Monitoring apparatus and method for monitoring an AC voltage source, which is DC-decoupled from a reference potential

Abstract

A method for monitoring an AC voltage source, which is DC-decoupled from a reference potential and which generates an AC voltage between two AC voltage lines. Each of the AC voltage lines is coupled to the reference potential by way of a respective capacitive voltage divider and a respective component voltage signal is tapped at the voltage dividers and at least one root mean square value signal is generated therefrom and a check is carried out to determine whether the respective root mean square value signal meets a predetermined triggering criterion. When the triggering criterion is met, a fault signal is generated.

Claims

1. A method for monitoring an AC voltage source, which is DC-decoupled from a reference potential and which generates an AC voltage between two AC voltage lines, the method comprising: during operation of the AC voltage source, continuously coupling each of the AC voltage lines to the reference potential by way of a respective capacitive voltage divider such that the AC voltage source causes a respective current to flow between the AC voltage source and the reference potential on each of the AC voltage lines; tapping a respective component voltage signal at the voltage dividers; generating at least one root mean square value signal from the respective component voltage signal at the voltage dividers; carrying out a check to determine whether the respective root mean square value signal meets a predetermined triggering criterion; and when the triggering criterion is met, generating a fault signal and switching off the AC voltage source in response to the fault signal.

2. The method according to claim 1, which comprises generating a summation root mean square value signal from a sum of the component voltage signals.

3. The method according to claim 2, wherein, as triggering criterion, a check is carried out to determine whether a signal value of the summation root mean square value signal is located outside of a predetermined value interval, which describes predetermined rated voltage values of the AC voltage.

4. The method according to claim 1, which comprises generating a delta root mean square value signal from a difference of the component voltage signals.

5. The method according to claim 4, wherein, as triggering criterion, a check is carried out to determine whether a signal value of the delta root mean square value signal is located in a predetermined value interval.

6. The method according to claim 4, which comprises scaling a signal value of the delta root mean square value signal using a gain factor and, as a result, identifying an impedance value of a leakage impedance that is effective between the AC voltage lines and the reference potential.

7. The method according to claim 6, wherein, as triggering criterion, a check is carried out to determine whether the impedance value is lower than a predetermined triggering value.

8. The method according to claim 6, which comprises: identifying a gain factor K by way of a calibration and the calibration comprises; while a predetermined calibration impedance couples the AC voltage lines to the reference potential, identifying a signal value rms of the delta root mean square value signal and, on the basis of the signal value rms and an impedance value Zcal of the calibration impedance calculating the gain factor K as
K(Zcal)=Zcal*rms(Vd).

9. The method according to claim 8, wherein the calibration includes a DC test by checking, before and/or after determining the signal value of the delta root mean square value signal, whether or not a direct current is flowing between at least one of the AC voltage lines and the reference potential, and, if a direct current is flowing, terminating the calibration.

10. The method according to claim 8, wherein the calibration comprises, before and after determining the signal value of the delta root mean square value signal, identifying a root mean square voltage value of the AC voltage of the AC voltage source in each case, and terminating the calibration if a difference between the root mean square voltage values is greater than a predetermined tolerance value.

11. A monitoring apparatus for monitoring an AC voltage source, which is DC-decoupled from a reference potential and which generates an AC voltage between two AC voltage lines, the monitoring apparatus comprising: a respective capacitive voltage divider for each of the two AC voltage lines, said respective capacitive voltage dividers continuously capacitively coupling the AC voltage lines to the reference potential during operation of the AC voltage source; and a measuring circuit (M) for detecting component voltage signals at said voltage dividers, said monitoring apparatus being configured to carry out the method according to claim 1.

12. An apparatus, comprising: an AC voltage source, which is DC-decoupled from a reference potential and which is configured to generate an AC voltage between two AC voltage lines; and a monitoring apparatus coupled to the two AC voltage lines and to the reference potential, said monitoring apparatus including: a respective capacitive voltage divider for each of the two AC voltage lines, said respective capacitive voltage dividers continuously capacitively coupling the AC voltage lines to the reference potential during operation of the AC voltage source; and a measuring circuit (M) for detecting component voltage signals at said voltage dividers, said monitoring apparatus being configured to carry out the method according to claim 1.

13. The apparatus according to claim 12, configured as a smart-glass window arrangement or as a high-voltage on-board power supply system of a motor vehicle or as a hybrid drive of a motor vehicle or as a medical appliance.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING

(1) An exemplary embodiment of the invention is described below. In this regard, in the figures:

(2) FIG. 1 shows a schematic illustration of an embodiment of the apparatus according to the invention;

(3) FIG. 2 shows a graph having time profiles of a summation signal and of a difference signal;

(4) FIG. 3 shows a graph having signal values of a delta root mean square value signal and characteristic values formed therefrom;

(5) FIG. 4 shows a basic sketch for explaining the functioning of the apparatus of FIG. 1;

(6) FIG. 5 shows a flowchart to illustrate calibration for the monitoring apparatus; and

(7) FIG. 6 shows a flowchart to illustrate a method that can be carried out by a monitoring apparatus of the apparatus of FIG. 1.

DESCRIPTION OF THE INVENTION

(8) The exemplary embodiment explained below is a preferred embodiment of the invention. In the exemplary embodiment, the described components of the embodiment each represent individual features of the invention which are to be considered independently of one another and which each also develop the invention independently of one another and can therefore also be considered to be a component of the invention, either individually or in a combination other than that shown. Furthermore, further features of the invention which have already been described can also be added to the described embodiment.

(9) In the figures, functionally identical elements are in each case provided with the same reference designations.

(10) FIG. 1 shows an apparatus 10, in which said apparatus can be, for example, a smart-glass window arrangement. The apparatus 10 can be integrated in a motor vehicle, for example. The apparatus 10 can have a smart-glass panel 11, the transparency of which is prescribed by means of an AC voltage U.sub.SG or is set in a closed control loop even in multiple stages or without stages. The transparency of the smart glass can therefore also be set without stages or the transparency can be controlled. The AC voltage U.sub.SG is generated by an AC voltage source 12, which can be coupled to the panel 11 by means of 2 AC voltage lines 13, 14. The AC voltage U.sub.SG can have, for example, a root mean square value or a peak-to-peak value (Vpp) or more than 60 volts.

(11) The AC voltage source 12 and the AC voltage lines 13, 14 are electrically isolated or DC-decoupled from a reference potential 15 if the apparatus 10 is functioning as intended, that is to say has a predetermined setpoint state. As a result of this, each of the AC voltage lines 13, 14 has a voltage V1, V2 with respect to the reference potential 15, which is not fixed, that is to say can change freely during operation. The reference potential 15 can be, for example, a ground potential, such as can be fixed in a motor vehicle by way of the bodywork or design thereof, for example.

(12) In the event of a fault, a leakage impedance Z.sub.leak can be effective between the AC voltage lines 13, 14 one the one hand and the reference potential 15 on the other hand. In order to recognize or to detect this, a monitoring apparatus 16 is provided in the apparatus 10. The monitoring apparatus 16 illustrated in FIG. 1 is only an exemplary embodiment.

(13) The monitoring apparatus 16 can have a capacitive voltage divider 17, by means of which the AC voltage line 13 can be coupled to the reference potential 15. Furthermore, a capacitive voltage divider 18 can be provided, by means of which the AC voltage line 14 can be coupled to the reference potential 15. The leakage impedance Z.sub.leak can have a conductive portion as leakage resistance R.sub.leak and a capacitive portion as leakage capacitance C.sub.leak.

(14) A measurement voltage can be tapped at each of the capacitive voltage dividers 17, 18, as a result of which two component measurement signals V.sub.meas_1, V.sub.meas_2 (meas—measurement) are available. The component voltage signals V.sub.meas_1, V.sub.meas_2 are tapped between two series-connected capacitors C1, C2 and C3, C4, respectively, of the respective voltage divider 17, 18.

(15) The monitoring apparatus 16 can have a supply voltage source 19 for generating a supply voltage Vcc. Two voltage followers or buffers 20 of a measuring circuit M can cause decoupling of the voltage dividers 17, 18 from a remaining part of a measuring circuit M.

(16) The monitoring apparatus 16 can have, as a further part of the measuring circuit M, a summing unit 21 for forming a summation signal or, in short, a sum Vs from the component voltage signals V.sub.meas_1, V.sub.meas_2. Furthermore, a subtraction unit 22 can be provided, by means of which a difference signal or, in short, a difference Vd can be formed from the component voltage signals V.sub.meas_1, V.sub.meas_2.

(17) The summation signal Vs and the difference signal Vd can be received by a digital processor device 23 and, for example, be digitized. The processor device can have a microcontroller μC for this purpose. For example, a root mean square value or RMS value can be calculated in each case for different times from the signals Vs, Vd by the processor device 23. The root mean square value function is denoted in the following text as rms( ). Furthermore, the processor device 23 can carry out monitoring, as illustrated in FIG. 6, and calibration, as illustrated in FIG. 5.

(18) The monitoring apparatus 16 can have a leakage impedance measuring circuit M as a result. The terminal connections of the “floating” AC voltage source U.sub.SG are monitored using capacitive voltage dividers 17, 18 by virtue of the illustrated operational amplifiers of the voltage followers 20 of the measuring circuit M buffer-storing the component voltage signals V.sub.meas_1, V.sub.meas_2 of the voltage dividers 17, 18. The subtraction unit 22 forms the difference voltage or difference Vd; the summing unit 21 forms the summation voltage or sum Vs.

(19) The signals Vd, Vs can subsequently be evaluated in analog fashion, for example using an RMS detector and comparators, or, as illustrated, in digital fashion.

(20) The monitoring apparatus 16 can be designed, for example, for detection of a leakage impedance Z.sub.leak with parameter variations for the leakage impedance components R.sub.leak (200 kOhm to 4 MOhm) and C.sub.leak (10 pF to 200 pF).

(21) FIG. 2 shows signal profiles of the summation voltage Vs and of the difference voltage Vd for different values of R.sub.leak. The summation voltage Vs is directly proportional to the floating AC voltage U.sub.SG and therefore makes it possible to measure U.sub.SG precisely by means of the voltage dividers 17, 18. The difference voltage Vd is approximately proportional to the leakage impedance Z.sub.leak and therefore makes it possible to measure the leakage resistance R.sub.leak precisely, for example by forming the RMS value rms(Vd) as delta root mean square value signal.

(22) FIG. 3 shows the RMS values, that is to say the delta root mean square value signal rms(Vd), of the difference voltage Vd as a function of the leakage resistance R.sub.leak in the case of a varying leakage capacitance C.sub.leak and the virtually linear course of a leakage current indicator 26, defined as the reciprocal of the RMS value of the measured leakage voltage, that is to say 1/rms(Vd). As FIG. 3 shows, the variation 24 of the leakage capacitance C.sub.leak does not have a significant influence.

(23) By multiplying the leakage current indicator 1/rms(Vd) by a calibration value, it is possible to identify the present value of the leakage impedance Z.sub.leak with a predeterminable accuracy for a predetermined interval of values of the leakage resistance R.sub.leak and the leakage capacitance C.sub.leak. Z.sub.leak can then be compared with a threshold value Z0 and, when the threshold value Z0 is exceeded, a fault signal 25 can be generated (see FIG. 1).

(24) To this end, in the following text, the principle functioning of the monitoring apparatus 16 is explained with reference to FIG. 4.

(25) By using two capacitive voltage dividers 17, 18, two AC measurement voltages, that is to say the component voltage signals V.sub.meas_1 and V.sub.meas_2, can be formed.

(26) It can be shown that, using the circuit from FIG. 4, the RMS value rms(Vd) of the vector difference Vd=V_meas_2−V_meas_1 supplies information about the leakage impedance and the vector sum Vs=V_meas_2+V_meas_1 is directly proportional to U.sub.SG.

(27) It is therefore possible to recognize a possibly present leakage current, to measure the leakage impedance Z.sub.leak precisely and to measure the amplitude of the floating AC voltage U.sub.SG,
while the measuring circuit of the monitoring apparatus 16 always remains DC-isolated (through the use of the capacitors C1-C4).

(28) On the basis of the measuring circuit of FIG. 4, the following analytical observation can be made and the described methods for measuring the isolation impedance and leakage impedance Z.sub.leak can be derived.

(29) The impedances of the capacitors can be calculated as

(30) $\begin{array}{cc}{Z}_C\left(C,f\right)=\frac{1}{2.Math.\pi .Math.f.Math.j.Math.C}& \mathrm{Eq}.1\end{array}$
wherein the symbol .Math. represents multiplication. For the further calculations, an auxiliary function is required, which defines the impedance of the RC parallel circuits
Z.sub.rc(C,R,f)=Z.sub.p(Z.sub.C(C,f),R)  Eq. 2

(31) The current I.sub.meas flowing to the proposed measuring circuit and illustrated in FIG. 4 is therefore equal to

(32) $\begin{array}{cc}{I}_{\mathrm{meas}}=\frac{U_{\mathrm{SG}}}{\begin{array}{c}{Z}_p\left[Z_{\mathrm{leak}}\left(R_1,C_1\right),\left(Z_C\left(C_1,f\right)+Z_{\mathrm{rc}}\left(C_2,R_1,f\right)\right)\right]+\\ Z_{\mathrm{rc}}\left(C_4,R_2,f\right)+Z_C\left(C_3,f\right)\end{array}}& \mathrm{Eq}.3\end{array}$

(33) The individual voltages at the DC-isolated AC voltage source 12 are calculated as
V.sub.2=I.sub.meas.Math.(Z.sub.rc(C4,R2,f))+Z.sub.C(C3,f)  Eq. 4
V.sub.1:=U.sub.AC−V.sub.2  Eq. 5

(34) The leakage current through the (possibly present) leakage impedance Z.sub.leak is therefore

(35) $\begin{array}{cc}{I}_{Z_{\mathrm{leak}}}\approx \frac{V_1}{Z_{\mathrm{leak}}\left(R_{\mathrm{leak}},C_{\mathrm{leak}}\right)}& \mathrm{Eq}.6\end{array}$

(36) Assuming some exemplary values for certain component parts, the proposed circuit can be analyzed in a practical manner. The following table indicates these exemplary values for the components of the proposed circuit in FIG. 4:

(37) TABLE-US-00001 Voltage U.sub.SG = 100 V f ≈ 100 Hz source Capacitors C1 ≈ 4.7 nF C2 ≈ 220 nF C3 ≈ C1 C4 C2 Tolerances TOL.sub.C1 ≈ TOL.sub.C2 ≈ 10% TOL.sub.C3 ≈ TOL.sub.C4 ≈ of the 2% TOL_C1 TOL_C2 capacitors Resistors R1 ≈ R2 ≈ R1 110 kΩ Tolerances TOL.sub.R1 ≈ TOL_R2 ≈ of the 2% TOL_R1 resistors Test R.sub.leak ≈ C.sub.leak ≈ parameters 300 kΩ 0.05 nF R.sub.leak, C.sub.leak

(38) With the aid of the above exemplary values, it is possible to calculate the leakage current as a function of Z.sub.leak. The result is the leakage current indicator 26 of FIG. 3. In this case, in the variation 24, the leakage capacitance C.sub.leak of 5 pF to 500 pF (with nominally 50 pF) has been varied in order to also investigate the influence thereof. The virtually linear characteristic curves of the leakage current indicator 26 of FIG. 3 represents a leakage impedance estimation.

(39) The output voltages of the measuring circuit can be derived as follows
V.sub.meas_2:=V.sub.2−I.sub.meas.Math.Z.sub.C(C3,f)  Eq. 7
V.sub.meas_2:=I.sub.meas.Math.Z.sub.rc(C4,R,f)  Eq. 8

(40) Furthermore, the current through C1 results from:
I.sub.C1:=I.sub.meas.Math.I.sub.Z_leak  Eq. 9
V.sub.meas_1:=V.sub.1−I.sub.C1.Math.Z.sub.C(C1,f)  Eq. 10

(41) The equations for the leakage current indicator and the amplitude of the isolated voltage source U.sub.SG can therefore be defined:
V.sub.Δ.sub.meas(U.sub.SG,R.sub.leak,C.sub.leak,f):=V.sub.meas.sub.2−V.sub.meas_1  Eq. 11
V.sub.Σ_meas(U.sub.SG,R.sub.leak,C.sub.leak,f):=V.sub.meas.sub.2+V.sub.meas.sub.1  Eq. 12

(42) This produces the summation voltage Vs and the difference voltages Vd of eq. 11 and eq. 12 illustrated in the graph of FIG. 2. The summation voltage Vs=V.sub.Σ_meas is directly proportional to the amplitude of the floating AC voltage U.sub.SG. The difference voltage Vs=V.sub.Δ_meas in the case of different leakage resistances R.sub.leak produces information about the leakage impedance Z.sub.leak.

(43) It can be assumed and proven that the difference voltage Vd=V.sub.Δ_meas formed in this way is in wide ranges virtually inversely proportional to the leakage impedance:

(44) $\begin{array}{cc}{Z}_{\mathrm{leak}}=K\left(Z_{\mathrm{leak}}\right).Math.\frac{1}{V_{\mathrm{\Delta \_}\mathrm{meas}}}& \mathrm{Eq}.13\end{array}$
wherein K is a proportionality factor. The still present non-linearity in the case of low values of the leakage impedance is depicted by a variable coefficient K(Z.sub.leak).

(45) It is now possible to define a transfer function, which represents the ratio (analogously an amplification or a gain) between the actual leakage impedance Z.sub.leak and the measured leakage impedance indicator:

(46) $\begin{array}{cc}\mathrm{Gain}\approx \frac{1}{.Math.V_{\mathrm{\Delta \_}\mathrm{meas}}.Math..Math.\mathrm{Z\_leak}}.Math.K\left(Z_{\mathrm{leak}}\right)& \mathrm{Eq}.14\end{array}$

(47) A calibration method, as is explained below in connection with FIG. 5, can be applied to identify a matching K(Z.sub.leak) in practice so that the aforementioned transfer function becomes approximately equal to one through the suitable selection of K(Z.sub.leak) in the range of interest. The measurement error of Z.sub.leak in the relevant measurement range (for example R.sub.leak of 180 kOhm to 220 kOhm) is therefore minimized.

(48) The purpose of the calibration is therefore to find, according to equation 14, a scaling factor or gain factor K(Z.sub.cal), which, in the relevant measurement range of Z.sub.cal, determines a gain factor of equal to one (gain=1):
K(Z.sub.cat)=1.Math.|V.sub.Δ_Meas_cal|.Math.Z.sub.cal  Eq. 15

(49) FIG. 5 illustrates a flowchart for calibration 27.

(50) In a calibration step K1, the calibration can be triggered, for example, by a calibration signal (START CAL). In a calibration step K2, it is possible to check whether a direct current DC is flowing between the AC voltage lines 13, 14 on the one hand and the reference potential 15 on the other hand. If this is the case (symbolized in FIG. 5 by +), in a calibration step K3, the calibration 27 can be terminated, since it is recognized that the apparatus 10 is not in order (NOK—not OK), that is to say does not have DC-decoupled AC voltage lines 13, 14 as intended.

(51) In the case of a lack of direct current DC (symbolized in FIG. 5 by −), in a calibration step K4, a first root mean square value Vs1 of a summation root mean square value signal rms(Vs) of the summation signal Vs can be generated.

(52) In a calibration step K5, a leakage impedance as a calibration impedance having a known calibrated impedance value Z.sub.cal can be connected in a targeted manner between the AC voltage lines 13, 14 on the one hand and the reference potential 15 on the other hand. The impedance value Z.sub.cal is preferably located in an interval of impedance values, which represent those in the relevant measurement range. In particular, the impedance value corresponds to said threshold value Z0.

(53) In the case of an interconnected calibration impedance, in a calibration step K6, a second root mean square value Vs2=rms(Vs) of the summation signal Vs can be generated. The root mean square values Vs1, Vs2 can be formed as single measurements or respectively, for example, as an average value from a plurality of measurements.

(54) In a calibration step K7, a root mean square value rms(Vd) of the difference signal Vd can be identified.

(55) If it is recognized in a calibration step K8 that the root mean square values Vs1 and Vs2 are not equal (symbolized in FIG. 5 by −), the calibration 27 can be terminated in accordance with calibration step K3. In this case, it is recognized that the AC voltage source 12 generates the AC voltage U.sub.SG with a floating root mean square value.

(56) If the root mean square values Vs1, Vs2 are equal or at least the magnitude of their difference is lower than a predetermined maximum value (symbolized in FIG. 5 by +), the gain factor K(Z.sub.cal) can be calculated in accordance with equation eq. 15 in a calibration step K9.

(57) After calibration has taken place, the actual measurement process can be carried out by the monitoring apparatus 16 periodically or continuously according to the monitoring 28 illustrated in FIG. 6.

(58) The monitoring apparatus 16 can be started, for example, by a start signal START in a monitoring step O1. In a monitoring step O2, the gain factor K(Z.sub.cal) can be provided, for example it can be loaded from a data store. In a monitoring step O3, a present signal value of the difference signal Vd can be identified. In a monitoring step O4, a present signal value of the root mean square value signal rms(Vd) can be identified. In a monitoring step O5, an impedance value of the leakage impedance Z.sub.leak can be calculated or estimated on the basis of the gain factor K(Z.sub.cal) as
Z.sub.leak=K(Z.sub.cal)/rms(Vd).

(59) In a monitoring step O6, it is possible to check whether the estimated leakage impedance is lower than the threshold value Z0. In this case, Z0 represents a triggering criterion. The comparison can be carried out, for example, as a comparison of the magnitudes of the impedances. If the estimated leakage impedance is lower than the threshold value (symbolized in FIG. 6 by +), the fault signal 25 can be generated in a monitoring step O7. The AC voltage source 12 can then be switched off depending on the fault signal 25, for example.

(60) If the estimated leakage impedance is not lower than the threshold value (symbolized in FIG. 6 by −), the monitoring 28 can be resumed with monitoring step O3.

(61) When dimensioning the measuring circuit of FIG. 1 and FIG. 3, and when implementing the monitoring 28, whether it be analog or digital, it should be taken into account that the leakage current indicator 26 is dependent on frequency due to the reactance of the capacitors C1-C4 in the voltage dividers 17, 18. If the AC voltage source 12 that is to be monitored should be operated at different frequencies, this must be taken into account.

(62) A tolerance analysis (not illustrated in more detail here) shows that a worst-case tolerance of the leakage indicator of lower than 4 percent is produced at a leakage resistance R.sub.leak=200 kOhm using the circuit of FIG. 1 and FIG. 3. An average tolerance of the leakage indicator is 2.6 percent given R.sub.leak=200 kOhm.

(63) Overall, the apparatus 10 has, in particular, the following advantages: DC-isolated, separate measuring circuit (owing to the use of capacitors), cost-effective solution, managed without an expensive isolation amplifier, high-resolution, sensitive owing to the introduction of a linear, steep leakage current indicator function, precise measurement of the leakage impedance after calibration, the reactive component of the leakage impedance (the capacitance C.sub.leak) can also be detected, the amplitude of the AC voltage U.sub.SG is also measured, the signals can be evaluated in analog or digital fashion, DC-neutral circuit, a standard isolation test (which is carried out using DC voltage), for example in the production and manufacturing or else during maintenance, is therefore easily possible, without the appliance having to be activated.

(64) Overall, the example therefore shows how the invention can provide a method and a circuit for identifying the leakage impedance of an isolated AC voltage source.

LIST OF REFERENCE NUMERALS

(65) 10 Apparatus 11 Smart-glass panel 12 AC voltage source 13 AC voltage line 14 AC voltage line 15 Reference potential 16 Monitoring apparatus 17 Voltage divider 18 Voltage divider 19 Supply voltage source 20 Voltage follower 21 Summing unit 22 Subtraction unit 23 Processor device 24 Variation 25 Fault signal 26 Leakage current indicator 27 Calibration 28 Monitoring