch corresponds to} a ratio of the molar masses M _NH ^ of

Ammonia and M _NO ^ forms of nitrogen dioxide.

Accordingly, the stoichiometric factor f _{stöch is calculated} according to formula 3:

_ mNH ₃ _M _ms - n _ms _ M _ms (formula 3) mNOx Μ _Νθ2 - η _Νθ2 M _NOi

Exceeds 208 either the actual ammonia mass mN Halst or the modeled ammonia mass mNHsMod a predetermined limit mass

mNHsGrenz, which is NOxKonvert depending on the nitrogen oxide conversion rate, a metering factor facDos is calculated 209 as the quotient of the actual ammonia mass mNHslst and the modeled ammonia mass mNHsMod, according to formula 4.

, _^ mNH st,, facDos = (formula 4) mNH ₃ Mod

In FIG. 3 this is explained on the basis of an exemplary embodiment. Here is the time course of the actual mNHslst and the modeled

Ammonia mass mNHsMod shown. The limiting mass mNHsGrenz was chosen at this point at 6 g. At point 301, the actual ammonia mass mNHslst exceeds the predetermined limit mass mNHslimit for a time t of approximately 22.5 minutes. In the flowchart of FIG. 2, the I-controllers used in the integrations 206 and 207 are reset in a sibling step 210. It should be noted that these two integrations 206 and 207 run permanently, but the calculation 209 of the dosing factor facDos is done discretely. The frequency of the calculation 209 of the dosing factor facDos is dependent on a duration until, in step 208, the limit mass mNhtaGrenz has been exceeded, that is to say 22.5 minutes. In this case, the calculation 209 of the dosing factor facDos takes place via an exponential weighted moving average (EWMA) filter 211, which works on the basis of the formula 5.

S _t = a - Y _t + (l - a) - S _t _, where S, = Y _l r r t> \ (formula 5)

S _t indicates the value of the EWMA at the time t, Y _t are the input raw values and a represents a sensitivity coefficient of the EWMA filter, wherein this is varied as a function of a time interval of a last night Nachankereignisses the reducing agent. In an exemplary embodiment, this is possible on the basis of an integrator, which detects a nitrogen oxide mass mNOx and is reset in the case of refueling. The evaluation by the EWMA filter is activated on initial initialization and / or resetting only when enough raw values Y _{t have been} recorded.

The dosing factor facDos represents a manipulated variable for a nitric oxide regulator. To obtain a lower quality Q of the reducing agent, the predetermined high

To achieve the nitrogen oxide conversion rate NOxKonvert, the nitrogen oxide regulator must increase the ammonia filler mass FN H3 (see FIG. 1) by increasing the metered reducing agent mass. In other words, to achieve the same NOx conversion rate, NOxConvert is higher

Reducing agent mass required, which leads to an increase in the dosing factor facDos.

In a further step 212, it is checked whether the dosing factor facDos, as shown in Figure 4, has exceeded a first threshold 311. This takes place at marked point 312 at approx. 25 minutes. In the flowchart in FIG. 2 it is subsequently provided to generate an error state in the system 213, which attributes a poor quality Q to the reductant solution and then creates a fault memory entry 214.

In an alternative embodiment, a decreasing metering factor facDos is further provided, which corresponds to an improvement in the quality Q, e.g. by replacing the reducing agent. It is checked in an additional step 215 whether the dosing factor facDos, as shown in Figure 5, has fallen below a second threshold 321. This takes place at the marked point 322 at approx. 70 minutes. Now, as shown in the flowchart of Figure 2, the error state is removed 216.

Claims

claims

A method of quality control of a reductant solution for a nitrogen oxide regulator S CR catalyst, comprising the steps of:

Control (200) of the S CR catalyst to a predetermined

Nitrogen oxide conversion rate (NOx conversion);

Integrating (206) an actual ammonia mass flow to an actual ammonia mass (mlNH-Ust);

Integrating (207) a modeled ammonia mass flow into a modeled ammonia mass (mNHsMod);

Calculation (208) of a metering factor (facDos) as a quotient of the actual ammonia mass (mN Halst) and the modeled

(mNHsMod) ammonia mass; and

Generating (213) an error condition which is the

Reducing agent solution a poor quality (Q) as the dosing factor (facDos) has exceeded a predetermined first threshold (311) (212).

2. The method according to claim 1, characterized in that a

Query (208) is whether the actual ammonia mass (mN Halst) and / or the modeled ammonia mass (mNHsMod) is a

exceeds specified limit mass (mNHsGrenz) and the

Calculation (208) of the dosing factor (facDos) is carried out only if the actual ammonia mass (mN Halst) and / or the modeled ammonia mass (mNHsMod) has exceeded the limit mass (mNHslimit). A method according to claim 1 or 2, characterized in that the error state is canceled (216) when the metering factor (facDos) falls below a fixed second threshold (321) (215).

Method according to one of the preceding claims, characterized

characterized in that an opening duration (töff) and a

Opening frequency (föff) of a Reduktionsmitteldosierventils in the determination (206) of the actual ammonia mass (m lNH-Ust) is included.

Method according to one of the preceding claims, characterized

characterized in that the modeled ammonia mass (m N HsMod) corresponds to a calculated pilot mass (m N htavor), which by multiplying a nitrogen oxide concentration {c _NOx ) upstream of the

S CR catalyst with an exhaust gas mass flow {rh _Ab ) and a stoichiometric factor {f _stöch ) and a modeled nitrogen oxide efficiency {η _ΝΟχ ) is calculated.

Method according to one of the preceding claims, characterized

characterized in that the integration (206, 207) of the actual ammonia mass (mlNH-Ust) and the modeled (mN HsMod)

Ammonia mass permanently expires and the calculation (209) of the metering factor (facDos) is discrete in time.

A method according to claim 6, characterized in that the

Frequency of the calculation (209) of the dosing factor (facDos) is dependent on a duration until the integration (206, 207) of the actual ammonia mass (mN Hslst) or the modeled ammonia mass (mN HsMod) exceeds a predetermined limit value (mN H3 limit).

A method according to claim 7, characterized in that a discrete filter for calculating (208) of the metering factor (facDos) is used (211).

9. The method according to claim 8, characterized in that a sensitivity of the filter depends on a time interval to a last night Nachankereignis of the reducing agent.

10. Computer program which is set up every step of the

Method according to one of claims 1 to 9 perform.

11. Machine-readable storage medium on which a

Computer program according to claim 10 is stored.

12. Electronic control device which is set up to perform a quality control of a reducing agent solution by means of a method according to one of claims 1 to 9.