Method and device for vehicle diagnostics

Abstract

Vehicle diagnostic procedure, in particular for detecting a delayed response behavior of an internal combustion engine (8), wherein the procedure comprises the steps: Measuring at least one input parameter (EGR, n) mot , q inj , POS soll ) and at least one output parameter (p 2,mess ) with at least one measuring device (10, 12, 14, 16, 18), wherein the parameters (EGR, n mot , q inj , POS soll , P 2,mess ) in particular parameters of an internal combustion engine (8) are; Calculate at least one expected value (P) 2,Observ ) for at least one output parameter (p 2,mess ) from at least one input parameter (EGR, n mot , q inj , POS soll ) by an expected value calculation device (3); Calculating a measure of agreement (G) between the at least one measured output parameter (p) 2,mess) and the output parameter for this (P 2,mess ) calculated expected value (P 2,Observ ) by a conformity measurement calculation device (4a); Comparing the measure of agreement (G) with a first limit value (G1) stored in a storage device (4b) by a comparison device (4c); and Output an error message by an output device (5) if the degree of conformity (G) is less than the stored first limit value (G1).

Description

[0001] The invention relates to a method and a device for vehicle diagnostics, in particular for detecting delayed response behavior by predicting expected measured variables. State of the art

[0002] Modern engine control units must include functions that allow for the detection of malfunctions in individual components or assemblies and, for example, enable appropriate diagnosis if the impact on exhaust emissions is too severe. The goal is to reliably identify defective systems without incorrectly classifying intact systems as faulty. This is achieved using methods that, for example, detect electrical defects or trace persistent control deviations back to specific phenomena. Difficulties arise when actuators experience mechanical degradation (friction, sticking, play, etc.) or when changes occur within the overall system that lead to a delayed response. Such defects must also be reliably detected if they affect exhaust emissions.

[0003] DE 10 2011 109 307 A1 discloses a system and method for testing a control module, wherein a test module receives a throttle position signal and, based on this and a user input, generates a test signal in SENT format. This test signal is output to the control module of a vehicle to test and verify its response to various fault conditions.

[0004] DE 10 2009 018 152 A1 describes an electronic control system for vehicles that stores diagnostic trouble codes (DTCs) in non-volatile memory (EEPROM). The DTCs are stored in either a first or a second memory area, depending on whether a specific condition is met that indicates the completion of vehicle assembly and the start of use by the end customer. Technical task

[0005] The object of the invention is to provide an improved method and an improved device for vehicle diagnostics that make it possible to reliably detect the malfunction of individual components or assemblies, in particular a delayed response behavior of components of an internal combustion engine, without falsely classifying intact systems as defective. Disclosure of the invention

[0006] A fundamental concept of the invention is to perform a comparison between a measured quantity and a prediction of this quantity, calculated under the assumption that the system is intact. Such a comparison makes it possible to make a statement about the overall system's condition. The prediction is based on the simplest possible model, which provides a prediction of the quantity in question with sufficient accuracy to reliably distinguish between defective and intact systems.

[0007] A vehicle diagnostic method according to the invention, in particular for detecting a delayed response behavior of an internal combustion engine, comprises the following steps: Measuring at least one input parameter and at least one output parameter with at least one measuring device, wherein the parameters are in particular parameters of an internal combustion engine; Calculating at least one expected value for the at least one output parameter from the at least one input parameter using an expected value calculation device; Calculating a measure of agreement between the at least one measured output parameter and the expected value calculated for that output parameter using an agreement measure calculation device; Comparing the degree of conformity with a first limit value stored in a storage device by a comparison device; and Output an error message via an output device if the degree of conformity is less than the stored first limit value.

[0008] A device according to the invention for vehicle diagnostics, in particular for detecting a delayed response behavior of an internal combustion engine, has: at least one first measuring device, which is designed to measure at least one input parameter, at least a second measuring device designed to measure at least one output parameter, an expected value calculation device configured to calculate an expected value for the at least one output parameter from at least one input parameter; a conformity measure calculation device designed to calculate a measure of conformity between the at least one measured output parameter and the expected value calculated for that output parameter; a comparison device configured to compare the measure of conformity calculated by the conformity measure calculation device with a first limit value stored in a storage device; and an output device configured to issue an error message when the measure of conformity calculated by the conformity calculation device is less than the first limit value stored in the storage device.

[0009] Calculating the measure of agreement may include calculating the cross-correlation between the expected value and the measured value and / or the squared errors of both quantities.

[0010] In one embodiment, the output parameter is the boost pressure of an internal combustion engine, and the input parameters include one or more of the following parameters: exhaust gas recirculation rate, engine speed, injection quantity, target position of the boost pressure regulator, and throttle valve position. In this way, a delayed boost pressure response can be reliably detected by comparing the calculated output parameter "boost pressure" with the actual measured boost pressure.

[0011] In this system, a number of independent input variables, such as the exhaust gas recirculation rate, engine speed, injection quantity, the target position of the boost pressure regulator, and / or the throttle valve position, determine the behavior of an output variable under investigation (e.g., boost pressure). This is therefore a "multiple input - single output" system ("MISO system"), which, as a first approximation, can be assumed to exhibit linear transfer characteristics within the considered operating range. In this case, a transfer function for predicting the expected output variable can be represented by a linear superposition of several transfer sub-functions.

[0012] In one embodiment, calculating the at least one expected value involves using at least one linear transfer function for the calculation, wherein the linear transfer function is, in particular, a linear superposition of several linear transfer subfunctions. Each of the several linear transfer subfunctions is assigned to one of the input parameters, such that the input parameter is an input variable of the transfer subfunction assigned to it. Linear transfer (sub)functions are easy to implement and require only minimal computational effort. Within a limited operating range around a given operating point, linear transfer (sub)functions are sufficient to predict the output parameter with sufficient accuracy. In particular, the linear transfer (sub)functions can each comprise a first-order timing element with a gain factor and a time constant.

[0013] In one embodiment, the method includes calculating and storing the difference between the calculated expected value and the corresponding measured output parameter at a first time point in time and comparing it with the difference between a calculated expected value and the corresponding measured output parameter at a second, later time point in order to detect a fault of the vehicle, in particular a delayed response behavior of the internal combustion engine, on the basis of a change in the difference between the at least one measured output parameter and the at least one calculated expected value.

[0014] In this way, the reliability of the diagnosis can be increased, since the analysis is not based on absolute values ​​or their difference, but on the temporal development of the deviation between the measured value and the predicted value.

[0015] The procedure may include first calculating a first difference between the calculated expected value and a measured output parameter in a difference calculation device at a first time point in time and storing the calculated first difference in a difference value storage device, and then adding the first difference stored in the difference value storage device at the first time point to a second expected value calculated from input parameters measured at a second, later time point in an addition device in order to generate an adapted expected value for the second time point in time.

[0016] The agreement measure calculation device then calculates an adapted measure of agreement as a measure of the agreement between the output parameter measured at the second time point and the adapted expected value; the adapted measure of agreement is compared by the comparison device with a second limit value stored in the storage device, and an error message is issued by the output device if the adapted measure of agreement is less than the stored second limit value.

[0017] The method is particularly effective when the time interval between the first and second points in time is between 5s and 20s, especially 10s.

[0018] Alternatively or additionally, the method may include calculating, by means of the difference calculation device, at least a second difference between a second expected value, calculated from input parameters measured at the second time point, and an output parameter measured at the second time point; calculating a temporal change of the difference from the first difference and the at least one second difference; comparing the absolute value of the temporal change of the difference calculated in this way (the so-called adaptation range) in the comparison device with a third limit value stored in the storage device; and outputting an error message via the output device if the absolute value of the temporal change of the difference calculated in this way is greater than the third limit value.

[0019] The invention has the following advantages in particular: The method can easily be integrated into the software of a vehicle control unit, as the calculations to be carried out are not very complex and therefore do not require high computing power.

[0020] The individual transfer functions can be simple because the relevant operating range, defined by the parameters of the test procedure, is limited, and the model only needs to reliably predict the output within this range. Therefore, parameterizing such a function is quick and easy. Not only absolute values ​​but also their time-dependent behavior can be considered. This information enables the application of additional comparison methods.

[0021] The method according to the invention can be extended by further input parameters without the existing relationships losing their validity.

[0022] The method is applicable to other signals from a system, and in particular is not limited to the delayed response behavior of internal combustion engines. Brief description of the characters: Fig. Figure 1 illustrates in a schematic representation a first embodiment of a method according to the invention for vehicle diagnostics based on the example of the boost pressure of an internal combustion engine. Fig. Figure 2 shows a schematic representation of a second embodiment, which additionally includes an adaptation of the predicted value for the boost pressure. Character description

[0023] Fig. Figure 1 illustrates in a schematic representation a first embodiment of a method according to the invention for vehicle diagnostics using the example of the boost pressure p2 of an internal combustion engine 8.

[0024] Several first measuring devices 10, 12, 14, 16, 18 are arranged on an internal combustion engine 8, each configured to measure an input parameter EGR, n mot , q inj , pos soll of the internal combustion engine 8. A second measuring device 20 arranged on the internal combustion engine 8 is configured to measure an output parameter p. 2,mess to eat.

[0025] In the Fig. In the exemplary embodiment shown in 1, the input parameters include the exhaust gas recirculation rate (EGR) and the engine speed n. mot , the injection quantity q inj , the target position pos soll of the boost pressure regulator and the setting of a throttle valve located in the intake manifold of the internal combustion engine 8 (not shown), and the output parameter is the boost pressure p 2,mess .

[0026] The input parameters EGR, n measured by the measuring devices 10-18 mot , q inj , pos sollare input variables of an expected value calculation device 3 and in particular each of the respective input parameter EGR, n mot , q inj , pos soll assigned transfer subfunction G11, G12, G13, G14, G15.

[0027] The transfer sub-functions G11, G12, G13, G14, G15 implemented in the expectation value calculation device 3 can in particular be considered linear transfer sub-functions, for example as first-order timing elements. y(x,t)=a*x*e−kt with a gain factor a and a time constant k, as proportional terms and / or more complex transfer functions, which for example include a dead time T tot include, be implemented.

[0028] Since the diagnosis is usually only carried out in a limited operating range of the motor 8, which is often specified by a standardized test procedure, and the diagnosis only needs to provide a reliable prediction within this limited operating range, it is possible to represent each of the transfer sub-functions G11, G12, G13, G14, G15 at least approximately by an easy-to-calculate first-order timing element.

[0029] The output values ​​of the transfer functions G11, G12, G13, G14, G15 are fed to a superposition device 2, which performs a linear superposition of the output values ​​of the transfer functions G11, G12, G13, G14, G15 and uses the output value p. 2,Obsvr an expected value of the boost pressure P 2,Obsr provides.

[0030] The expected value of the boost pressure p 2,Obsvris fed to a conformity measurement calculation device 4a. The conformity measurement calculation device 4a determines a measure of conformity G of the calculated expected value of the boost pressure p. 2,Obsvr with the output parameter p measured by the second measuring device 20 2,mess , in this case the boost pressure p 2,mess , which occurs simultaneously with the input parameters EGR, n mot , q inj , pos soll has been measured.

[0031] Possible calculation and comparison algorithms that can be used to calculate the measure of agreement G include the calculation of the cross-correlation between the predicted value p. 2,Obsvr and the measured value p 2,mess and / or the calculation of the squared errors of these two quantities.

[0032] The measure of conformity G calculated by the conformity measure calculation device 4a is compared by a comparison device 4c with a predetermined first limit value G1, which is stored in a storage device 4b. If the measure of conformity G calculated by the conformity measure calculation device 4a falls below the predetermined first limit value G1, an error message is displayed on a display device 5 to inform the user that a malfunction, for example a delayed response behavior of the internal combustion engine 8, has been detected.

[0033] The parameters of the transfer functions G11, G12, G13, G14, G15, for example the gain factors a and the time constants k of first-order timing elements, can be determined by measurements on an intact internal combustion engine 8, whereby in particular a so-called path identification can be used to determine the parameters of the transfer functions from the measured values.

[0034] For vehicle diagnostics, especially to detect delayed response behavior, the absolute value of the agreement between the predicted value p is less important. 2,Obsvr and the measured value p 2,mess , rather than the temporal progression of the agreement.

[0035] Fig. Figure 2 shows a second embodiment, which builds upon the first, in the Fig.The embodiment shown in Figure 1 is based on an additional adaptation device 6, which includes a difference calculation device 6a and an addition device 6b and are configured to calculate a static, i.e., time-constant, difference between the predicted value p. 2,Obsvr and the measured value p 2,mess to compensate, so that the diagnosis is based solely on a temporal change in the difference between the predicted value p 2,Obsvr and the measured value p 2,mess is carried out.

[0036] For this purpose, the difference calculation device 6a is designed to calculate the predicted value p at a first time t1. 2,obsvr (t1) with the value p measured at this first time point 2,mess (t4) to compare and find a difference between the predicted value p 2,obsvr (t1) with the measured value P 2,mess (t1) to determine.

[0037] This difference D(t1), determined at the first time t1, is stored in a difference value storage device 6c and calculated by an addition device 6b to an expected value p for a second, later time t2. 2,obsvr (t2) added to obtain an adapted expected value p for the second time point 2,dapt (t2) to generate, which then with the output parameter p measured at the second, later time 2,mess (t2) is compared.

[0038] Since a delayed response behavior detectable during vehicle diagnostics can usually be observed on a time scale of a few seconds, especially in the range of 1 to 30 seconds, the value p predicted for a second time point t2 is 2,Obsvr (t2) for example adapted with a value D(t2) which is derived from the difference between the predicted value p 2,Obsvr and the measured value p 2,messto an earlier time, e.g. by Δt=t2-t1=10s earlier, results in: p2,dapt(t2)=p2,Obsvr(t2)+[p2,mess(t2−10s)−p2,Obsvr(t2−10s)] p2,adapt(t2)=p2,Obsvr(t2)+D(t2).

[0039] The degree of agreement is then determined between the expected value p adapted in this way. 2,adapt (t2) for the second time t2 and the value p measured at this second time t2 2,mess (t2) is determined. If the two values ​​differ from each other by less than a predetermined second limit value G2, the system is intact; if the values ​​differ from each other by more than the predetermined second limit value G2, a malfunction, e.g. a delayed response behavior, is detected and displayed on the display device 5.

[0040] The comparison between the adapted expected value P 2,adapt and the measured value p 2,messThe calculation (t2) can be performed at a single point in time, i.e., at the second time point t2, or averaged or integrated over a period that may include a few seconds before the second time point t2. Alternatively or additionally, the time course ΔD(t) of the difference between the calculated expected value p can be used. 2,Obsvr and the measured output parameter P 2,mess , e.g. by calculating the cross-correlation and / or the error squares, evaluated and included in the comparison.

[0041] The value of 10s for the adaptation time delay is only an example; any other value that seems suitable, and which usually covers a range of 1s to 30s, can also be used.

Claims

[1] Vehicle diagnostic procedure, in particular for detecting a delayed response behavior of an internal combustion engine (8), wherein the procedure comprises the steps: Measuring at least one input parameter (EGR, n) mot , q inj , POS soll ) and at least one output parameter (p 2,mess ) with at least one measuring device (10, 12, 14, 16, 18), wherein the parameters (EGR, n mot , q inj , POS soll , P 2,mess ) in particular parameters of an internal combustion engine (8) are; Calculate at least one expected value (P) 2,Observ ) for at least one output parameter (p 2,mess ) from at least one input parameter (EGR, n mot , q inj , POS soll ) by an expected value calculation device (3); Calculating a measure of agreement (G) between the at least one measured output parameter (p) 2,mess) and the output parameter for this output parameter (P 2,mess ) calculated expected value (P 2,Observ ) by a conformity measurement calculation device (4a); Comparing the measure of agreement (G) with a first limit value (G1) stored in a storage device (4b) by a comparison device (4c); and Output an error message by an output device (5) if the degree of conformity (G) is less than the stored first limit value (G1). [2] Method according to claim 1, wherein calculating the measure of agreement includes a cross-correlation between the expected value (p 2,Obsvr ) and the measured value (P 2,mess ) and / or to calculate the squared errors of both quantities. [3] Method according to claim 1 or 2, wherein the output parameter is the boost pressure (p2) of an internal combustion engine (8) and wherein the input parameters (EGR, n mot , q inj , POS soll) one or more of the parameters exhaust gas recirculation rate (EGR), engine speed, (n mot ), injection quantity (q inj ), target position (pos soll ) of the boost pressure regulator and position (DK) of a throttle valve. [4] Method according to one of the preceding claims, wherein the calculation of at least one expected value (P 2,Observ ) by the expectation value calculation device (3) includes using at least one linear transfer function, wherein the linear transfer function is in particular a linear superposition (2) of several linear transfer sub-functions (G11, G12, G13, G14, G15), and wherein each of the several linear transfer sub-functions (G11, G12, G13, G14, G15) corresponds to one of the input parameters (EGR, n) mot , q inj , POS soll ) is assigned. [5] Method according to claim 3 or 4, wherein at least one linear transfer function (G11, G12, G13, G14, G15) comprises a first-order timing element with a gain factor and a time constant. [6] Method according to one of the preceding claims, wherein the method includes calculating in a difference calculation device (6a) at a first time (t1) a first difference (D(t1)) between the calculated expected value (p 2,Observ ) and a measured output parameter (p 2,Mess ) to calculate and store in a differential value storage device (6c). [7] The method of claim 6, wherein the method additionally includes, in an addition device (6b) the first difference (D(t1)) calculated at the first time (t1) and stored in the difference value storage device (6c) to a second expected value (p 2,Observ(t2)), which is derived from input parameters (EGR, n) measured at a second, later time (t2). mot , q inj , POS soll ) has been calculated, to add to obtain an adapted expected value (p 2,adapt (t2)) to generate for the second time point (t2); by the conformity measure calculation device (4a) an adapted measure of conformity (aG) as a measure of conformity between the output parameter (p) measured at the second time (t2) 2,Mess (t2)) and the adapted expected value (p 2,adapt (t2)) to calculate for the second time point (t2); to compare the adapted measure of agreement (aG) by the comparison device (4c) with a second limit value (G2) stored in the storage device (4b); and to output an error message via the output device (5) if the adapted measure of conformity (aG) is less than the stored second limit value (G2). [8] The method of claim 6 or 7, wherein the method further comprises, by the difference calculation device (6a) at least a second difference (D(t2)) between a second expected value (p 2,Observ (t2)), the input parameters (EGR, n) measured at the second time point (t2). mot , q inj , pos soll ) has been calculated, and an output parameter (P) measured at the second time point (t2). 2,Mess (t2)) to calculate; to calculate a time change of the difference (ΔD(t)) from the first difference (D(t1)) and the at least one second difference (D(t2)); to compare the absolute value of the time-dependent change of the difference (ΔD(t)) calculated in this way in the comparison device (4c) with a third limit value (G3) stored in a storage device (4b); and to output an error message via the output device (5) if the absolute value of the time change of the difference (ΔD(t)) calculated in this way is greater than the third limit (G3). [9] Method according to claim 7 or 8, wherein the time interval (Δt) between the first time (t1) and the second time (t2) is 5s to 20s, in particular 10s. [10] Device for vehicle diagnostics, in particular for detecting a delayed response behavior of an internal combustion engine (8), with at least one first measuring device (10, 12, 14, 16, 18) for measuring at least one input parameter (EGR, n) mot , q inj , POS soll ) is trained, at least a second measuring device (20) for measuring at least one output parameter (p 2,Mess ) is trained, an expected value calculation device (3) which is configured to consist of at least one input parameter (EGR, n)mot , q inj , POS soll ) an expected value (P 2,Observ ) for at least one output parameter (P 2,Mess to calculate; a conformity measure calculation device (4a) for calculating a measure of conformity (G) between the at least one measured output parameter (p 2,Mess ) and the output parameter for this output parameter (p 2,Mess ) calculated expected value (P 2,Observ ) is trained; a comparison device (4c) configured to compare the measure of conformity (G) calculated by the conformity measure calculation device (4a) with a first limit value (G1) stored in a storage device (4b); and an output device (5) configured to output an error message when the measure of conformity (G) calculated by the measure of conformity calculation device (4a) is less than the first limit value (G1) stored in the storage device (4b). [11] Device according to claim 10, wherein the agreement measure calculation device (4a) is configured to perform a cross-correlation between the predicted value (P 2,Obsrv ) and the measured value (p 2,mess ) and / or the squared errors of these two quantities. [12] Device according to claim 11, wherein the expected value calculation device (3) is configured to calculate at least one expected value (P 2,Observ) to calculate using a linear transfer function, wherein the linear transfer function is in particular a linear superposition (2) of several linear transfer subfunctions (G11, G12, G13, G14, G15) and wherein each of the several linear transfer subfunctions (G11, G12, G13, G14, G15) takes one of the input parameters (EGR, n) mot , q inj , POS soll ) as an input variable. [13] Device according to one of claims 11 or 12, wherein the device additionally comprises a difference calculation device (6a) configured to calculate the difference (D(t1)) between the calculated expected value (p 2,Observ (t1)) and the measured output parameter (p 2,Mess (t1)) to calculate at a first time point (t1) and to store in a differential value storage device (6c). [14] Device according to claim 13, which additionally comprises an addition device (6b) configured to add the difference (D(t1)) calculated at the first time (t1) and stored in the difference value storage device (6c) to a second expected value (p 2,Observ (t2)), which is derived from input parameters (EGR, n) measured at a second, later time (t2). mot , q inj , POS soll ) has been calculated, to add to obtain an adapted expected value (p 2,adapt (t2)) to calculate for the second time point (t2); where the conformity measure calculation device (4a) is configured to provide an adapted measure of conformity (aG) as a measure of conformity between the output parameter (p) measured at the second time (t2). 2,Mess (t2)) and the adapted expected value (p 2,adapt (t2)) to calculate for the second time point (t2); the comparison device (4c) is configured to compare the adapted measure of conformity (aG) with a second limit value (G2) stored in the storage device (4b); and the output device (5) is configured to output an error message if the adapted measure of conformity (aG) is less than the second limit value (G2). [15] Device according to claim 13 or 14, wherein the difference calculation device (6a) is additionally designed, a second difference (D(t2)) between a second calculated expected value (p 2,observ (t2)), which is derived from input parameters (EGR, n) measured at a second, later time (t2). mot , q inj , POS soll ) has been calculated, and an output parameter (P) measured at the second, later time (t2). 2,Mess (t2)) to calculate; to calculate a time change of the difference (ΔD(t)) from the first difference (D(t1)) and the second difference (D(t2)); to compare the absolute value of the time-dependent change of the difference (ΔD(t)) calculated in this way in the comparison device (4c) with a third limit value (G3) stored in the storage device (4b); and the output device (5) is designed to output an error message if the time change of the difference (ΔD(t)) is greater than the third limit (G3).

Images