Method for operating a conveying and metering system

Abstract

This invention relates to a method for operating a conveying and metering system, the conveying and metering system having a conveying module, a metering module, a pressure line connecting the conveying module and the metering module, and a return section of the conveying module. If during the operation time (t... a1L t a1 -t a5 During the subsequent pressure formation period, the pressure (p) in the pressure line did not reach the predetermined pressure level (p). S Then, the exhaust of the pressure side portion of the system is carried out through at least one ventilation cycle (152, 153, 154, 155, 156), in which the metering module is turned on. This invention specifies that the control time (t) is obtained through an empirical function. a1 -t a5 ).

Description

Technical Field

[0001] This invention relates to a method for operating a delivery and metering system with a blocked return line. Here, venting is performed and the control time is obtained through an empirical function. Furthermore, this invention relates to a computer program and a machine-readable storage medium, wherein when the computer program is run on a computing device, the computer program implements each step of the method, and wherein the machine-readable storage medium stores the computer program. Finally, this invention relates to an electronic controller configured to implement the method according to the invention. Background Technology

[0002] Currently, the SCR (Selective Catalytic Reduction) method is used in the aftertreatment of exhaust gases from internal combustion engines. S elective C atalytic R The principle of reducing nitrogen oxides (NOx) in exhaust gas is described in DE 103 46 220 A1. Here, a 32.5% aqueous urea solution (HWL), commercially known as AdBlue®, is dispensed into the exhaust gas. Typically, a delivery and metering system with a delivery module and a metering module is configured for this purpose. The HWL is delivered from the reducing agent tank to the delivery module via a suction line by means of a delivery pump and to the metering module via a pressure line. The metering module dispenses the HWL upstream of the SCR catalyst into the exhaust gas stream. Ammonia is released from the HWL and then combines with the reaction surface of the SCR catalyst. There, the ammonia combines with nitrogen oxides, thereby forming water and nitrogen. The delivery and metering system has a return section from the delivery module to the reducing agent tank, through which excess HWL can be returned to the reducing agent tank.

[0003] When the internal combustion engine stops, the HWL from the delivery and metering system is pumped back to the reducing agent tank by means of a delivery pump. At this point, the metering module is opened to allow air to enter the pressure line and the suction line. Meanwhile, HWL often remains in the return line. Under certain conditions, such as when the external temperature is too low, HWL may crystallize and thus block the return line.

[0004] When the internal combustion engine restarts, pressure is formed when the metering valve is closed. Because air cannot escape through the return section, pressure cannot be formed to a predetermined level. It is known that a ventilation cycle is performed in this situation, during which the metering module is opened to release air from the pressure side of the system. Thus, the reducing agent solution can be delivered from the reservoir to the suction line, delivery module, and pressure line. Correspondingly, new reducing agent solution also flows to the return section, where it encounters a blockage caused by crystallized reducing agent. The new reducing agent solution removes this blockage, in particular by dissolving the crystallized reducing agent, thereby reopening the return section. Now, air is no longer contained in the delivery and metering system, the reducing agent solution reaches the pressure level required for the dosage, and the delivery and metering system is ready. Traditionally, a fixed operating time is waited until the ventilation cycle is performed. Fixed operating times are obtained beforehand for the delivery and metering system without specificity and then applied to different configurations.

[0005] As known from DE 10 2013 218 552 A1, the pressure side portion of the delivery and metering module is vented via at least one ventilation cycle, for which a time-independent release condition is pre-defined, during which the metering unit is opened. However, in the mentioned method, the release condition is a pressure threshold that must be exceeded before venting, in order to prevent erroneous metering into the exhaust gas system and unnecessary ventilation cycles. Summary of the Invention

[0006] This invention relates to a delivery and metering system comprising a delivery module and a metering module interconnected by a pressure pipeline. A liquid medium is delivered from a storage tank via a suction pipeline by a delivery pump of the delivery module and supplied under pressure to the metering module via a pressure pipeline, where it is then metered by a metering valve. Furthermore, a return flow section is provided between the delivery module and the storage tank, through which the liquid medium is returned to the storage tank. Particularly relevant is a delivery and metering system for a reducing agent solution, such as an aqueous urea solution, used in an SCR catalyst, which is metered from an upstream metering module into the exhaust gas system of an internal combustion engine.

[0007] When the internal combustion engine stops, the reducing agent solution is pumped back from the delivery and metering system to the reducing agent tank. Here, the reducing agent solution often remains in the return section, where it crystallizes under certain conditions, such as when the external temperature is too low, and thus blocks the return line.

[0008] When the internal combustion engine restarts, pressure is established when the metering valve is closed. The pressure of the reducing agent solution on the metering module is crucial for accurate and satisfactory metering. Therefore, the system should operate at a predetermined pressure level. Because air cannot escape through the return flow section, pressure cannot be established up to the predetermined pressure level during pressure formation. Consequently, if the pressure in the pressure line during pressure formation does not reach the predetermined pressure level after a specific time (also referred to here as the control time), it can be assumed that air is present in the pressure side portion of the delivery and metering valves and that this air cannot escape through the return flow section, i.e., the return flow section is blocked.

[0009] The method specifies that if a predetermined pressure level is not reached during the pressure formation period after the control time, the pressure-side portion of the system is vented. Here, the metering module is opened in a ventilation cycle. Air is now able to escape from the pressure-side portion of the system via the metering module.

[0010] This specification stipulates that the manipulation time is obtained through an empirical function. As an empirical function, a reproducible relationship between empirical values ​​based on past observations, measurements, or other analyses is used, rather than directly obtaining the functional relationship.

[0011] Therefore, for the delivery and metering system, the control time until the ventilation cycle is triggered can be obtained based on past ventilation cycles and their operation, as well as the corresponding time until the pressure reaches a predetermined pressure level. Furthermore, for the same type of delivery and metering system with the same or at least similar configuration, the control time until the ventilation cycle is triggered can be obtained based on past ventilation cycles and their operation, as well as the corresponding time until the pressure reaches a predetermined pressure level. Especially in such a configuration, the geometry of the delivery and metering system is important. For vehicles of the same model, it can be assumed that their delivery and metering systems have the same or at least similar configuration and geometry. The past ventilation cycles considered at the time of acquisition can be used for a "normal" state, a "problematic" state, a "defective" state, or any combination of the aforementioned states, wherein in the "normal" state, pressure formation and dispensing are performed routinely and, in particular, there is no pipe blockage; in the "problematic" state, the return pipe is blocked and the blockage is relieved by venting; and in the "defective" state, the return pipe is blocked and the blockage cannot be relieved even by venting.

[0012] Such control offers advantages over conventional control with fixed control times, namely: the actual configuration of the delivery and metering systems is considered in the selection of control times, where fixed control times are pre-acquired and applied to various different delivery and metering systems. As a result, control times up to the completion of ventilation cycles can be specifically selected for the corresponding delivery and metering systems, significantly reducing these control times compared to fixed control times. Due to the reduced control time, exhaust occurs relatively earlier, thereby allowing the reducing agent solution to be delivered to the suction line, delivery module, pressure line, and return section earlier, thus clearing blockages in the return section earlier, reaching the required pressure level for the dosage earlier, and ultimately making the delivery and metering systems ready earlier to facilitate exhaust gas aftertreatment. Furthermore, the cost of calibrating different delivery and metering systems is reduced through the use of empirical functions.

[0013] The above description applies primarily to the operating time of the first ventilation cycle. If the ventilation cycle is insufficient to completely remove air from the delivery and metering system or to remove blockages in the return section, an additional ventilation cycle can be performed. If the pressure in the pressure line does not reach a predetermined pressure level during another pressure formation period after the end of the previous ventilation cycle and following another operating time, an additional ventilation cycle is performed. The predetermined pressure level is preferably kept the same, as it corresponds to the feed rate required for operating the SCR catalyst, but it can also be matched purely theoretically. The additional operating times for the respective ventilation cycles can be different and obtained separately through empirical functions. See the above description for empirical functions.

[0014] By using empirical functions, as described above, each operation time can be reduced compared to the conventional fixed operation time, thereby reducing the total operation time. This reduces the total time until exhaust, allowing the reducing agent solution to be delivered to the suction line, delivery module, pressure line, and return section earlier, thereby clearing blockages in the return section earlier, thereby reaching the required pressure level for the dosage earlier, and ultimately preparing the delivery and metering system earlier to facilitate exhaust gas aftertreatment.

[0015] Preferably, the empirical function (or each function) is a self-learning function that finds the optimal control time for the "normal state" based on the pressure change curve during dispensing. This optimal control time corresponds to the time until the pressure in the corresponding delivery and metering system reaches a predetermined pressure without air in the pressure line and without blockage in the return section. The self-learning function is preferably executed using a neural network. The configuration of the delivery and metering system, its control, and the pressure change curve are used as input parameters, and the optimal control time for the delivery and metering system is obtained as the output parameter. Using the self-learning function, the optimal control time can be found and learned, thereby minimizing the time until venting is performed and ultimately until the delivery and metering system is ready. Specifically, the optimal control time is obtained for each configuration of the delivery and metering system, and thus particularly for each vehicle type, or rather for each vehicle.

[0016] As a supplementary or alternative approach, if it is determined by a function that air is contained within the pressure-side portion of the system, ventilation circulation can be performed. Preferably, this involves measuring the gas volume in the pressure-side portion of the system, particularly in the pressure lines, and comparing the measured gas volume to a pre-defined threshold. If the measured gas volume is greater than the threshold, ventilation circulation is performed. For example, the threshold can be chosen to be zero. In this case, ventilation circulation is performed as soon as gas, primarily air, is present in the pressure-side portion of the system. However, other thresholds can also be defined. These thresholds determine the acceptable gas volume for the proper operation of the delivery and metering module.

[0017] Preferably, the database is located on an online computer network (cloud). This database can store the time until a predetermined pressure level is reached, and the acquired control time. The empirical function, and especially the self-learning function, can retrieve the stored data and then use this data to obtain the current control time for the current system.

[0018] In cases where the pressure fails to reach a predetermined pressure level after a predetermined number of ventilation cycles, and the delivery and metering system is not ready, an error can be output and emergency measures can be initiated, such as throttling of an internal combustion engine.

[0019] A computer program is configured to perform each step of the method, particularly when executed on a computing device or controller. This computer program enables the implementation of the method in a conventional electronic controller without requiring structural modifications. For this purpose, the computer program is stored on a machine-readable storage medium.

[0020] An electronic controller is obtained by loading the computer program onto a conventional electronic controller, which is configured to operate the conveying and metering system. Attached Figure Description

[0021] Embodiments of the present invention are shown in the accompanying drawings and explained in detail in the following description.

[0022] Figure 1a and Figure 1b The following are illustrations of a delivery and metering system with a blocked return line, during the pressure build-up period (…). Figure 1a ) and the period of exhaust ( Figure 1b A schematic diagram of ( ).

[0023] Figure 2 A graph showing three pressure variation curves for different situations, based on existing technology, and multiple ventilation cycles with fixed operating times is presented.

[0024] Figure 3 A graph showing the pressure change curve is provided, from which the matched control time for the first ventilation cycle is obtained by means of a self-learning function according to a first embodiment of the invention.

[0025] Figure 4 A graph showing the pressure change curve is provided, and according to a second embodiment of the invention, the matched control time for the first ventilation cycle is obtained from the pressure change curve by means of a function that identifies the air in the system.

[0026] Figure 5 A pressure change curve is shown—for which a first ventilation cycle with a matched operating time is performed according to the first embodiment—and a graph of a pressure change curve with a fixed operating time for the first ventilation cycle according to the prior art.

[0027] Figure 6 A graph showing the pressure change curve is provided, and according to a third embodiment of the invention, the matched control time for the second ventilation cycle is obtained from the pressure change curve by means of a self-learning function.

[0028] Figure 7 A pressure change curve is shown—for which a second ventilation cycle with a matching operating time is performed according to the third embodiment—and a graph of a pressure change curve with a fixed operating time for the second ventilation cycle according to the prior art.

[0029] Figure 8A graph showing a pressure change curve according to an embodiment of the method according to the invention, and multiple ventilation cycles with matching operating times, is shown. Detailed Implementation

[0030] Figure 1a and Figure 1b The conveying and metering systems are shown separately during the pressure formation period. Figure 1a ) and during the exhaust period ( Figure 1b A schematic diagram of the above is shown. The delivery and metering system is part of the SCR system of a motor vehicle and is configured to dispense urea aqueous solution (HWL) from the reductant tank 1 into the exhaust gas system (not shown) of an internal combustion engine. The delivery and metering system has a delivery module 2 and a metering module 3. HWL is delivered from the reductant tank 1 via a suction line 5 by means of a delivery pump 4 in the delivery module 2, and then supplied to the metering module 3 via a pressure line 6 arranged between the delivery module 2 and the metering module 3. The pressure of the HWL on the metering module 3 is crucial for accurate and demanding metering. Therefore, a pressure level p must be predetermined. S (See the attached diagram below), the system should operate at this pressure level. In this example, the pre-given pressure level p S The pressure is 7.5 bar. The delivery and metering system also has a return section 7, which connects the delivery module 2 to the reducing agent tank 1 and is configured to return excess HWL back to the reducing agent tank 1.

[0031] In the illustrated case, a blockage 8 exists in the reflux section 7. This blockage 8 is, for example, composed of a crystallizing reducing agent, which may be generated by HWL under unfavorable conditions, such as at low temperatures below 11.5°C. Figure 1a The diagram illustrates a situation where the delivery and metering system is emptied and pressure formation is performed during internal combustion engine startup to achieve the aforementioned pressure level p. S And it is prepared. In the emptied state, air L is in the suction line 5 and in the pressure line 6. The metering module 3 is closed in this state and the return section 7 is blocked as described above. As a result, air L cannot escape and cannot form pressure. This will combine Figure 2 To describe in detail again. In Figure 1bThe exhaust gas from the conveying and metering system is shown in the figure. During the ventilation cycle, the metering module 3 is opened, allowing air L to escape from the system into the exhaust gas system, while simultaneously, HWL is drawn from the reducing agent tank 1 and conveyed to the conveying module 2 and the return section 7. The blockage 8 is cleared by the new HWL. Finally, air L is completely expelled from the system, and HWL also enters the pressure line 6, thereby creating pressure. In this regard, refer to the following figures.

[0032] exist Figure 2 In the graph of pressure p, three pressure change curves (10, 11, and 12) are shown with respect to time t. Additionally, multiple ventilation cycles (90 to 95) are shown, where... Figure 1b As shown, the metering module 3 is opened to allow air L to escape from the system. According to the prior art, during a fixed operation time t... fest The first ventilation cycle 90 is then triggered, the control time of which is determined in advance, not specifically for the delivery and metering system, and is, for example, 60 seconds. After the first ventilation cycle 90 is executed, subsequent ventilation cycles 91-95 can be executed as needed (see below). To keep this example as simple as possible, a fixed control time t is waited between each of the successively executed ventilation cycles 90-95. fest Generally, the control time can vary depending on the existing technology, but it is always predetermined each time. Furthermore, a pre-defined pressure level p is plotted. S It should be at the pre-given pressure level p S To run the system for precise and required dosing. In this example, the pre-given pressure level p S It is 7.5 bar.

[0033] The first pressure change curve 10 represents a situation where pressure formation has occurred normally without ventilation circulation. During the control time t... fest Before the end, the first pressure change curve 10 had exceeded the predetermined pressure level p. S Therefore, ventilation cycles are not triggered and dispensing can be performed normally. The second pressure change curve 11 represents such a case where four ventilation cycles 90-93 have been performed before pressure can be established. These four ventilation cycles 90-93 occur at the determined control time t. fest The actions were then carried out sequentially. This resulted in a total time t. fest _ g Until the pressure change curve 11 has exceeded the predetermined pressure level p SAnd thus the system is prepared to dispense quantities, the total time being a fixed control time t. fest More than four times, or more than 240 seconds in this example. The third pressure change curve 12 represents a situation where the system never reaches the metering readiness state. After the sixth ventilation cycle 95, the third pressure change curve 12 also failed to reach the predetermined pressure level p. S An error will be output and the power of the internal combustion engine will be restricted. However, in this case, it will only occur after a fixed control time t. fest The error could only be identified after six times, or in this example, after more than 360 seconds.

[0034] exist Figures 3 to 7 The pressure graph shows the pressure change curves over time t.

[0035] The following uses Figure 3 and 4 To describe the matched control time t used to obtain the first ventilation cycle from pressure change curves 100 and 110. a1e t a1L Examples of implementations. In Figure 3 In the first embodiment, a self-learning function is provided that, for multiple "normal" dispensings—in which there is no air in the system and the return section 7 is not blocked—learns the pressure change curve 100 and a pre-given pressure level p for the current delivery and metering system. S The condition that the subject belongs to is 101 (in Figure 3 (The example shown here refers only to the dosage). Then, the learned data is used to obtain up to a pre-given pressure level p101. S The optimal time is determined and used as the learned, matched manipulation time t for the first ventilation cycle. a1e Stored. Here, learning is achieved using neural networks. For example, in... Figure 3 As can be seen, the learned matched manipulation time t a1e Shorter than the typical fixed control time t fest .

[0036] exist Figure 4 In the second embodiment, a function is provided that determines, by means of a pressure change curve 110, whether air L is present in the pressure line 6. If the pressure change curve 110 rises to a predetermined pressure threshold p (1.5 bar in this example) during the pressure formation period... L(Or remain below it) and then drop again, then air L in the pressure line 6 is identified. The time until air L in the pressure line 6 is identified is taken as the matching control time t. a1L The first ventilation cycle is stored, wherein the matching control time t is used. a1L Air L is identified. If re-pressure formation is performed, the function is reset. (As in...) Figure 4 As can be seen, the matched control time t a1L —Based on the matched control time t a1L Identify the air L—the pressure change curve 110 until it reaches a pre-defined pressure level p. S The time is shorter and is particularly noticeable compared to the typically used fixed control time t. fest short.

[0037] exist Figure 5 In the common graph of pressure p, pressure change curve 120 obtained by the method according to the invention and pressure change curve 20 according to the prior art are shown with respect to time t. Thus, the two pressure change curves 20 and 120 and the control time t can be used to... a1e t fest Comparison. In the pressure change curve 20 according to the prior art, the first ventilation cycle 22, as usual, only occurs during a fixed operating time t. fest Then, the fixed control time is 60 seconds in this example. However, in the pressure change curve 120 according to the invention, the first ventilation cycle 122 is performed according to the first embodiment at the learned, matched control time t for the first ventilation cycle. a1e It has already begun. The learned, matched manipulation time t a1e Less than a fixed control time t fest Thus, according to the method according to the invention, compared with the conventional method, the first ventilation cycle 122 is performed at the learned matched control time t. a1e With a fixed control time t fest The time difference Δt1 occurs earlier.

[0038] Figure 6 The pressure change curve 130 is shown after the first ventilation cycle 132, which reaches a predetermined pressure level p 131. S In the third embodiment, a self-learning function is provided, which is a multi-volume dispensing function—in which a ventilation cycle is performed to exhaust gas from the system—learning the pressure change curve 130 and a pre-given pressure level p for the current delivery and metering system. SThe condition reached is 101. The first ventilation cycle 132 is learned from the learned data and is associated with a pre-given pressure level p. S The optimal time between condition 131 is determined and used as the learned control time t for the second ventilation cycle. a2e The information is then stored. Here, the learning is performed using a neural network, preferably the same neural network.

[0039] A function can also be used for the second ventilation cycle, which identifies the air L in the pressure line in order to obtain a matched control time. A fourth embodiment using this function is not shown here. Here, after the first ventilation cycle, compared with the second embodiment (see...),... Figure 2 Similarly, air L was identified.

[0040] exist Figure 7 In the common graph of pressure p, pressure change curve 140 obtained by the method according to the invention and pressure change curve 40 according to the prior art are shown with respect to time t. Thus, the two pressure change curves 40 and 140 and the control time t can be used to... a2e t fest Comparison. In the pressure change curve according to existing technology, the second ventilation cycle 43, as usual, only occurs during a fixed operating time t. fest This then occurs immediately after the first ventilation cycle 42, wherein the fixed operation time t fest In this example, it is 60 seconds. However, in the pressure change curve 140 according to the invention, the second ventilation cycle 143, according to the third embodiment, has already been trained and matched for the second ventilation cycle at a controlled time t. a2e This then proceeds immediately after the first ventilation cycle 142. The learned, matched control time t is used for the second ventilation cycle 143. a2e Less than a fixed control time t fest This allows the second ventilation cycle 143 to operate according to the method according to the invention, compared to conventional methods, within the learned, matched control time t. a2e With a fixed control time t fest The time difference Δ between t2 To do it earlier.

[0041] exist Figure 8 The pressure p graph shows a pressure change curve 150 relative to time t according to an embodiment of the method according to the invention. Furthermore, multiple ventilation cycles 152 to 156 are shown. The first ventilation cycle 152 is performed according to the method at a matched operating time t obtained according to the first or second embodiment. a1It was then triggered. After executing the first ventilation cycle 152, within the matched control time... ta2 This then triggers the second ventilation cycle 153. The control time t for the second ventilation cycle 153 is... a2 This can be obtained according to the third or fourth embodiment. After the second ventilation cycle 153, another ventilation cycle 154-156 can be performed. After the previous ventilation cycles 153-155, another matched control time t can be obtained similarly to the third or fourth embodiment. a3 To t a5 Matched control time t a1 To t a5 They are matched to the corresponding ventilation cycles and are therefore different from each other. Furthermore, a pre-defined pressure level p is plotted. S The system should be operated at this pre-given pressure level to ensure accurate and demand-compliant dosing. In this example, the pre-given pressure level p S It is 7.5 bar.

[0042] As a result, the matched manipulation time t is obtained. a1 To t a5 The control time is specifically matched to the current conveying and metering system. Therefore, a relatively small total time t is obtained for each conveying and metering system, and thus for each vehicle type or even each vehicle. a_g Until the pressure p reaches a predetermined pressure level p S The acquired manipulation time t a1 -t a5 and the total time t obtained a_g It is then sent to an online computer network (cloud) 200 and stored in a database there. The self-learning function can be updated using this database.

[0043] The pressure p does not reach the predetermined pressure level p after a predetermined number of ventilation cycles—for example, after six ventilation cycles. S In cases such as when an error is transmitted to the vehicle's driver and the internal combustion engine's power is throttled, the time until the error is detected is also relatively short.

Claims

1. A method for operating a conveying and metering system, the conveying and metering system having a conveying module (2), a metering module (3), a pressure line (6) connecting the conveying module (2) and the metering module (3), and a return section (7) of the conveying module (2). in, If during the manipulation time (t) a1e t a2e t a1L t a1 -t a5 During the subsequent pressure formation period, the pressure (p) in the pressure line (6) did not reach the predetermined pressure level (p). S Then, the exhaust of the pressure side portion of the system is carried out through at least one ventilation cycle (122, 142, 152, 153, 154, 155, 156), in which the metering module (3) is turned on. The characteristic is that the control time (t) a1e t a2e t a1 -t a5 The optimal control time for "normal state" is obtained through an empirical function, which is a self-learning function that finds the optimal control time based on the pressure change curve during dispensing. This optimal control time corresponds to the time until the pressure in the corresponding delivery and metering system reaches a predetermined pressure without air in the pressure line and without blockage in the return section.

2. The method according to claim 1, wherein, If after the ventilation cycle (142, 152, 153, 154) ends, at another control time (t) a5 During the subsequent period of pressure formation, the pressure (p) did not reach the predetermined pressure level (p). S If this occurs, then the pressure side portion of the system is vented through at least one additional ventilation cycle (143, 155). Characterized by, the other control time (t) a2e t a5 This is obtained through empirical functions.

3. The method according to claim 1 or 2, characterized in that, If the air (L) trapped in the pressure side portion of the system is identified by the function (119), ventilation circulation is performed (152, 153, 154, 155, 156).

4. The method according to claim 1 or 2, characterized in that, Until a predetermined pressure level (p) is reached S ) time (t) a_g ) and the acquired manipulation time (t) a1e t a2e t a1L t a1 -t a5 The data is stored in the database, and the empirical function retrieves the stored data and is used to obtain the current manipulation time (t). a1e t a2e t a1L t a1 -t a5 ).

5. A computer program product configured to perform each step of the method according to any one of claims 1 to 4.

6. A machine-readable storage medium having a computer program stored thereon, the computer program being configured to perform each step of the method according to any one of claims 1 to 4.

7. An electronic controller configured to operate a conveying and metering system by means of any one of claims 1 to 4.

Images