DE102019202126A1 - Method for determining at least one parameter of a piezo sensor of a fuel injector - Google Patents

Abstract

The invention relates to a method (100) for determining at least one parameter (SSF, τ) of a piezo sensor (300) of a fuel injector with the following steps: measuring (120) a profile of a pressure (p) inside the fuel injector; Measuring (140) an output voltage (U) of the piezo sensor (300); Determining (160) a modeled output voltage (U) of the piezo sensor (300); and determining (180) a measure (J) for a deviation of the modeled output voltage (U) from the measured output voltage (U) of the piezo sensor (300).

Description

The present invention relates to a method for determining at least one parameter of a piezo sensor of a fuel injector, a computer program, a machine-readable storage medium and an electronic control device.

State of the art

In the prior art, common rail injection is known, in which high-pressure fuel injection valves or injectors are used. In the prior art, control functions were developed for this purpose in order to improve the injection accuracy of the fuel injection valves or to maintain it over the service life. These control functions are based on a sensor signal or its evaluation in order to determine characteristic variables of the injector.

In the case of a piezo injector of a common rail system of the prior art, this sensor is an active piezo sensor which is attached to the outside of a holding body at the level of a high pressure bore. The sensor reacts to a change in the mechanical tension, which is triggered due to changes in pressure in the high-pressure bore. To detect the sensor signal curve, the sensor is connected to a fast analog-digital converter, the so-called ΔΣ converter, via suitable input circuitry.

According to a simplified equivalent circuit diagram of the sensor and the ΔΣ converter, the sensor generates a voltage U _{NCS, Ini} , which is proportional to the pressure that prevails in the high-pressure bore. The proportionality factor, also called sensor scaling factor, is abbreviated as SSF. The voltage applied to the sensor is discharged through the resistance of the input circuit. The two components of the equivalent circuit that are most relevant for the discharge and its temporal spread are the sensor capacitance C _NCS and the internal resistance R _{ΔΣ, in of} the ΔΣ converter. Their product gives the time constant τ of the corresponding RC element, which determines the course of the discharge.

Since the parameters contained in the equivalent circuit diagram, ie SSF, C _NCS and R _{ΔΣ, vary widely,} the time constant cannot be precisely determined.

In one of the above-mentioned control functions, which is called SQA (Small Quantity Adaption), the sensor output is required as an absolute value. In this case, sensor scaling is therefore required, which can be determined with great effort using a test injection.

Other control functions rely on the detection of characteristic features of the injector behavior such as the needle closing time. Different discharge time constants hardly influence the recognized times.

The pamphlet DE 10 2014 204 629 A1 discloses a fuel injector, which has a measuring device on its injector housing for detecting the fuel pressure in a fuel-carrying bore, which measuring device is arranged on the outside of the injector housing.

Disclosure of the invention

The method is used to determine at least one parameter of a piezo sensor of a fuel injector. Here, a parameter of a piezo sensor is understood to mean a parameter which characterizes at least one physical property of the piezo sensor.

The piezo sensor measures a mechanical tension which prevails in the high pressure bore of a high pressure fuel injection valve.

According to a first step of the method, a pressure curve in the interior of the fuel injector is measured. Here, the appropriate pressure curve can be viewed as the input signal of the sensor.

The fuel injector is preferably a high-pressure fuel injection valve, which can thus be used in a common rail system.

The piezo sensor is preferably an active piezo sensor which reacts to changes in the mechanical tension.

According to a second step of the method, an output voltage of the piezo sensor is measured. The first and second steps are preferably carried out identically.

According to a third step of the method, a modeled or simulated output voltage of the piezo sensor is determined. This is preferably done with stimulation with the measured input signal.

According to a fourth step of the method, a measure for a deviation of the modeled simulated output voltage from the measured output voltage of the sensor is determined.

The method advantageously ensures that at least one parameter of the piezo sensor can be determined without the possibility of a test injection. Furthermore, the present method a sensor scaling, whereby the above-mentioned SQA control function can be implemented. In addition, other detection functions are improved by compensating for the sensor discharge. Another detection function can be a function called Pilot Quantity Detection (POD). The present sensor scaling of the described method is only a first part of the POD. The signal is then integrated, which can be improved.

According to a preferred embodiment, a parameter of the piezo sensor can be a parameter of an equivalent circuit of the piezo sensor.

According to a preferred embodiment, according to a fifth step of the method, it is determined whether the measure for the deviation of the modeled output voltage from the measured output voltage of the sensor is smaller than a predefined variable. This advantageously creates the prerequisite that a decision can be made as to whether the modeled output voltage is close enough to the measured output voltage, which can serve as a termination criterion.

According to a further preferred embodiment, a sixth step of the method is carried out if the measure for the deviation of the modeled output voltage from the measured output voltage of the sensor is greater than the predefined variable. According to the sixth step, in this case the measure for the deviation of the modeled output voltage from the measured output voltage of the sensor is reduced by varying at least one parameter of the piezo sensor. This feature advantageously ensures that the value of the modeled output voltage approximates the measured output voltage of the sensor.

According to yet another preferred embodiment, the method returns to the third step of the method, in which the modeled output voltage of the piezo sensor is determined. In this case, the modeled output voltage is now determined with at least one varied parameter of the piezo sensor, so that the value of the modeled output voltage generally differs from the value of the previously determined modeled output voltage.

This feature advantageously creates the prerequisite for an iterative repetition of the determination of the value of the modeled output voltage to differ the value of the modeled output voltage from the value of the measured output voltage by only a predetermined amount.

According to a preferred embodiment, the sixth, third and fourth steps are carried out iteratively until the measure for the deviation of the modeled output voltage from the measured output voltage of the sensor is smaller than the predetermined value. This advantageously means that the value of the modeled output voltage differs from the value of the measured output voltage only by a predetermined amount.

According to a preferred embodiment, the at least one, preferably all, parameters are determined after the end of the iterative implementation. The at least one parameter can be output in the method or stored in a suitable device.

The step of reducing the measure for the deviation of the modeled output voltage from the measured output voltage of the sensor by varying at least one parameter of the piezo sensor is preferably implemented by an optimization algorithm. The optimization algorithm is preferably implemented using a gradient method. This is a standard method which advantageously recognizes by forming a gradient in which direction the at least one parameter is to be changed in order to achieve an optimal value as quickly as possible.

The computer program is set up to carry out each step of the method, in particular when it runs on an electronic control device or computing device. This enables the implementation of the method in a conventional control device without having to make structural changes to it. For this purpose, the computer program is stored on a machine-readable storage medium. By uploading the computer program to a conventional electronic control device, the electronic control device is obtained, which is set up to carry out a method for determining at least one parameter of a piezo sensor of a fuel injector.

Figure list

• 1 zeigt ein schematisches Ablaufdiagramm eines Verfahrens gemäß einem Ausführungsbeispiel der Erfindung.
• 2 zeigt ein Ersatzschaltbild des Piezo-Sensors inklusive einer Messschaltung.

An embodiment of the invention is shown in the drawings and is explained in more detail in the following description.

• 1 shows a schematic flow diagram of a method according to an embodiment of the invention.
• 2 shows an equivalent circuit diagram of the piezo sensor including a measuring circuit.

Embodiment of the invention

1 shows a procedure 100 for determining parameters of an equivalent circuit of a piezo sensor of a fuel injector. The procedure 100 starts with the first step 120 , in which a profile of a pressure p inside the high-pressure fuel injection valve is measured.

At the second step 140 of the procedure 100 an output voltage U _{NCS, measurement of} the piezo sensor is measured.

In the third step 160 of the procedure 100 a modeled output voltage U _{NCS, Sim of} the equivalent circuit is determined.

A simple model of the piezo sensor is included for this purpose. Equivalent circuit created, which in the present case has a first-order high pass.

2 shows an equivalent circuit diagram of the piezo sensor 300 including a measuring circuit 320 . The one on the piezo sensor 300 acting compressive force p is indicated by the arrow 302 shown. The piezo sensor 300 creates one by the arrow 304 shown voltage U _{NCS, Ini} , which is proportional to the pressure force p, which prevails in the high pressure bore. Here the proportionality factor or sensor scaling factor is abbreviated with the designation SSF.

The one on the piezo sensor 300 applied voltage U _{NCS, Ini} discharges through the total resistance of the input circuit 320 . Two components of the equivalent circuit that are relevant for the discharge and its temporal spread are those with the reference symbol 324 marked sensor capacitance C _NCS and with the reference symbol 322 marked internal resistance R _{ΔΣ, in of} the ΔΣ converter 326 . The product of the sensor capacitance C _NCS and the internal resistance R _{ΔΣ, in} gives the time constant τ of the corresponding RC element, which determines the course of the discharge. From an electrical perspective, the piezo sensor is a voltage source, the voltage of which is dependent on the change in pressure, and a capacitance.

Digital recursive filters can simulate analog filters made up of resistors and capacitors. chapter 19th of the book "The Scientist &Engineer's Guide to Digital Signal Processing", 1st edition by Steven W. Smith discloses a recursive calculation method for such filters. A discretized calculation at equidistant points in time is used here. The time interval between the points in time is constant and each point in time is numbered with integer indices n. The calculation takes place in a loop over all indices n. The current value y (n) of a filter function is dependent on the previous value y (n-1). The above first-order high pass is on page 322 ff. under the heading “Single Pole Recursive Filters”. The one on side 323 disclosed equation 19-3 gives the recursion coefficients for the first order high pass as follows: $$a_0=\frac{\left(1+\text{x}\right)}{2},a_1=-\frac{\left(1+\text{x}\right)}{2}=-a_0,b_1=x$$

The parameter x is a value between zero and one that controls the properties of the recursive filter. The value of the parameter x is the amount of change or decay between neighboring values. Thus x is given by the following formula: $$x=e^{-\frac{\Delta t}{\tau }}$$

U(n) und U(n - 1) repräsentieren indizierte Werte der simulierten Ausgangsspannung U_NCS,Sim. In der Ersatzschaltung bilden Sensorkapazität C_NCS und des Innenwiderstands R_ΔΣ,in einen Hochpass erster Ordnung.Here, Δt is the time difference between adjacent discrete values of the recursive calculation. By substituting the coefficient of formula (1) and the value for x of formula (2) in the recursion formula 19-1 which on side 320 of the above book is disclosed, the following formula (3) results in the present case in order to implement the above-mentioned simple model. The formula (3) is well suited for calculation in real time. $$U\left(n\right)=a_0\cdot p\left(n\right)\cdot \mathrm{S.}\mathrm{S.}\mathrm{F.}+a_1\cdot p\left(n-1\right)\cdot \mathrm{S.}\mathrm{S.}\mathrm{F.}+b_1\cdot U\left(n-1\right)$$

U (n) and U (n - 1) represent indexed values of the simulated output voltage U _{NCS, Sim} . In the equivalent circuit, the sensor capacitance C _NCS and the internal resistance R form _{ΔΣ in} a first-order high-pass filter.

In the fourth step 180 of the procedure 100 becomes a measure J for a deviation of the modeled output voltage U _{NCS, Sim} from the measured output voltage U _{NCS, measurement of} the piezo sensor 300 determined.

For this purpose, a change in pressure p is brought about during operation of the high-pressure fuel injection valve, this change is measured and, at the same time, the output signal U _{NCS, measurement of} the piezo sensor 300 measured. The change in pressure p is measured with a rail pressure sensor. The pressure increase that occurs as a result of the delivery strokes of the high-pressure pump can be used as the pressure change event.

The measured rail pressure profile is fed into the previously described simple model as an input signal, whereby the modeled output voltage U _{NCS, Sim} can be determined.

In the fifth step 200 of the procedure 100 it is determined whether the measure J for the deviation of the modeled output voltage U _{NCS, Sim} from the measured output voltage U _{NCS, measurement of} the piezo sensor 300 is smaller than a predetermined size J ₀ . If this is not the case, the procedure continues with the sixth step 220 away. If this is the case, the procedure continues with step 240 continues, in which the two parameters SSF and τ are output and then the method 100 is terminated.

The deviation J between the simulated output signal U _{NCS, Sim} and the measured sensor signal U _{NCS, Mess} can be determined using the formula (4) below: $$J={\displaystyle \sum {}_i{\left(U_{N\mathrm{C.}\mathrm{S.},\mathrm{S.}im}\left[i\right]-U_{N\mathrm{C.}\mathrm{S.},\mathrm{M.}ess}\left[i\right]\right)}^2}$$

Accordingly, the deviation J between the simulated output signal U _{NCS, Sim} and the measured sensor signal U _{NCS, Mess is} a sum of the squared deviations of the simulated output signal U _{NCS, Sim} from the measured sensor signal U _{NCS, Mess,} summed over all discretization points. The discretization points are provided with the index i. The deviation J is a function of the two parameters SSF and τ.

In the sixth step 220 the dimension J for the deviation of the modeled output voltage U _{NCS, Sim} from the measured output voltage U _{NCS, measurement of} the piezo sensor 300 reduced by varying at least one parameter of the two parameters SSF and τ of the equivalent circuit.

The deviation between the simulated output signal U _{NCS, Sim} and the measured sensor signal U _{NCS, Mess} is reduced by means of an optimization algorithm of the following formula (5). $$min\left(J=f\left(\mathrm{S.}\mathrm{S.}\mathrm{F.},\tau \right)\right)$$

The two parameters SSF and τ are determined here.

After step 220 reverses the procedure 100 to the third step 160 of the procedure 100 back, in which in turn the modeled output voltage U _{NCS, Sim of} the equivalent circuit is determined.

The modeled output voltage U _{NCS, Sim} is now determined with at least one varied parameter of the two parameters SSF and τ of the equivalent circuit, so that the value of the modeled output voltage U _{NCS, Sim} generally differs from the value of the previously determined modeled output voltage U _{NCS , Sim is} different.

The sixth step 220 , the third step 160 and the fourth step 180 are repeated iteratively until the measure J for the deviation of the modeled output voltage U _{NCS, Sim} from the measured output voltage U _{NCS, measurement of} the sensor is smaller than the specified variable J ₀ . In this case the procedure continues with step 240 gone and is thus over.

QUOTES INCLUDED IN THE DESCRIPTION

This list of the documents listed 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 102014204629 A1 [0008]

Claims (11)

Method (100) for determining at least one parameter (SSF, τ) of a piezo sensor (300) of a fuel injector, comprising the following steps: measuring (120) a profile of a pressure (p) inside the fuel injector; Measuring (140) an output voltage (U _{NCS, Mess} ) of the piezo sensor (300); Determining (160) a modeled output voltage (U _{NCS, Sim} ) of the piezo sensor (300); and determining (180) a measure (J) for a deviation of the modeled output voltage (U _{NCS, Sim} ) from the measured output voltage (U _{NCS, Mess} ) of the piezo sensor (300). Method (100) according to Claim 1 , further comprising: determining (200) whether the measure (J) for the deviation of the modeled output voltage (U _{NCS, Sim} ) from the measured output voltage of the piezo sensor (300) is smaller than a predetermined value (J ₀ ). Method (100) according to Claim 2 , further comprising: if, the measure (J) for the deviation of the modeled output voltage (U _{NCS, Sim} ) from the measured output voltage (U _{NCS, Mess} ) of the piezo sensor (300) is greater than the specified value (J ₀ ) : Reduce (220) the measure (J) for the deviation of the modeled output voltage (U _{NCS, Sim} ) from the measured output voltage (U _{NCS, Mess} ) of the piezo sensor (300) by adding at least one parameter (SSF, τ) of the Piezo sensor (300) is varied. Method (100) according to Claim 3 , characterized in that the method (100) returns to the step of determining (160) the modeled output voltage (U _{NCS, Sim} ) of the piezo sensor (300). Method (100) according to Claim 3 or 4th , characterized in that the steps of reducing (220) the measure (J) for the deviation of the modeled output voltage (U _{NCS, Sim} ) from the measured output voltage (U _{NCS, Mess} ) of the piezo sensor (300) by at least one Parameters (SSF, τ) of the piezo sensor (300) is varied; determining (160) the modeled output voltage (U _{NCS, Sim} ) of the piezo sensor (300); determining (180) the measure (J) for the deviation of the modeled output voltage (U _{NCS, Sim} ) from the measured output voltage (U _{NCS, Mess} ) of the piezo sensor (300); be carried out iteratively until the measure (J) for the deviation of the modeled output voltage (U _{NCS, Sim} ) from the measured output voltage (U _{NCS, Mess} ) of the piezo sensor (300) is smaller than the specified value (J ₀ ). Method (100) according to the preceding claim, characterized in that the at least one parameter (SSF, τ) is determined after the end of the iterative implementation. Method (100) according to Claim 3 , characterized in that the step of reducing (220) the measure (J) for the deviation of the modeled output voltage (U _{NCS, Sim} ) from the measured output voltage (U _{NCS, Mess} ) of the piezo sensor (300) by at least one Parameters (SSF, τ) of the piezo sensor (300) is varied, is realized by an optimization algorithm. Method (100) according to the preceding claim, characterized in that the optimization algorithm is implemented by a gradient method. Computer program which is set up, each step of the method (100) according to one of the Claims 1 to 8th perform. Machine-readable storage medium on which a computer program according to the preceding claim is stored. Electronic control device which is set up to carry out each step of the method (100) according to one of the Claims 1 to 8th perform.

Images