DE102020130375A1 - Method and test system for a waveform test - Google Patents

Abstract

The invention relates to a method for checking a signal curve (s), the signal curve (s) being specific to a time curve of a measured variable (201) in a technical system (1), the following steps being carried out: - providing an adaptation operator (c) for an approximation function (w) for approximating the signal curve (s),- detecting at least one measured value (202) of the signal curve (s) at at least one measurement time (tm),- determining (101) the adaptation operator (c) for the approximation of the approximation function (w) to the detected measured value (202), - carrying out a comparison (102) of the determined adaptation operator (c) with at least one threshold value (sw) in order to determine a comparison result (205) of the comparison (102), - Carrying out an evaluation (103) of the comparison result (205) in order to determine at least one state of the technical system (1).

Description

The present invention relates to a method and test system for testing a waveform.

From Scripture DE 10 2015 014 873 A1 a method for checking the tightness of a tank is known, in which an integral of a signal curve is determined.

In addition, other systems and methods are also known for checking a tank system for leaks. For this purpose, a desired target pressure is achieved, for example, by pumping air into the tank. A drop in pressure is then detected in the closed tank system. Depending on the presence and an evaluation of a pressure drop, leaks in the tank system can be inferred. With this evaluation of the pressure drop, however, it is first necessary to always produce the same target pressure in the tank system.

Traditional methods and testing systems often require achieving a target pressure and/or venting a tank to ambient pressure in order to perform a leak diagnosis.

Furthermore, similar problems are known in other types of systems in which a waveform must be evaluated.

It is therefore an object of the present invention to at least partially eliminate the disadvantages described above. In particular, it is the object of the present invention to propose an improved solution for checking a signal curve.

The above object is achieved by a method having the features of claim 1 and by a test system having the features of claim 10. Further features and details of the invention result from the respective dependent claims, the description and the drawings. Features and details that are described in connection with the method according to the invention naturally also apply in connection with the test system according to the invention, and vice versa, so that the disclosure of the individual aspects of the invention is or can always be referred to reciprocally.

The object is achieved in particular by a method for checking a signal curve, in particular a decaying signal curve such as a pressure drop.

It is provided in particular that the signal curve is specific to a time curve of a measured variable in a technical system, in particular after a drop in the signal curve, such as a drop in pressure, has been initiated and/or detected manually or automatically in the technical system.

It may be possible for the initial value of the waveform to remain unchanged prior to the initiation of the decay. For example, there is no need to establish a predetermined target pressure before initiating the pressure drop. The existing initial value of the signal curve is used for the test. If there is now a drop in the values of the signal curve, this can be detected as a drop in the signal curve.

• - Bereitstellen eines (insbesondere zeitkonstanten und somit nicht zeitabhängigen) Anpassungsoperators, insbesondere Anpassungsfaktors oder Anpassungstupels, für eine (insbesondere zeitabhängige) Annäherungsfunktion zur Annäherung an den Signalverlauf (d. h. auch zumindest an einen Teil des Signalverlaufs), bevorzugt in digitaler und/oder numerischer und/oder mathematischer Form, vorzugsweise zumindest teilweise durch eine Bereitstellungskomponente in der Form eines Datenspeichers, insbesondere durch das Speichern, vorzugsweise Zwischenspeichern, und/oder Auslesen des Anpassungsoperators und/oder einer Funktion zur Berechnung des Anpassungsoperators in digitaler Form durch eine Verarbeitungskomponente,
• - Erfassen wenigstens oder genau eines Messwertes (oder wenigstens zwei oder wenigstens drei Messwerte) des Signalverlaufs (jeweils) zu wenigstens einem Messzeitpunkt, also insbesondere auch zu mehreren Messzeitpunkten, vorzugsweise durch eine Erfassungskomponente in der Form einer Messvorrichtung, wie eines Drucksensors, welche in das technische System integriert ist,
• - Ermitteln (d. h. Bestimmen, z. B. numerisches Bestimmen) des Anpassungsoperators für die Annäherung der Annäherungsfunktion an den (wenigstens einen oder den zeitlichen Verlauf der) erfassten Messwert(e), bevorzugt mittels einer numerischen Approximation, insbesondere durch die Verarbeitungskomponente, bevorzugt durch die Verwendung der gespeicherten Funktion zur Berechnung des Anpassungsoperators, wobei hierzu vorzugsweise ein Vergleich des (wenigstens oder genau einen) erfassten Messwertes mit der Annäherungsfunktion anhand des bereitgestellten Anpassungsoperators erfolgt,

• - Durchführen eines Vergleichs des ermittelten Anpassungsoperators mit wenigstens einem Schwellenwert, um ein Vergleichsergebnis (des Vergleichs) zu bestimmen,
• - Durchführen einer Auswertung des Vergleichsergebnisses, um wenigstens einen Zustand des technischen Systems zu bestimmen.

The following steps can then be carried out for the test, preferably one after the other in the specified order or in any order, with individual or all steps also being able to be carried out repeatedly, with the steps also being able to be carried out for each measurement time:

• - Providing a (particularly time-constant and therefore not time-dependent) adjustment operator, in particular adjustment factor or adjustment tuple, for a (particularly time-dependent) approximation function for approximating the signal curve (ie also at least part of the signal curve), preferably in digital and/or numerical and/or or mathematical form, preferably at least partially by a provision component in the form of a data memory, in particular by storing, preferably temporarily storing, and/or reading out the adaptation operator and/or a function for calculating the adaptation operator in digital form by a processing component,
• - Detection of at least or exactly one measured value (or at least two or at least three measured values) of the signal curve (each) at least at one measurement time, i.e. in particular also at several measurement times, preferably by a detection component in the form of a measuring device, such as a pressure sensor, which is integrated into the technical system is integrated,
• - Determining (ie determining, e.g. numerically determining) the adaptation operator for the approximation of the approximation function to the (at least one or the temporal course of) measured value(s), preferably by means of a numerical approximation, in particular by the processing component, preferably by the use of the saved cherten function for calculating the adjustment operator, for which purpose a comparison of the (at least or exactly one) measured value recorded with the approximation function is preferably carried out using the adjustment operator provided,

• - Carrying out a comparison of the adjustment operator determined with at least one threshold value in order to determine a comparison result (of the comparison),
• - Carrying out an evaluation of the comparison result in order to determine at least one state of the technical system.

This has the advantage that the use of the adjustment operator for the test means that there is no need to create a fixed initial value for the signal curve. Rather, the procedure described can be applied flexibly to different forms of a signal curve.

The signal profile can include a time profile of a signal. In this case, a signal is understood to mean, in particular, a measurement signal, that is to say a recorded value of the measurement variable at a specific measurement time.

The adjustment operator can be provided, for example, by the adjustment operator being calculated using a predefined algorithm (ie a function), in particular using the recorded measured value as a parameter for the algorithm. Furthermore, the algorithm and/or also the approximation function can be stored as a mathematical function and/or an algorithm in a data memory of the system according to the invention. The adjustment operator can be defined as a factor or offset of the approximation function, with the approximation function being physically predefined as a function of the technical system, and (by determining the adjustment operator) being adjusted to the recorded measured values when the method steps are carried out.

The determination of the adjustment operator is used in particular to approximate the approximation function to the signal profile in that the approximation takes place at least or only for the at least one recorded measured value. In this case, the approximation can be limited to the at least one previously recorded measured value, ie to a short section of the signal curve. In other words, it is not necessary (but possible) for a further part of the signal profile or a longer period before and/or after the at least or precisely one measurement time to be taken into account for this purpose. If, for example, at least or exactly two or three measured values were recorded, and therefore measured values were recorded at different measurement times, the adjustment operator can be determined in the form of an approximation using this short time profile of the recorded measured values. Furthermore, the determination can be continued for further measurement times as soon as further measurement values have been recorded.

The steps of a method according to the invention can be carried out repeatedly, so that successive measured values can be recorded and used for the method steps. For example, for each iteration, the last at least two or three most recently acquired measured values are used to determine the adaptation operator in order to carry out the approximation of the approximation function thereto. Furthermore, the determination and/or the implementation of the comparison and/or the evaluation can take place on the basis of the currently recorded, chronologically consecutive measured values. Furthermore, the measured variable can also be continuously monitored, e.g. B. by repeated recording and evaluation of the measured values in order to determine the status of the system. For example, a drop in pressure or the like can be detected during the evaluation based on an anomaly such as a kink in the signal curve, in order to then carry out the check for a leak by executing the method steps according to the invention.

Measurements can be taken for the entire waveform, or for just a portion of the waveform. For example, it is conceivable that measured values are not recorded at the beginning of the signal curve and/or at the end of the signal curve and are therefore not taken into account. It is thus possible to hide predefined time ranges of the signal profile for carrying out the method steps. If the signal curve represents a drop in pressure, this may be in areas where it is not possible to reliably detect a leak based on the measured values. To suspend the collection of readings for these areas, a time registration can be performed. The recorded time can then be compared to a predefined masking template to determine whether measurements should be recorded or whether the recording and/or the determination and/or the comparison and/or the evaluation should be suspended. The masking template thus defines the masking areas of the signal curve.

Provision is advantageously made for the approximation function to be implemented as a time-dependent function which is dependent (in particular only) on the parameters time and the adjustment operator. In other words, the approximation function can be implemented as a function that depends on the adjustment operator as a parameter of the function - and otherwise only on the "time" parameter. This applies, for example, to an exponential function that is multiplied by the adjustment operator (as a factor) and to which the measurement time is passed as a function parameter. If the approximation function is embodied as a polynomial, the adjustment operator can also be embodied as a parameter tuple according to a possible embodiment. The adjustment operator can be implemented as a time-constant parameter of the function and/or as a parameter of the function. If the waveform is embodied as a descending waveform, the fitting operator may be specific to, in particular, characterize the decay rate of the waveform.

It is conceivable that the approximation function is a function dependent on the measurement time, e.g. B. in the form of an exponential function and / or a logarithm or a polynomial, and advantageously the adjustment operator is designed as an adjustment factor of the function or exponential function. Specifically, the approximation function can be defined by an exponential function (e-function) or a logarithm, which is calculated, for example, using the measurement time as a function parameter, and is then adjusted and, in particular, multiplied with an adjustment operator. Thus, according to a further embodiment, the adjustment operator can be implemented as an adjustment factor. In principle, however, the approximation function can also have another function which is dependent on the measurement time and which is multiplied by the adjustment operator or processed in some other way. In principle, the approximation function can be dependent on the measurement time and the adaptation operator. The definition of the approximation function can define the type or the calculation rule for determining the adjustment operator.

The use of a time-dependent approximation function with a time-constant adjustment operator has the advantage that the approximation function—like the signal curve or the underlying process in the technical system—is dependent on time, but can be characterized and/or identified by the adjustment operator in a time-independent manner. The adjustment operator is thus a parameter both for the approximation function and, after approximation, also for the signal curve. The approximation can be done, for example, by numerically varying the adjustment operator and comparing the resulting results of the approximation function with the recorded measured value. An adjustment operator can then be sought for the approximation, which (on the basis of the comparison) yields the greatest agreement between the function result and the measured value. This adjustment operator can then be used to identify the signal curve and, depending on the value, also indicate the state of the system.

Carrying out the determination of the adjustment operator can include the step that the adjustment operator is calculated using the approximation function for the (at least one) measurement time, in particular is calculated in such a way that the approximation function is adjusted with the adjustment operator to the curve of the recorded measured values. The approximation function can first be calculated for various adjustment operators in order to approximate the measured value and/or signal curve for this purpose. The respective results can then be compared with the recorded measured value. The result that comes closest to the measured value is then selected. In this way, the adjustment operator used to calculate this result is specific to the concrete signal curve or measured value and is therefore defined as the determined adjustment operator. Using this adjustment operator, e.g. B. by the comparison with at least one or more threshold values, the comparison result, and by means of the comparison result when carrying out the evaluation of the state of the technical system can be determined.

The course of the measured variable can denote a change in the measured variable that is caused by a real physical process, such as pressure equalization. The signal curve includes the measured values, which are repeated during the change in the measured variable and, if necessary, recorded regularly, e.g. B. by the acquisition component. The signal profile is in particular a curve profile which can be mathematically approximated by the approximation function. The method according to the invention can be used to evaluate the course of the curve, for example to determine error states in a diagnostic function, by continuously evaluating the changes in the parameter, which can be newly determined with each calculation step. Within the scope of the invention, this parameter is also referred to as “adaptation operator”.

The waveform can z. B. be provided in the form of a pressure drop curve and / or a decaying waveform.

The method according to the invention can make it possible for the signal profile in the form of a pressure drop curve to be used in order to differentiate between different leak sizes (the larger the leak, the faster the pressure drop). In order to reduce computing time, instead of a complete e-function as an approximation function, only the value for the decrease rate can be considered, i.e. the adjustment operator. This can result directly from a logarithm of the measured pressure values. This calculated quantity, ie the adjustment operator, can then be checked against thresholds in order to grade the slope of the curve (and thus the underlying leakage). The threshold values can be predefined as a function of environmental variables, with the environmental variables being able to influence the absolute pressure (ambient pressure and temperature or tank temperature and internal tank pressure). The rate of decrease can also be correlated to the size of the leak and be constant over the pressure drop curve and time. Furthermore, the large difference in decay rate between a tight system and different leak sizes such as 0.5mm and 1.0mm diameter allows good selectivity.

The comparison of the adjustment operator determined with the at least one threshold value can also be carried out by first calculating a value dependent on the adjustment operator, e.g. the square of the adjustment operator, and comparing the value of this value, which is dependent on the adjustment operator, directly with the at least one threshold value.

It is also conceivable that when the at least one measured value is recorded, a time profile of at least two or at least three measured values is recorded for at least two or at least three different measurement times. When determining the adjustment operator, the adjustment operator can then be calculated on the basis of the recorded course over time, ie the at least two or at least three measured values. For example, an approximation with several points (ie measured values at specific measurement times) such as a two-point or three-point approximation can be used for this purpose. An approximation method (in the sense of a "best fit", i.e. an optimization method) can also be used. The approximation with a number of points can be understood to mean that the approximation function is approximated to the course of a number of measured values which are recorded at different measurement times. For example, the last two or last three recorded measured values can be used here as the measured values of the course over time, in particular as the points for the approximation.

The adjustment operator can be determined in order to adjust the approximation function to the recorded measurement values. The approximation function is specified, for example, in such a way that the approximation function can physically represent the signal curve. A physical reference course of the measured values can thus always be used as a basis for the evaluation. A function of the same type (e.g. the e-function) can be determined continuously on the basis of the recorded measured values. The methods that can be used for the determination (such as two-point/three-point approximation, with/without min/max adaptation) are described in more detail below.

• - eine Zwei-Punkt-Approximation,
• - eine Approximation nach Maximum-Adaption,
• - eine kontinuierliche Max-Min-Adaption,
• - eine Drei-Punkt-Approximation.

When determining the adjustment operator, at least one of the following calculation methods can be used to determine the adjustment operator:

• - a two-point approximation,
• - an approximation after maximum adaptation,
• - a continuous max-min adaptation,
• - a three-point approximation.

The various calculation methods are described in more detail below. Depending on the calculation methods, only one measured value determined at exactly one measurement time or other measured values at subsequent measurement times can be used for determining the adaptation operator.

Put more generally, it is conceivable that the determination of the adjustment operator includes an approximation of the approximation function to the recorded measured value only at the measurement time or also to at least or exactly two or three measured values at subsequent measurement times. The approximation can e.g. B. by varying the adjustment operator of the approximation function. For example, an initial value of the signal curve is also taken into account here, as will be described below using the example of a pressure curve. The adjustment operator can also be determined taking into account the measured value at the measurement time (ie the current measured values at the measurement time as the time of determination) and the time difference between the measurement time of the initial value and the measurement time of the current measurement value. As the time difference increases, the specific adjustment operator or the decrease rate represented thereby can assume an almost constant value. However, in the case of an investigation of a pressure drop and with a smaller time difference, the specific adjustment operator should not be taken into account, since in this range the mapping of the pressure drop using an exponential function can deviate from the real pressure drop. This area of the signal curve can be hidden accordingly. Furthermore, correction factors taken into account in order to improve the selectivity. Possible correction factors are, for example, information about the tank fill level and/or about a pressure increase in the tank system due to fuel outgassing or heat input and/or about an ambient temperature and/or about an ambient pressure and/or about an inside tank temperature and/or an inside tank pressure.

In the case of the two-point approximation, the adjustment operator of the approximation function (e.g. in the form of an exponential function) can be calculated using two measured values which were recorded at subsequent measurement times. For subsequent determinations of the adjustment operator, the measurement times should in particular have a constant time interval from one another.

In the case of the approximation after maximum adaptation, the adaptation operator can be determined taking into account an initial value of the signal curve. The initial value is, for example, the measured value that was recorded before a drop in the signal curve is present or recognized on the basis of the subsequent measured values. Furthermore, as described above, the time difference between the measurement time of the initial value and the measurement time of the current measured value can be taken into account.

The continuous max-min adaptation can take place in a corresponding manner, but in this case a minimum value together with a maximum value of the signal curve can also be taken into account.

In the case of the three-point approximation, at least or exactly three measured values recorded at successive measurement times can be used to determine the adaptation operator. Furthermore, based on the two-point approximation, there is no need to adapt values for calculating the adaptation operator.

Provision can be made for the steps of acquiring the measured value and/or determining the adaptation operator and/or carrying out the comparison and/or carrying out the evaluation to be carried out cyclically and/or to be part of an observation phase, in particular a pressure observation phase, in which the signal curve ( e.g. continuously). The observation phase can B. be initiated without actively bringing about a predefined initial value of the signal curve or the measured variable. For example, the observation phase is initiated when such a change in the signal curve is detected that indicates a drop in the signal curve (a kink in the curve or the like).

• - Einlesen des wenigstens einen Messwertes p(t), insbesondere Druckwertes p(t), zum Messzeitpunkt tm,
• - Ermitteln eines Maximalwertes p_max anhand des wenigstens einen eingelesenen Messwertes p(t) und/oder anhand wenigstens eines vorangegangenen eingelesenen Messwertes,
• - Anpassen (Korrigieren) des eingelesenen Messwertes p(t) anhand des ermittelten Maximalwertes p_max, z. B. durch die Berechnung p_neu (t) = p_max - p(t), wobei p_neu der angepasste Messwert ist.

The acquisition of the measured value can include at least one of the following steps:

• - reading in the at least one measured value p(t), in particular pressure value p(t), at the measurement time tm,
• - Determination of a maximum value p_max based on the at least one read measured value p(t) and/or based on at least one previously read measured value,
• - Adaptation (correction) of the measured value p(t) read in using the determined maximum value p_max, e.g. B. by calculating p_new (t) = p_max - p(t), where p_new is the adjusted measured value.

Determining the maximum value p_max can e.g. B. for the first measurement or the first measurements, so also before the first detection of a measured value p(t), which is used for the comparison and the evaluation and / or for the observation phase. For this purpose, the measurement can be carried out repeatedly in order to define the largest measured value of the measurement as the maximum value p_max. As soon as a drop in the measured value is identified as a result of the repeated measurement, the observation phase can be initiated, ie the steps of acquiring the measured value and/or performing the comparison and/or performing the evaluation can be carried out repeatedly.

In contrast to conventional methods, the solution according to the invention has the advantage that a targeted pumping up of the tank system to a defined target pressure (as the initial value of the measured variable pressure) is not required in order to obtain the signal curve. If a system other than a tank system is used as the technical system, it is accordingly not necessary to bring about an initial value of the measured variable in order to obtain the signal curve. The drop in the signal curve can thus be initiated from different initial values, e.g. B. the pressure equalization in the tank system can be initiated. An integration of the differential measured values (in particular the differential pressure) of the signal curve over time is also not necessary. In order to be able to deal with the different signal curves, an approximation function can be used according to the invention.

In the method according to the invention, the signal curve, e.g. B. in the form of a pressure drop curve, at different initial values (initial pressures), in particular independently of the initial value, e.g. by continuously calculating the curvature of the waveform.

• - Bestimmen eines ersten Zustands des Systems, wenn der (ermittelte) Anpassungsoperator oder ein davon abhängiger, z. B. daraus berechneter, Wert größer ist als der Schwellenwert, insbesondere ein erster Schwellenwert, und/oder Bestimmen eines dritten Zustands des Systems, wenn der (ermittelte) Anpassungsoperator oder der davon abhängige Wert größer ist als ein zweiter Schwellenwert, und andernfalls Bestimmen eines zweiten Zustands des Systems.

According to a further advantage, it can be provided that carrying out the evaluation of the comparison result includes the following step:

• - Determining a first state of the system if the (determined) adaptation operator or one dependent thereon, e.g. B. value calculated therefrom is greater than the threshold value, in particular a first threshold value, and/or determining a third state of the system if the (determined) adjustment operator or the value dependent thereon is greater than a second threshold value, and otherwise determining a second state of the system.

The result of the comparison can be the result of whether the (determined) adaptation operator or the dependent one, e.g. B. value calculated therefrom is greater than the threshold value - and in the case of several threshold values possibly also above which of the threshold values is exceeded. Correspondingly, when carrying out the comparison, it is also possible, if necessary, to carry out a number of comparisons and/or first of all to calculate the value dependent thereon from the adjustment operator, e.g. B. by squaring the adjustment operator. It may also be possible for the second condition to indicate a non-leak condition, the first condition to indicate a leak of a first leak diameter, and the third condition to indicate a leak of a second leak diameter. Thus, different leak diameters can be distinguished by the evaluation, e.g. B. by using appropriate threshold values.

Furthermore, it is optionally provided that the method for testing is carried out in the form of a tank leak diagnosis, the measured variable being a pressure in the technical system in the form of a tank system, the signal curve being specific for pressure equalization within the tank system, and the state of the technical system is specific to the presence of a leak in the tank system. The pressure profile when there is a leak in the tank system can depend on the differential pressure between the tank pressure and the ambient pressure. Based on the flow behavior through an orifice, the mass flowing through decreases with decreasing differential pressure. In particular, this leads to the pressure falling very quickly at the beginning of the pressure monitoring phase and flattening out as the differential pressure falls. According to the invention, the course of this pressure drop curve can be used to differentiate between different leak sizes (the larger the leak, the faster the pressure drop).

• - Schließen eines Druckventils am luftseitigen Einlass eines Aktivkohlebehälters des Tanksystems (folglich existieren im Tanksystem zwei verschlossene Räume mit unterschiedlichen Druckniveaus und verschieden großen Volumina),
• - Öffnen eines Tankabsperrventils des Tanksystems, um den Druckausgleich zu initiieren.

It is also advantageous if a change in the measured variable is actively initiated in the technical system in the form of a tank system, for example by switching at least one pressure valve of the system. To trigger a change in the measurand in the system in the form of a tank system, the following steps can be taken:

• - Closing a pressure valve on the air-side inlet of an activated charcoal canister of the tank system (therefore there are two closed spaces with different pressure levels and different volumes in the tank system),
• - Opening a tank shut-off valve of the tank system to initiate pressure equalization.

Triggering the change in the measured variable can define the start of the pressure monitoring phase. The triggering can take place independently of the initial pressure.

The tank system can B. as components have an activated charcoal canister and / or a tank vent valve and / or a tank shut-off valve. The pressure equalization can, for. B. in the subspace between these components.

Furthermore, it is optionally possible within the scope of the invention for the method to be carried out in a vehicle, with the tank system being part of the vehicle. The vehicle is z. B. a motor vehicle and / or a passenger vehicle and / or a hybrid vehicle with an internal combustion engine and an electric motor.

It is also conceivable within the scope of the invention for the approximation function to have the adjustment operator in the form of a variable adjustment operator for approximating the signal curve, which is used to characterize the signal curve and/or the state of the system. The adjustment operator can therefore characterize different waveforms depending on its value.

Furthermore, it is conceivable that the approximation function has a function dependent on the measurement time, which is adjusted by the adjustment operator in the form of a variable adjustment operator, for example multiplied or added to the adjustment operator for this purpose. Alternatively or additionally, it is possible that when the adjustment operator is determined, the adjustment operator is varied and/or that adjustment operator is determined by which the approximation function yields (at least approximately) the recorded measured value at the time of measurement. In other words, the adjustment operator of the approximation function can be varied when determining, and thus that adjustment operator can be determined which leads to a result of the approximation function which comes closest to the signal curve and in particular the measured value. In this way, the determination can take place in that the signal profile is approximated by the approximation function. Furthermore, the approximation of the approximation function to the recorded measured value at the time of measurement is possible in a particularly reliable manner in this way.

It may be possible that the approximation function is denoted as "w(t)", and by "w(t) = c*f(t)" or "w(t)= f(t,c)", in particular " w(t)= e( ^-c*t )” is defined. "*" designates the operator for the multiplication, "f(t)" the function dependent on the measuring time tm, in particular exponential function "e" or logarithm. It can therefore be provided with further advantage that the approximation function has the function dependent on the measurement time in the form of an exponential function or a logarithm, “c” is the variable adaptation operator.

The result of the approximation function w(t) should correspond approximately to the measured value if a sufficient approximation has been made. For this purpose, for example, after each step of detection when carrying out the determination of the adjustment operator c, the course of the p_new values or also an individual p_new value can be approximated by the approximation function w(t).

The adjustment operator c can also be designed as a prefactor of the approximation function w(t) or as a decrease rate of the signal curve. The calculated adjustment operators can then be compared against the diagnostic thresholds of individual leaks. The threshold values defined for the diagnostic thresholds can be determined as a function of environmental variables that have an influence in particular on the absolute pressure (ambient pressure and temperature or tank temperature and internal tank pressure, tank fill level). Provision can also be made for the threshold value(s) for the comparison with the adjustment operator or a value dependent thereon to be defined in such a way that they are specific to the measured value and/or determined adjustment operator and/or the measurement time for the presence of a leak. The calculation of the adjustment operator c enables a simple and reliable distinction between different curve shapes of the signal curve.

The invention also relates to a test system, in particular for carrying out a method according to the invention.

• - eine Bereitstellungskomponente zum Bereitstellen eines Anpassungsoperators für eine Annäherungsfunktion zur Annäherung an den Signalverlauf,
• - eine Erfassungskomponente zum Erfassen wenigstens eines Messwertes des Signalverlaufs zu wenigstens einem Messzeitpunkt,
• - eine Verarbeitungskomponente zum Ermitteln des Anpassungsoperators für die Annäherung der Annäherungsfunktion an den erfassten Messwert und zum Durchführen eines Vergleichs des ermittelten Anpassungsoperators mit wenigstens einem Schwellenwert, um ein Vergleichsergebnis des Vergleichs zu bestimmen,
• - eine Auswertekomponente zum Durchführen einer Auswertung des Vergleichsergebnisses, um wenigstens einen Zustand des technischen Systems zu bestimmen.

The test system has the following components, which are preferably designed for integration into a vehicle:

• - a provisioning component for providing a fitting operator for an approximation function for approximating the waveform,
• - a detection component for detecting at least one measured value of the signal curve at at least one measurement time,
• - a processing component for determining the adjustment operator for the approximation of the approximation function to the recorded measured value and for carrying out a comparison of the adjustment operator determined with at least one threshold value in order to determine a comparison result of the comparison,
• - an evaluation component for carrying out an evaluation of the comparison result in order to determine at least one state of the technical system.

The test system according to the invention thus brings with it the same advantages as have been described in detail with reference to a method according to the invention. The detection component and/or processing component and/or evaluation component can be part of electronics, for example at least one microcontroller, and/or a computer program. Furthermore, the evaluation component can also be part of the processing component. The processing component and/or evaluation component can have a processor for data processing in order to carry out numerical calculations for the determination and/or comparison and/or the evaluation.

The vehicle with the test system is also placed under protection.

• 1 eine schematische Darstellung eines erfindungsgemäßen Prüfsystems
• 2 beispielhafte Kurven mit der zugehörigen Abnahmerate,
• 3 und 4 beispielhafte Kurven des Signalverlaufs und der Annäherungsfunktion.

Further advantages, features and details of the invention result from the following description, in which exemplary embodiments of the invention are described in detail with reference to the drawings. The features mentioned in the claims and in the description can each be essential to the invention individually or in any combination. Show it:

• 1 a schematic representation of a test system according to the invention
• 2 exemplary curves with the associated decrease rate,
• 3 and 4 exemplary curves of the signal course and the approximation function.

In the following figures, the same reference numbers are used for the same technical features of different exemplary embodiments.

In 1 a test system 300 according to the invention is shown schematically. The test system 300 can be integrated into a vehicle 2 . Furthermore, a technical system 1 can be provided in the vehicle 2, such as a tank system 1.

The test system 300 can be coupled to the tank system 1 or integrated therein in order to test a signal curve s, the signal curve s being specific to a time curve of a measured variable 201 in a technical system 1 . For this purpose, the test system 300 can comprise a provision component 310 for providing an adaptation operator c for an approximation function w for approximating the signal curve s. Furthermore, an acquisition component 320 for acquiring at least one measured value 202 of the signal curve s at at least one measurement time tm can be provided. In addition, it is possible that a processing component 330 is provided for determining 101 the adjustment operator c for the approximation of the approximation function w to the recorded measured value 202 and for carrying out a comparison 102 of the determined adjustment operator c with at least one threshold value sw in order to obtain a comparison result 205 of the Comparison 102 to determine. An evaluation component 340 of the test system 300 can also be used to carry out an evaluation 103 of the comparison result 205 in order to determine at least one state of the technical system 1 . In this way, the test system 300 can also be used to carry out the method steps (e.g. provided with the reference symbols 101 to 103) of a method according to the invention.

It is possible that the method according to the invention for testing is carried out in the form of a tank leak diagnosis, with the measured variable 201 being a pressure 201 in the technical system 1 in the form of a tank system 1, and the signal curve s being specific for a pressure equalization within the tank system 1 is. The state of the technical system 1 can then be specific to the presence of a leak in the tank system 1, ie, for example, a first state with and a second state without a leak.

mit p_min als den minimalen Druck, also der Druck, dem sich der Kurvenverlauf langfristig annähert. p_max entspricht dem maximalen Druck, also dem Druck zu Beginn des Druckabfalls. p₀ ist der maximale Druck-Zuwachs, d. h. die Differenz von maximalen und minimalen Druck. Ferner ist -a die (Druck-)Abnahmerate, welche durch den Anpassungsoperator c annäherungsweise repräsentiert werden kann. Der Wert a bzw. c ist somit ein Maß dafür, wie schnell der Druckabnahmeprozess stattfindet. Es ist daher möglich, diesen für Diagnosekriterien zur Erkennung von Leckagen in Druckleitungen einzubeziehen.If the pressure drops, the pressure values follow an e-function (with the measurement time tm = t): $$p\left(t\right)=p_0{e}^{-a\ast t}+p_{min}=\left(p_{max}-p_{min}\right)\cdot e^{-a\ast t}+p_{min}$$

with p _min as the minimum pressure, i.e. the pressure to which the curve progresses in the long term. p _max corresponds to the maximum pressure, i.e. the pressure at the beginning of the pressure drop. p ₀ is the maximum pressure increase, ie the difference between maximum and minimum pressure. Furthermore, -a is the (pressure) decrease rate, which can be approximately represented by the adjustment operator c. The value a or c is therefore a measure of how quickly the pressure reduction process takes place. It is therefore possible to include this for diagnostic criteria for detecting leaks in pressure lines.

The larger the value a for the rate of decrease, the faster the pressure falls. In closed systems, this is an indication of a leak. Similar relationships apply to other technical systems, so that the state of the system can generally be inferred from the rate of decrease.

In the 2 until 4 curves of a pressure drop are shown, in which the signal curve s thus corresponds to a pressure curve by way of example. Here p denotes the pressure difference with the unit hPa, t the time and s1 to s4 pressure curves for different states of the technical system. The adjustment operators c for the associated pressure curves s1 to s4 are indicated by c1 to c4.

In 2 the pressure curve s1 is given for a condition without a leak, the pressure curve s2 for a condition with a leak (at a diameter of 0.5 mm) and the pressure curve s3 for a condition with a leak (at a diameter of 1.0 mm). . It can be seen that the presence of the resulting matching operator c in a first area b1 can indicate a leak greater than or equal to 1.0 mm, in a second area b2 a leak greater than 0.5 mm and in a third area v3 can indicate a leak smaller than 0.5 mm.

The determination 101 of the adjustment operator c for the approximation of the approximation function w to the recorded measured value 202 at the current measurement time tm (ie recorded at the current measurement time tm) is carried out in particular to characterize the signal curve s at the current measurement time tm using the adjustment operator c. Subsequently, the adjustment operator c determined can be compared with at least one threshold value sw in order to determine a Ver to determine the same result 205 of the comparison 102 . In this case, the threshold value can be fixed or determined as a function of environmental conditions, for example as a function of temperature and pressure. In addition, the threshold values can optionally also be defined as a function of time and/or as a function of the signal curve s. Further conditions can be used in the subsequent evaluation of the comparison result and/or already in the comparison 102 . For example, the determined adjustment operator can be checked for plausibility (it is checked whether meaningful values were determined for it).

It can also be possible for a plurality of adjustment operators c (at successive measurement times) to be determined one after the other, and for the rate of change of the determined adjustment operators c to be determined. The rate of change can then be compared to an upper and/or lower threshold. If the rate of change falls below the lower threshold and/or exceeds the upper threshold, the adjustment operator c is considered to be non-constant in time, preventing the evaluation from determining the state of the system (i.e. not enabling diagnosis of a leak, for example) . Otherwise, the adjustment operator c can be regarded as sufficiently constant to be able to carry out the determination of the state.

Furthermore, the method steps according to the invention can only be carried out if a drop in pressure has previously been detected. It is also possible to check whether the pressure value is within a test interval, so that the method steps of the method according to the invention can only be carried out then. A specific time range of the signal curve s can also be blanked out so that the method steps are not carried out in this range either.

The method steps according to the invention can only be carried out, for example, if all conditions such as a detected drop in pressure and/or a sufficiently constant rate of decrease and/or the presence of the pressure values in the test area and/or other release conditions are met. Depending on the status determined by the evaluation, an error message may be output.

• - eine Zwei-Punkt-Approximation,
• - eine Approximation nach Maximum-Adaption,
• - eine kontinuierliche Max-Min-Adaption,
• - eine Drei-Punkt-Approximation.

When determining 101 the adjustment operator c, at least one of the following calculation methods can be used to determine the adjustment operator c to c, in particular to calculate it, preferably to calculate it numerically by the processing component:

• - a two-point approximation,
• - an approximation after maximum adaptation,
• - a continuous max-min adaptation,
• - a three-point approximation.

p₁ ist ein erster Messwert 202 zu einem ersten Messzeitpunkt tm. p₂ ist ein zweiter Messwert 202 zu einem zweiten Messzeitpunkt tm, der nach dem ersten Messzeitpunkt tm folgt. t₂ - t₁ entspricht hierbei „dT“, also einem Zeitintervall zwischen dem ersten und zweiten Messzeitpunkten tm.For the two-point approximation, the adjustment operator c can be calculated as follows: $$c=\frac{\text{log}\left({\mathrm{<figure-callout\; id="1"\; label="p"\; filenames="DE102020130375A1\_0009.png,DE102020130375A1\_0010.png"\; state="\{\{state\}\}">p</figure-callout>}}_1-p_{min}\right)-\text{log}\left({\mathrm{<figure-callout\; id="2"\; label="p"\; filenames="DE102020130375A1\_0008.png,DE102020130375A1\_0010.png"\; state="\{\{state\}\}">p</figure-callout>}}_2-p_{min}\right)}{{\mathrm{<figure-callout\; id="2"\; label="t"\; filenames="DE102020130375A1\_0008.png,DE102020130375A1\_0010.png"\; state="\{\{state\}\}">t</figure-callout>}}_2-{\mathrm{<figure-callout\; id="1"\; label="t"\; filenames="DE102020130375A1\_0009.png,DE102020130375A1\_0010.png"\; state="\{\{state\}\}">t</figure-callout>}}_1}$$

p ₁ is a first measured value 202 at a first measurement time tm. p ₂ is a second measured value 202 at a second measurement time tm, which follows the first measurement time tm. In this case, t ₂ -t ₁ corresponds to “dT”, ie a time interval between the first and second measurement times tm.

$$c\approx \frac{log\left(p_1\right)-log\left(p_2\right)}{t_2-t_1}$$

Hierbei erfolgt also keine Berücksichtigung von p_min (dem „Min-Offset“) bzw. p_max. Um dennoch die Robustheit der Annäherungsfunktion zu verbessern, kann ein zulässiger Prüfbereich für den Signalverlauf s festgestellt sein.For convenience, the adjustment operator c can also be calculated as follows: $$c\cdot \mathrm{i.e}T\approx lOG\left(p_1\right)-lOG\left(p_2\right)$$

$$c\approx \frac{lOG\left(p_1\right)-lOG\left(p_2\right)}{{\mathrm{<figure-callout\; id="2"\; label="t"\; filenames="DE102020130375A1\_0008.png,DE102020130375A1\_0010.png"\; state="\{\{state\}\}">t</figure-callout>}}_2-{\mathrm{<figure-callout\; id="1"\; label="t"\; filenames="DE102020130375A1\_0009.png,DE102020130375A1\_0010.png"\; state="\{\{state\}\}">t</figure-callout>}}_1}$$

In this case, p _min (the "min offset") or p _max is not taken into account. However, in order to improve the robustness of the approximation function, a permissible test range for the signal curve s can be determined.

In 3 the result of the two-point approximation is shown as an example. Threshold values sw1 to s3 are also shown, which enable the determined adaptation operators c to be evaluated. As in 2 the ranges b1 to b3 are also given here, in which the determined adaptation operators c can lie, depending on the leak size. Signal curve s1 represents a pressure drop with a leak of 1.0 mm, signal curve s2 represents a pressure drop with a leak of 1.0 mm, but with a higher initial value. Signal curves s3 and s4 represent pressure drops a 0.5 mm leak, but with different initial values. The associated adaptation operators c1 to c4, which are determined at the various measurement times tm, are also shown.

Sobald es zum Druckabfall kommt, kann der zuvor erfasste Messwert als p_max gespeichert werden und für die weiteren Iterationen der Verfahrensschritte bzw. des Ermittelns 101 verwendet werden.For the approximation after maximum adaptation, the adaptation operator c can be calculated as follows (with the measurement time tm=t): $$\begin{array}{c}c=\frac{lOG\left(p_{max}-p_{min}\right)-lOG\left(\left(p\left(t\right)-p_{min}\right)\right)}{t}\\ \approx \frac{lOG\left(p_{max}\right)-lOG\left(p\left(t\right)\right)}{t}\end{array}$$

As soon as the pressure drops, the previously recorded measured value can be stored as p _max and used for the further iterations of the method steps or the determination 101 .

To further increase the accuracy of the calculation of c, both the maximum and minimum pressure can be taken into account (continuous max-min adaptation): $$c=\frac{lOG\left(p_{max}-p_{min}\right)-lOG\left(p\left(t\right)-p_{min}\right)}{t}$$

It is possible to estimate the minimum value p _min using a characteristic as soon as the maximum value p _max is known. The minimum value after the end of the pressure drop can also be determined using the signal curve s. Furthermore, the maximum and minimum values p _max and p _min can also be updated during the iterations based on the current measured values. If the current measured value is smaller than the previously recorded measured values, this current measured value can be buffered as the new minimum value for the further iterations until an even smaller measured value is recorded. If the current measured value is greater than the previously recorded measured values, this current measured value can be buffered as the new maximum value for further iterations until an even larger measured value is recorded.

In 4 the result of the approximation after maximum adaptation is shown as an example. Threshold values sw1 to s3 are also shown, which enable the determined adjustment operators c to be evaluated. As in 2 and 3 the ranges b1 to b3 are also given here, in which the determined adaptation operators c can lie, depending on the leak size. Signal curve s1 represents a pressure drop for a leak of size 1.0 mm, and signal curve s2 represents a pressure drop for a leak of size 1.0, but with a higher initial value. Signal curves s3 and s4 represent pressure drops for a 0.5 mm leak, but with different initial values. The associated adaptation operators c1 to c4, which are determined at the various measurement times tm, are also shown. It can also be seen that when using the max value adaptation method, the values for the rate of decrease a calculated at the beginning of the pressure drop fluctuate greatly and are not suitable for an evaluation. Therefore, this area A is hidden for the evaluation here.

For the three-point approximation, the adjustment operator c can be calculated as follows $$c=\frac{\left(lOG\left({\mathrm{<figure-callout\; id="1"\; label="p"\; filenames="DE102020130375A1\_0009.png,DE102020130375A1\_0010.png"\; state="\{\{state\}\}">p</figure-callout>}}_1-p_2\right)-lOG\left({\mathrm{<figure-callout\; id="2"\; label="p"\; filenames="DE102020130375A1\_0008.png,DE102020130375A1\_0010.png"\; state="\{\{state\}\}">p</figure-callout>}}_2-p_3\right)\right)}{\mathrm{i.e}T}$$

In order to increase the robustness of the method, values should be used at larger time intervals, so that dT is effectively larger than one calculation step.

It is also possible that dT is the step size. The increment dT can be set variably for the implementation of the algorithm. The increment dT can also optionally designate the distance between two iterations in which the method steps are carried out, or the distance in which the method steps of detecting and/or determining and/or comparing and/or evaluating are carried out.

The increment dT can be variable. The comparison with pressure curves in the presence of different leaks shows that different increments are optimal for the different leak sizes. In the case of large leaks, the increment can be selected smaller than in the case of small leaks. The three-point approximation can also be carried out several times, with different values for the increment dT.

The above explanation of the embodiments describes the present invention exclusively in the context of examples. It goes without saying that individual features of the embodiments can be freely combined with one another, insofar as this makes technical sense, without departing from the scope of the present invention.

Reference List

1

system, tank system

2

vehicle

101

Determine

102

comparison

103

Evaluation

201

measurement, pressure

202

reading

205

comparison result

300

test system

310

deployment component

320

acquisition component

330

processing component

340

evaluation component

p

Pressure in hPa

s

waveform

bw

threshold

t

time

tm

time of measurement

w

approximation function

A

blanking area

b

areas

c

adjustment operator

QUOTES INCLUDED IN DESCRIPTION

This list of the documents cited by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent Literature Cited

• DE 102015014873 A1 [0002]

Claims (10)

Method for checking a signal curve (s), the signal curve (s) being specific to a time curve of a measured variable (201) in a technical system (1), the following steps being carried out: - providing a fitting operator (c) for an approximation function (w) for approximating the waveform (s), - detecting at least one measured value (202) of the signal curve (s) at at least one measurement time (tm), - determining (101) the adjustment operator (c) for the approximation of the approximation function (w) to the recorded measured value (202), - Carrying out a comparison (102) of the determined adaptation operator (c) with at least one threshold value (sw) in order to determine a comparison result (205) of the comparison (102), - Carrying out an evaluation (103) of the comparison result (205) in order to determine at least one state of the technical system (1). procedure after claim 1 , characterized in that the approximation function (w) has a function dependent on the measurement time (tm) in the form of an exponential function and/or a logarithm, and the adaptation operator (c) is designed as an adaptation factor (c) of the function. Method according to one of the preceding claims, characterized in that the approximation function (w) is implemented as a time-dependent function which is only dependent on the parameters time (t) and the adaptation operator (c). Method according to one of the preceding claims, characterized in that the approximation function (w) has a function which is dependent on the measurement time (tm) and which is adapted by the adaptation operator (c) in the form of a variable adaptation operator (c), wherein when determining (101) of the adjustment operator (c) that adjustment operator (c) is determined by which the approximation function (w) results in the recorded measured value (202) at the measurement time (tm). Method according to one of the preceding claims, characterized in that carrying out the evaluation (103) of the comparison result (205) comprises the following step: - determining a first state of the system (1) if the adaptation operator (c) or a value dependent thereon is greater than the threshold (sw), and otherwise determining a second state of the system (1). Method according to one of the preceding claims, characterized in that the method for testing is carried out in the form of a tank leak diagnosis, the measured variable (201) being a pressure (201) in the technical system (1) in the form of a tank system (1). , wherein the signal curve (s) is specific for a pressure equalization within the tank system (1), and the state of the technical system (1) is specific for the presence of a leak in the tank system (1), the method being used in a vehicle (2 ) is carried out, the tank system (1) being part of the vehicle (2). Method according to one of the preceding claims, characterized in that when the at least one measured value (202) is detected, a time profile of at least two or at least three measured values (202) is detected for at least two or at least three different measurement times (tm), with the determination (101) of the adaptation operator (c) the adaptation operator (c) is calculated on the basis of the course over time. Method according to one of the preceding claims, characterized in that the approximation function (w) has the adjustment operator (c) in the form of a variable adjustment operator (c) for approximating the signal curve (s), which characterizes the signal curve (s) and/or or the state of the system (1) is used. Method according to one of the preceding claims, characterized in that in the technical system (1) in the form of a tank system (1) a change in the measured variable (201) is actively initiated. Test system (300) for carrying out a method according to one of the preceding claims, having: - a provision component (310) for providing an adaptation operator (c) for an approximation function (w) for approximating the signal curve (s), - a detection component (320) for detecting at least one measured value (202) of the signal curve (s) at at least one measurement time (tm), - a processing component (330) for determining (101) the adjustment operator (c) for the approximation of the approximation function (w) to the detected measured value ( 202) and for carrying out a comparison (102) of the determined adaptation operator (c) with at least one threshold value (sw) in order to obtain a comparison result (205) of the ver to determine the same (102), - an evaluation component (340) for carrying out an evaluation (103) of the comparison result (205) in order to determine at least one state of the technical system (1).

Images