EP0979556A1

EP0979556A1 - Method and device for digitizing a high-dynamic analog measuring signal

Info

Publication number: EP0979556A1
Application number: EP98931923A
Authority: EP
Inventors: Stephan Mohr; Stefan Hain; Thomas Bosselmann
Original assignee: Siemens AG
Current assignee: Siemens AG
Priority date: 1997-04-28
Filing date: 1998-04-15
Publication date: 2000-02-16
Also published as: JP2001522550A; WO1998049776A1

Abstract

The invention relates to a method or a device for digitizing an analog measuring signal (M). According to the invention, the measuring signal (M) is scaled by means of multiplication by a scaling factor (A(ti)), and the signal amplitudes of the scaled measuring signal (M'(ti)) are converted at successive sampling instants (ti) into a data word (D(ti)), by an analog to digital converter. The data word (D(ti)) is regulated by setting the scaling factor (A(ti)) to a predetermined data word setpoint (DS) from the control range of the analog to digital converter (3) at each sampling instant (ti).

Description

description

Method and device for digitizing an analog measurement signal with high dynamics

The invention relates to a method and a device for digitizing an analog measurement signal. An analog measurement signal is understood to mean a measurement signal whose amplitude is proportional to the measured size (amplitude-analog measurement signal).

In some areas of measurement technology, analog measurement signals with high dynamics occur, which must be recorded and digitally transmitted to avoid interference, such as electromagnetic interference. For example, new types of inductive converters for measuring nominal currents and overcurrents in a high-voltage power conductor (e.g. overhead line) generate analog current measurement signals with a dynamic range of 1: 2000.

For digital transmission, the analog measurement signals are first converted into a transmission data word of a word length (number of bits per word) predetermined by the digital resolution of the ADC, for example a 12-bit data word, in an analog-digital converter (ADC). Via a digital transmission path (data bus, information channel) of a corresponding bit width, the transmission data word is transmitted to an evaluation unit and further processed there, for example in a digital-to-analog converter (DAC) of the same digital resolution as the ADC, or converted back into an analog measurement signal also processed and / or output as a digital measured value.

On the digital transmission channel for the measurement signals, which comprises the ADC and the transmission link, absolute errors in particular, such as quantization errors in particular, limit the relative accuracy that can be achieved. The relative over The smaller the quantization error on the one hand and the higher the transmission channel is controlled, the smaller the transmission error. The upper limit control of the transmission channel, in particular the ADC, limits the achievable dynamics for a given absolute error. The dynamics of the digital measurement signal transmission can now be increased by reducing the absolute error of the transmission channel by using a higher-resolution ADC (with a larger number of bits of the data words). However, this increases the complexity of the circuitry and, above all, the energy consumption.

The relative accuracy of a described digital transmission channel is highest in the upper modulation range of the transmission channel, which corresponds to the upper measuring range of the analog measuring signal. However, the situation is exactly the opposite when it comes to measuring current in energy technology. There a high relative accuracy of typically 0.1% in the nominal current range, i.e. in the lower current measuring range, and a significantly lower accuracy of typically 5% in the overcurrent and short-circuit current range, i.e. in the upper measuring range, and at the same time a high dynamic range is required to detect both nominal currents and the significantly higher overcurrents.

To expand the dynamics of a digital measurement signal transmission, the analog measurement signal can be scaled with a few discrete scaling factors, corresponding to a measurement range selection for an ampere / voltmeter, so that the transmission channel is controlled within predetermined limits.

In this way, several analog-to-digital converters, each with an upstream amplifier, can be connected in parallel, the amplifiers having different fixed gain factors in pairs. The measurement signal is applied to the inputs of the amplifiers, and the output data word of the analog-digital converter that is processed by the associated , ß ß -.

T φ 0 »φ φ -H rö ß rö ß φ -H ε ra rö x! 4- ⁾ <tf ß CQ ε rö Φ ß Φ φ & Di P

©

00 ov

ofo

speed. Compared to a device and a method with only a few discrete scaling factors, the analog / digital converter in the system according to the invention can be optimally controlled at any time over a large dynamic range, regardless of the input signal.

Advantageous refinements and developments of the method and the device according to the invention result from the claims dependent on claims 1 and 12, respectively.

To further explain the invention, reference is made to the drawings, in which exemplary embodiments according to the invention are each shown schematically. 1 shows a device for digitizing and digitally transmitting an analog measurement signal in a basic structure, FIG. 2 shows a device for digitizing and digital

Transmission of analog measurement signals from a sensor head to a ground station and FIG. 3 a scaler with a multiplying digital-to-analog converter. Corresponding parts are provided with the same reference numerals in FIGS. 1 and 2.

The digital transmission system according to FIG 1 comprises a scaler 2, an analog-digital converter 3, a control device 4, a deskaler 5 and a digital-analog converter 6. The analog measurement signal M to be transmitted is present at an input 21 of the scaler 2 . This analog measurement signal M is a measure of a measurement variable measured by a sensor, for example an electrical current or an electrical voltage or a temperature or a pressure, and is in usually a band-limited and generally also continuous signal from a given measuring range. Even with a relatively wide measuring range for the measuring signal M and a correspondingly high signal dynamic, the measuring signal M should be digitized with a high relative accuracy and preferably be transmitted thereafter. This is achieved with the following measures.

A variable scaling factor A, which is a function A (t) of the time t, is supplied to the scaler 2 by the control device 4. With this scaling factor A, the measurement signal M from the scaler 2 is now multiplied (scaled). The correspondingly scaled measurement signal M '= A -M can be tapped at an output 24 of the scaler 2 and is fed from there to an input 31 of the analog-digital converter 3. Before

Input 21 of the scaler 2 can be switched to suppress interference, an anti-aliasing filter, not shown. The analog-digital converter 3 samples the scaled measurement signal M 'at sampling times ti with the natural number i as a counting index and forms the sampled ones

Signal amplitudes M (ti) of the scaled measurement signal M 'each have corresponding data words (hereinafter referred to as transmission data words) D (ti) which are present at an output 32 of the analog-digital converter 3 until the next sampling time ti + 1 (digitizing or quantizing the analog scaled measurement signal M '). These transmission data words D (ti) are firstly supplied to the control device 4 via a data bus 11. The control device 4 stores the transmission data words D (ti) with the associated sampling times ti in a memory 41 and forms a tten, current sampling time tn from a part and at least two of the last stored transmission data words D (ti) and the associated sampling times ti with i <n the scaling factor A (tn) at the current sampling time tn for the scaler 2.

In one embodiment, the current scaling factor A (tn) can be calculated as the ratio of an extrapolated prospective transmission data word PD (tn) for the considered sampling time tn and a predetermined data word setpoint DS for the transmission data words D (ti) and D (tn ) be calculated. The prospective transmission data word PD (tn) can be calculated by means of an extrapolation method known per se from at least two of the earlier stored value pairs (D (tn-1), tn-1), (D (tn-2), tn-2), in particular by linear extrapolation (eg gradient method) or also extrapolations with higher derivatives and also taking into account the differential equation of the anti-aliasing filter mentioned.

In another embodiment, the signal amplitudes M (ti) of the analog measurement signal M at the previous sampling times ti with i <n are determined from the known associated scaling factors A (ti) and the known transmission data words D (ti) and it is determined from these signal amplitudes M (ti) predicts a prospective measurement signal PM (tn) at a current sampling time tn. The current scaling factor A (tn) is now determined so that the product

A (tn) -PM (tn) is mapped by the analog-digital converter 3 to the data word setpoint DS. To determine the prospective measurement signal PM (tn), the time profile of the measurement signal M or the scaled measurement signal M 'can also be tracked and evaluated via an analog circuit, for example a delay line (delay circuit). Such an analog time evaluation can in particular also be used in combination with the described digital evaluation to protect the analog-digital converter 3 from being overloaded by transient interference signals. The additional analog signal line 7 shown in FIG. 1 is then provided, via which the control device 4 is supplied with the analog measurement signal M. The control device 4 then contains an analog differentiator for determining the signal change and a downstream comparator for comparing this signal change with a maximum permissible change and sets a predetermined maximum or minimum value for the prospective measurement signal PM or the prospective transmission data word D instead of the respectively calculated prospective value, if the predicted prospective value exceeds a predetermined limit due to the high signal gradient.

Transient interference signals, in particular at higher frequencies (e.g. interference edges or other discontinuous signal changes) can also be suppressed by a band limitation by means of a low pass, for example the already mentioned anti-aliasing low pass, in front of the input 21 of the scaler 2.

The data word setpoint DS is in the upper modulation range and preferably as close as possible to the outer limit (edge) of the modulation range (value range for the transmission data Words D (ti)) of the analog-digital converter 3 are selected, a small distance between the data word setpoint DS and this control limit being advantageous in order to take possible transients into account. With a resolving power of p bits / word of the analog-digital converter 3, the data word setpoint DS is, for example, in a range between a corresponding numerical value 2 ^{P ~ 1} and the numerical value 2 ^P -1 with maximum control of the analog-digital converter 3 for the transmission data words D (ti). A consistently high relative accuracy of the digitization is achieved over the entire measuring range of the analog measuring signal. In the case of a bipolar analog measurement signal M, two data word setpoints, for example + DS and -DS, can be taken into account, or with only one data word setpoint DS the amount or an effective value of an original measurement signal can be used as the measurement signal M or by selecting the sign of the scaling factor A ( ti) the scaled measurement signal M 'is always kept at one polarity (a sign).

The bit width (word length) of the transmission data words (D (ti)) determined by the digital resolution capacity of the analog-digital converter 3 is generally at least 10 bits, preferably at least 12 bits, and can also be higher, for example 16 bits.

The scaling factor A (ti) is preferably a binary data word. The number of bits m of the digital scaling factor A (ti) determines the fineness of the scaling, since the number 2 ^{m of} the individual numerical values which the scaling factor A (ti) can assume naturally increases considerably with increasing number of bits m. The bit number m of the scaling factor A (ti) is in generally not less than 8, in particular at least 10 bits and preferably at least 12 bits or 16 bits. With the correspondingly large number 2 8, 210, 2 ¹² or 2 ¹ of adjustable values of the scaling factor A (ti), a practically continuous (quasi-continuous) scaling of the measurement signal M in the scaler 2 and a correspondingly practically continuous control of the output signal (the Transmission data words D (ti)) of the analog-digital converter 3 to the data word setpoint DS.

Such continuous control can also be achieved in an embodiment not shown with an analog scaling factor A and an associated analog controllable scaler, for example a controllable operational amplifier.

The transmission data words D (ti) at the sampling times ti, in particular the current transmission data word D (tn), can now be fed directly to signal processing or, as shown in FIG. 1, can be transmitted to the deskaler 5 via a digital transmission path 10. The deskaler 5 multiplies each transmitted transmission data word D (ti) or D (tn) by the reciprocal value l / A (ti) or 1 / A (tn) of the scaling factor A (ti) or A (tn) and reconstructs it thereby the original measurement signal value M (tn) at the current sampling time tn in the form of a measurement data word (digital measurement signal value), which is denoted by E (ti) or E (tn). This measurement data word E (ti) or E (tn) can now be converted with an (or more parallel-connected) digital-to-analog converter 6 into an analog evaluation signal M ″ whose signal amplitude M ″ (ti) Signal amplitude M (ti) of the original analog measurement signal M corresponds.

2 shows an embodiment of a device for digitally transmitting an analog electrical measurement signal for an electrical current or an electrical voltage in a high-voltage current conductor, not shown, for example an energy transmission network (overhead line or the like). Such measurement signals generally have a limited bandwidth because of the properties of the network. The device is partially integrated in a sensor head 30, which is at the high voltage potential and additionally a current sensor, not shown, for example an inductive converter, a shunt sensor or an optical Faraday sensor, or voltage converter, in particular an optical Pockels sensor. Sensor, which provides the measurement signal M, if necessary with the help of an optoelectric converter in electrical form. The scaler 2 contains a multiplying digital-to-analog converter 20 with the measurement signal M as an analog input signal at an input 21. At a digital input 23 of the multiplying digital-to-analog converter 20 there is the scaling factor A (ti) as a digital word (binary number) with a sufficiently high number of bits as in the embodiment according to FIG. 1, so that a practically continuous change in the scaling factor A (ti) is possible. This digital scaling factor A is converted with the analog measurement signal M as a multiplier into an output signal M 'as a scaled measurement signal which is present at the output 24 of the scaler 2. An operational amplifier for further (fixed) amplification of the scaled measurement signal M 'is preferably connected between the multiplying digital-to-analog converter 20 and the output 24 of the scaler 2.

The scaled measurement signal M 'is fed from the output 24 of the scaler 2 to the input 31 of the analog-digital converter 3. The analog-digital converter 3 scans (as in FIG. 1) the measurement signal M 'regularly at sampling times ti and converts the corresponding measurement signal amplitude M' (ti), which can be positive or negative, into a corresponding digital transmission data word D (ti) with a predetermined resolution.

A control device 8 is arranged in a ground station 40 which is at ground potential and is connected to the sensor head 30 via a bidirectional digital transmission path 12. The digital transmission path 12 branches off in the sensor head 30 into a unidirectional digital transmission path 13 and a further, bidirectional digital transmission path 14. The data word of the scaling factor A (ti) is transmitted from the control device 8 to the multiplying digital-to-analog converter 20 via the digital transmission path 12 and then the digital transmission path 13. The sampling times ti in the form of digital values are transmitted from the control device 8 to the analog-digital converter 3 via the bidirectional digital transmission paths 12 and 14 and the transmission data words D (ti) are transmitted from the analog-digital converter 3 to the control device 8. The control device 8 comprises a microprocessor which, in addition to determining the scaling factors A (ti), also specifies the sampling times ti (sampling clock) for the analog / digital converter 3 and provides the sequence control for the sensor head 30 and forms an interface to the periphery. Furthermore, the microprocessor also takes over the function of the digital signal reconstruction of the deskaler 5 according to FIG. 1 and for this purpose forms the quotients E (ti) = D (ti) / A (ti) from the received transmission data words D (ti) by the associated scaling factor A. (ti).

The downstream digital-to-analog converter 6, which receives the measurement data words E (ti) from the control device 8 via the data line 15, converts the measurement data words E (ti) back into the signal amplitudes M '' (ti) of an analog measurement signal M ''. A higher resolution than the analog-digital converter 3 can also be provided for this conversion, for example have at least a 16-bit resolution to achieve greater accuracy. Furthermore, instead of the one digital-to-analog converter 6, a plurality of digital-to-analog converters can also be provided, for example a converter for the nominal current range and a converter for the protective range (overcurrents).

If a digital interface is provided, the measurement data words E (ti) can also be used directly as digital measurement signals without D / A conversion.

As a multiplying digital-to-analog converter 20, all known embodiments can be used, for example according to the book Tietze, Schenk, "semiconductor switching technology", 6th edition, Springer publishing house, 1983, Be th 742 and 743 and 750 and 751, the content of which is incorporated in the disclosure of the present application, or the commercially available components AD 7943 / AD7945 / AD7948 from the company Analog Devices, these known multiplying digital-to-analog converters comprise a digitally controllable current divider network, in particular an R-2R Resistor network, and a downstream operational amplifier.

A particularly advantageous embodiment of a multiplying digital-to-analog converter is shown in FIG. 3 in an equivalent circuit diagram. An operational amplifier 50 with an inverting input 54 and a non-inverting input 53 and an output 55 is provided. Furthermore, a current divider network 57 is shown with an input resistor 56 and two digitally controllable subnetworks 51 and 52, which are shown in the equivalent circuit diagram as controllable resistors. The subnetworks 51 and 52 can, for example, each comprise an R / 2R resistance network with assigned switches, the switch positions of which are determined by the individual bits of the digital control data word. The output 55 of the operational amplifier 50 is via the sub-network 52 with the non-inverting input 53 and

Claims

claims

1. Method for digitizing an analog measurement signal (M), in which a) the analog measurement signal (M) is scaled by multiplication by a scaling factor (A (ti)), b) the signal amplitudes of the scaled analog measurement signal

(M '(ti)) at successive sampling times (ti) are each converted by an analog-digital converter (3) into a data word (D (ti)), c) the data word (D (ti)) of the analog -Digital converter (3) by setting the scaling factor (A (ti)) at each sampling time (ti) to a predetermined data word setpoint (DS) from the modulation range of the analog-digital converter (3) is regulated, d) the Data words (D (ti)) with the associated sampling times (ti) are continuously stored, e) before an upcoming, current sampling time (tn) from at least two earlier, stored sampling times (tn-1, tn-2, etc.) and the associated stored data words (D (tn-1), D (tn-2), etc.) a prospective data word (P (tn)) or a prospective scaled measurement signal (PM '(tn)) at the current sampling time (tn ) is determined.

2. The method of claim 1, wherein the scaling factor (A (tn)) for the current sampling time (tn) is calculated as a quotient from the associated prospective data word (P (tn)) and the predetermined data word setpoint (DS).

3. The method of claim 1, wherein the scaling factor (A (tn)) for the current sampling time (tn) is determined such that the product of the prospective scaled measurement signal (PM '(tn)) and the scaling factor (A (tn )) would be converted by the analog-digital converter (3) into the specified data word setpoint (DS).

4. The method according to any one of the preceding claims, wherein said data word setpoint (DS) is in the outer digital modulation range of the analog-digital converter (3).

5. The method according to any one of the preceding claims, in which the prospective data word (P (tn)) or the prospective scaled measurement signal (PM '(tn)) at the current sampling time (tn) from the two last stored data words (D (tn- l), D (tn-2)) of the two previous sampling times (tn-1, tn-2) is determined by linear extrapolation.

6. The method according to any one of the preceding claims, in which a multiplying digital-to-analog converter (20) is used for scaling the analog measurement signal (M), wherein at a digital input (23) of the multiplying digital-to-analog converter ( 20) the scaling factor (A (ti)) as a digital data word and to an analog input (21) of the multiplying digital-to-analog converter (20) the analog measurement signal (M) is applied and to an analog output (22) of the multiplying digital -Analog converter (20) the scaled analog measurement signal (M ') is present.

7. The method according to any one of the preceding claims, in which a) the data words (D (ti)) are transmitted via a digital transmission path (4), b) a measurement data word (E (ti)) belonging to a sampling time (ti)) as the quotient of the transmitted data word (D (ti)) at this sampling time (ti) and for this

Sampling time (ti) set scaling factor (A (ti)) is formed.

8. The method according to claim 7, wherein the associated measurement data word (E (tn)) at each sampling time (ti) is converted back into an analog electrical evaluation signal (M '') that corresponds to the original signal amplitude (M (ti)) of the analog measurement signal (M) is proportional to this sampling time (ti).

9. The method according to claim 8, wherein the digital-to-analog conversion of the measurement data words (E (ti)) takes place with a higher digital resolution than the analog-to-digital conversion of the scaled measurement signal (M ').

10. The method according to any one of the preceding claims, wherein the analog measurement signal (M) is a measure of an electrical current or an electrical voltage in a current conductor, in particular a high-voltage power distribution network.

11. The method according to claim 10, wherein at least partially an optical transmission path is used as the digital transmission path (4).

12. Device for digitizing an analog measurement signal (M) with a) a controllable scaler (2.20) with an input (21) for the analog measurement signal (M) and with an output (24), at which an output signal with a Scaling factor (A (ti)) multiplied analog measurement signal (M ') is present, b) an analog-to-digital converter (3) which converts the scaled measurement signal (M') from the scaler (2) into consecutive sampling times (ti) Converts data word (D (ti)), c) a control device (4,8) connected to the scaler (2,20) and the analog-digital converter (3), which controls the data word (D (ti)) Analog-digital converter (3) by setting the scaling factor (A (ti)) at each sampling time (ti) to a predetermined data word setpoint (DS) from the modulation range of the analog-digital converter (3), which controls continuously stores the data words (D (ti)) with the associated sampling times (ti) and those before an upcoming, current sampling time itpunkt (tn) from at least two earlier, stored sampling times (tn-1, tn-2, etc.) and the associated stored data words (D (tn-1), D (tn-2), etc.) is a prospective one Data word (P (tn)) or a prospective scaled measurement signal (PM '(tn)) is calculated in advance at the current sampling time (tn).

13. The apparatus of claim 12, wherein the control device (4,8) calculates the quotient from the prospective data word (P (tn)) for the current sampling time (tn) and the predetermined data word setpoint (DS) and the scaler (2 , 20) as a scaling factor (A (tn)) for this current sampling time (tn).

14. The apparatus of claim 12, wherein the control device (4) from the condition that the product of the scaling factor (A (tn)) and the prospective scaled measurement signal (PM '(tn)) at the current sampling time (ti) the data word setpoint (DS) is assigned as a data word by the analog-digital converter (3), the scaling factor (A (tn)) of the scaler (2.20) is determined for the current sampling time (tn).

15. The apparatus of claim 13 or claim 14, wherein the control device (4,8) the prospective data word (P (tn)) or the prospective scaled measurement signal (PM '(tn)) at the current sampling time (tn) from the two data words last stored (D (tn-1), D (tn-2)) of the two previous sampling times (tn-1, tn-2) determined by linear extrapolation.

16. The device according to one of claims 12 to 15, wherein the data word setpoint (DS) is in the outer modulation range of the analog-digital converter (3).

17. The device according to one of claims 12 to 16, wherein the scaler (2) comprises a multiplying digital-to-analog converter (20), the scaling factor being at a digital input (23) of the multiplying digital-to-analog converter (20) (A (ti)) as a digital data word and to an analog input (21) of the multiplying digital-analog Converter (20) the analog measurement signal (M) are applied and at an analog output (22) of the multiplying digital-to-analog converter (20) the analog measurement signal (M ') multiplied by the scaling factor (A (ti)) is present.

18. The apparatus of claim 17, wherein the multiplying digital-to-analog converter uses a scaling factor

(A (ti)) comprises a digitally controllable current divider network (57) and an operational amplifier (50) which is coupled back via the current divider network (57).

19. Device according to one of claims 12 to 18 with a) one of the analog-to-digital converter (3) downstream digital transmission path (10) for transmitting the data words (D (ti)) and e) a deskalizer (5) Form a measurement data word (E (ti)) belonging to a sampling time (ti) as the quotient of the data word (D (ti)) belonging to this sampling time (ti) by the scaling factor (A (ti)) at this sampling time (ti) .

20. The apparatus of claim 19, wherein the digital transmission link (4) is at least partially formed with an optical transmission link.

21. The apparatus of claim 19 or claim 20 with at least one digital-to-analog converter (6) which converts the measurement data word (E (ti)) back into an analog electrical evaluation signal (M '') which corresponds to the signal amplitude (M ( ti) of the original analog measurement signal (M) is proportional to the associated sampling time (ti).

22. The apparatus of claim 21, wherein the digital-to-analog converter (6) has a higher digital resolution than the analog-to-digital converter (3).

23. The device according to one of claims 12 to 22, wherein the control device (4,5,8) comprises a microprocessor.

24. Device according to one of claims 12 to 23, wherein the analog measurement signal (M) is a measure of an electrical

Current or an electrical voltage in a current conductor, in particular a high-voltage power distribution network.