Method for determining deviations in quantity in the case of a fluidic metering system

Abstract

A method for identifying deviations in quantity in the case of a fluidic metering system (100-165), in particular an internal combustion engine of a motor vehicle, in which at least one conveying pump (125) for conveying a fluid, and at least one pressure sensor (135) for determining a fluidic pressure in the metering system (100-165), are disposed, wherein it is provided in particular that a test admeasurement of fluid is carried out (205), that a temporal pressure drop in the metering system (100-165) is detected (210), that the detected temporal pressure drop is compared with a pressure drop (215) that is to be theoretically expected (220), and that a deviation in quantity of the metering system (100-165) is determined based on the result of the comparison (225).

Claims

1. A method for identifying deviations in quantity of a fluid in a fluidic metering system, the method comprising: providing at least one conveying pump for conveying a fluid, providing at least one pressure sensor for determining a fluidic pressure in the fluidic metering system, carrying out a test admeasurement of the fluid including, detecting a temporal pressure drop in the metering system, comparing the detected temporal pressure drop to a pressure drop that is to be theoretically expected (220), and determining a deviation in quantity of the metering system based on the result of the comparison (225), wherein as the result of the comparison, an additional conclusion is drawn as to which system parameters of the metering system have the strongest influence on the result of the test admeasurement.

2. The method according to claim 1, wherein the pressure drop that is to be theoretically expected is computed using an analytical approximate solution which corresponds to a pressure drop that theoretically results in the case of a test admeasurement.

3. The method according to claim 1, wherein in an analytical approximate solution, the following equation is used as a basis, the equation describing a fluidic pressure drop in the metering system (100-165) after the test admeasurement (205) performed at t=0 with the duration t:
p(t)p.sub.1(A.sub.V*t)/(2*k)+(A.sub.V*A.sub.D*t.sup.2)/(4*p.sub.1*k.sup.2)+ . . . , wherein p.sub.1 is an initial pressure prior to the test admeasurement (205), A.sub.v is a cross-sectional area of an injector that carries out the test admeasurement (205), A.sub.D is a cross-sectional area of a return flow throttle and pump leakage area, k is a compressibility of the metering system (100-165), and p is a density of the fluid.

4. A non-transitory, computer-readable storage medium storing instructions for carrying out each step of the method according to claim 1.

5. An electronic control apparatus which is specified for controlling a fluidic metering system in which at least one conveying pump for conveying a fluid, and at least one pressure sensor for determining a fluidic pressure in the metering system, are disposed, by means of a method according to claim 1.

6. The method according to claim 1, wherein the system parameters include at least one of an initial pressure prior to the test admeasurement, a cross-sectional area of an injector that carries out the test admeasurement, a cross-sectional area of a return flow throttle and pump leakage area, a compressibility of the metering system (100-165), and a density of the fluid.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) In the figures:

(2) FIG. 1 shows a block diagram of a UWS metering system of an SCR catalytic converter system according to the prior art;

(3) FIG. 2 shows a typical pressure profile in the metering of a fluid concerned here;

(4) FIG. 3 shows an exemplary embodiment of the method according to the invention by means of a flow diagram.

DETAILED DESCRIPTION

(5) The extended identification of deviations in quantity described hereunder for DNOX systems is capable of being applied in particular in the case of displacement pumps having an internal leakage and an external return flow, for example in the case of mentioned COR pumps.

(6) As is schematically illustrated in FIG. 1 by way of an example of an SCR catalytic converter system, metering systems equipped with conveying units and injectors are used in the exhaust gas post-treatment of internal combustion engines by means of metering AdBlue or UWS, respectively, into the exhaust gas flow. In many cases herein, a return flow into the AdBlue tank is also provided. In order for an operation of these systems that conforms to the exhaust gas legislation to be guaranteed, the quantity of AdBlue that is metered into the exhaust gas train has to be monitored.

(7) The fluidic metering system known per se of a motor vehicle, shown in FIG. 1, comprises a metering module 100 having an metering valve 105 that is presently controllable in a cycled manner. The UWS Floyd 115 that is stored in a tank 110 is supplied by means of a pump module 120, specifically by way of the discharge line 117, a mainline 118, and a supply line 119, to the metering module 100. The pump module 120 comprises a conveying pump 125 and a return conveying pump 130 which with a pressure sensor 135 arranged in the region of the mainline 118. The pressure sensor 135 by way of a signal or control line 140, respectively, is connected to a control apparatus 145. The return conveying pump 130 is connected to a return conveying line 131 that emanates from the main line 118, wherein excess fluid that is to be returned into the tank 110 is routed by way of a return flow line 132 that is connected to the tank 110, or opens into the tank 110, respectively.

(8) The conveying pump 125 which in the present exemplary embodiment is configured as a diaphragm pump, suctions the UWS fluid 115 from the tank 110 and compresses said UWS fluid to a system pressure of 4.5 to 8.5 bar that is required for the atomization. The metering module 100 admeasures the UWS quantity that is required for the NOx reduction and atomizes said UWS quantity into the exhaust gas flow (not shown), specifically ahead of the SCR catalytic converter. The controlling of the metering and heating strategy, and of an on-board diagnosis, is performed by a motor control apparatus (not shown) or by the control apparatus 145 shown in FIG. 1. By processing the current motor operating data and all required sensor data, the quantity of the reduction agent by means of the closed-loop control is tuned in a manner known per se precisely to the operating point of the internal combustion machine and to the catalytic converter specific properties, in order for nitrogen oxide to be converted to the maximum.

(9) It is to be noted that the mentioned metering valve 105 operating as a dosing means is actuated in a manner known per se by the control apparatus 145 by way of a control line 107.

(10) A sensor system 150 which by way of the signal or control line 155, respectively, is connected to the control apparatus 145 is disposed in the tank 110. The sensor system 150 in particular serves for reporting a low level of the fluid 115 that is optionally present to the control apparatus 145, in order for the vehicle driver to be able to optionally top up the fluid. The conveying pump 125 and the return conveying pump 130 are also connected to the control apparatus 145 by way of two further signal or control lines 160, 165, respectively, so as to actuate in a manner known per se the two pumps 125, 130 for the metering operation.

(11) In order for future metering systems to be homologated, deviations in quantity of 35% must be identified by means of a so-called Consumption Deviation Monitoring (CDM). Systems known per se have the potential of identifying these deviations in quantity only to a limited extent, and require the highly precise manufacturing of individual system components and/or the complicated computation of further variables (rigidity determination). In concrete terms, by way of the rigidity determination that is presently applied in a system having a displacement pump, the trade-off is a reliance of the result on the internal leakage of the pump. The reduction and delimitation of the internal leakage is not possible based on the current state of knowledge, since said leakage can vary intensely over the service life.

(12) It can be determined which system parameters have a major influence on the result of the test injection by means of the analytical approximate solution described hereunder of the pressure drop over the time of a test injection of fluid. A simple function that is universally applicable and has a higher accuracy can be used by way of the systematic combination of the analytical solution having existing and future fluidic or hydraulic components. This enables the reduction in production tolerances of the pump and of the return flow, and thus enables a reduction in costs.

(13) The following equation (1) describes in an analytical manner the fluidic or hydraulic pressure drop, respectively, for example in a UWS metering system after a metering test injection performed at t=0:
p(t)p.sub.1(A.sub.V*t)/(2*k)+(A.sub.V*A.sub.D*t.sup.2)/(4*p.sub.1*k.sup.2)+ . . .(1).

(14) According to the equation (1), the metering procedure results approximately in a series expansion in the opening/measuring duration t, specifically depending on the parameters

(15) initial pressure ahead of the opening: p.sub.1

(16) cross-sectional area of the injector (valve): A.sub.V

(17) cross-sectional area of the return throttle and the pump leakage area: A.sub.D

(18) system compressibility: k

(19) and AdBlue density: p.

(20) The qualitative correlation between the individual parameters in the equation (1) is illustrated in FIG. 2 by means of a typical pressure profile 300 in the metering of UWS solution by means of an injector. In a relatively short temporal window t of, for example. 0.2 s, at a pressure drop p.sub.1p(t) 305, proceeding from an initial value p.sub.1, an approximately linear pressure profile 310 results. In the case of an assumed increase in the system compressibility k, the curve 300 would be displaced to the top right (in the direction indicated by the upper arrow). In the case of an assumed enlargement of the cross-sectional area of the return throttle A.sub.D, the pressure curve 300 in the run-out region on the right would be displaced to the top (in the direction indicated by the lower arrow).

(21) In the determination of the deviation in quantity, the measured pressure drop is now compared with the pressure drop that is to be theoretically expected. In the metering systems known per se (DNOX2.2, DNOX6.x) the exact value of the system compressibility k has to be determined in order for the equation (1) to be able to be solved in the first place. However, this value is obtained only depending on the parameter A.sub.D. Therefore, a benefit in terms of accuracy would only be achieved in the case of a metering system in which A.sub.D is known to a sufficiently precise extent. However, in a metering system having a displacement pump, the variable A.sub.D also comprises the internal leakage of said pump. If the rigidity determination is dispensed with, and if the pressure drop to be expected is determined by way of a reference measurement, the variable A.sub.D appears only in the second order of the measuring duration.

(22) The present approach is based on the concept of being able to neglect the influence of the second series member or term, respectively, (return flow and pump leakage) in the equation (1) by way of a sufficiently short metering or measuring time. A further advantage of a relatively short metering time moreover lies in that inadvertent introduction of AdBlue into the exhaust gas train can be effectively prevented or at least reduced during the test injection.

(23) As an alternative to the shortening of the measuring time, in the case of the present approach for eliminating the second series member it can also be provided that the compressibility k of the system is increased, for example by means of the use of a pressure damper, of an air cushion, or by means of flexible lines.

(24) An exemplary embodiment of the method according to the invention is shown in FIG. 3. After the start 200 of the routine shown, a test injection is first carried out 205, and the pressure drop arising thereafter in the metering system is detected or measured, respectively, 210. On the basis of the starting or initial pressure p.sub.1, respectively, measured in the metering system shortly before the test injection, the pressure profile that is to be theoretically expected is computed, for example, by means of the equation (1). The pressure profile detected according to step 210 and the pressure profile computed according to step 215 are compared with one another in step 220, and in the case of a deviation in quantity being established, a value for the deviation in quantity is subsequently determined 225 by means of the resulting differences. If no deviation in quantity is established in the test step 220, the routine reverts back to the start, for example ahead of step 205, in order for the routine to be carried out once again. Alternatively, the routine can also first be terminated and subsequently be optionally restarted 200.

(25) An exemplary computation as to how an absolute value for the deviation in quantity can be determined from the differences mentioned, or as to how a comparison of the measured data can be performed by way of the analytical equation (1), respectively, is set forth hereunder. It is first determined by way of the curvature of the curve of the pressure profile herein whether a linear approximation according to the equation (1) is possible with the required accuracy. To this end, the quotient from the discreet second deviation of the square root of the pressure profile and from the gradient of the square root of the pressure profile is compared with a threshold value. If said threshold value is exceeded, a smaller temporal step has to be used for the temporal discretization of the pressure profile. As soon as a temporal step by way of which the threshold value is not exceeded is achieved, the linear approximation of the equation (1) is permissible and in the revised form is:
(p(t)p.sub.1)/tA.sub.V/(2*k).

(26) The measured gradient of the square root of the pressure profile is accordingly proportional to the cross-section of the metering valve, said cross-section being a measure of the metered quantity. An insufficient metering can thus be identified when the gradient deviates excessively from a previously measured reference gradient.
g.sub.rev=(p.sub.rev(t)p.sub.rev,1)/t.sub.rev

(27) The quotient from the measured gradient and from the reference gradient according to the equation above is equivalent to the following combined variable:
A.sub.V/A.sub.Vrev*(2.sub.rev*k.sub.rev)/(2*k).

(28) Assuming that p and k have not changed since a mentioned reference measurement, the result is reduced to the ratio A.sub.V/A.sub.vrev which represents the deviation in quantity.

(29) The method described can be implemented in the form of a control program for an electronic control apparatus for controlling an internal combustion engine, or in the form of one or a plurality of corresponding electronic control units (ECUs).