WO2007042367A1

WO2007042367A1 - Method for determining the level based on the propagation time of a high-frequency test signal

Info

Publication number: WO2007042367A1
Application number: PCT/EP2006/066377
Authority: WO
Inventors: Dominik Buser; Bernhard Michalski; Stefan Scherr
Original assignee: Endress+Hauser Gmbh+Co.Kg
Priority date: 2005-10-13
Filing date: 2006-09-14
Publication date: 2007-04-19
Also published as: DE102005049500B4; DE102005049500A1

Abstract

The invention relates to a method for determining the level (l) based on the propagation time (t) of a high-frequency test signal (SHF) that is transformed into a low-frequency intermediate frequency signal (SZF). According to the invention, the transformation factor (kT) is determined from a differential frequency (fSweep) between a pulse repetition frequency (fPRF) and a sampling frequency (fSampl), the pulse repetition frequency (fPRF) or the sampling frequency (fSampl) being modified based on a regulation process using a control variable (c_var) by means of an adequate control algorithm in such a way that a setpoint value of the differential frequency (fSweep-setpoint) is regulated. A gradient (grad) is determined from at least two values, an operating point (OP) of the regulation process being determined based on the gradient (grad) and the differential frequency (fSweep) or the differential time (tSweep) when the control variable (c_var) has been adjusted, and the control algorithm being adjusted accordingly. The aim of the invention is to ensure that the differential frequency between the frequencies of two oscillators is quickly, safely, and accurately adjusted to a predetermined setpoint value of the differential frequency with the aid of a control algorithm.

Description

description

Method for determining the fill level based on the transit time of a high-frequency measurement signal

The present invention relates to a method for determining the level on the basis of the duration of a high-frequency measurement signal, which is transformed by means of a transformation process with a certain transformation factor in a low-frequency intermediate frequency signal.

Such methods for detecting and monitoring the level in a container are often used in the measuring devices of automation and process control technology. By the applicant, for example, measuring devices under the name Micropilot or Levelflex are produced and distributed, which operate according to the transit time measurement method and serve to determine and / or monitor a level of a medium in a container. According to the guided microwave or time domain reflectometry method or the Time Domain Reflection (TDR) measuring method, a high-frequency pulse is emitted along a Sommerfeld or Goubauschen waveguide or along a coaxial waveguide, which at a discontinuity of the DK value

(Dielectric constant) of the medium surrounding the waveguide is partially reflected back. After the free-radiating transit time measurement method, for example, microwaves are transmitted via an antenna into a free space or process space, and the echo waves reflected at the medium surface are received again by the antenna according to the distance-dependent transit time of the signal. Based on the time duration between the emission of the high-frequency pulses and the reception of the reflected echo signals, the distance of the measuring device to the medium surface can be determined. Taking into account the geometry of the container interior of the level of the medium is then determined as a relative or absolute size. The transit time measurement method can essentially be divided into two determination methods: The first determination method is based on a time measurement which requires a pulse train modulated signal for the distance covered; A second widely used method of determination is the determination of the swept frequency difference of the transmitted continuous high frequency signal to the reflected, received high frequency signal (FMCW - Frequency-Modulated Continuous Wave). In general, the following is not limited to a specific investigation.

[0003] A general problem with all transit time measuring methods with high-frequency measuring signals in the GHz range (gigahertz) is that for evaluating the high-frequency measuring signals. Frequency measurement signals must be used high-frequency components that are designed for such high frequency ranges. These high-frequency components have the disadvantage that their production is complicated and the purchase is very expensive. One way to evaluate the high-frequency measurement signals with cheap low-frequency components is to map the high-frequency measurement signals by means of a sequential sampling in the low frequency range. The method for sequential scanning of high-frequency measurement signals represents a possibility of transformation into the low-frequency range, in which method a time-expanded intermediate-frequency signal is generated from a multiplicity of high-frequency, periodically sampled measurement signals. This additional processing step is carried out because there are no correspondingly cost-effective data processing units, eg DSPs (Digital Signal Processors), which can reliably process high-frequency measurement signals. One approach to generating a time-extended intermediate frequency signal is the mixer principle, where two oscillators produce two oscillations with slightly different frequencies. Due to the slight 'detuning' of the frequencies of the two oscillations, a phase shift increases linearly with each measurement period, which corresponds to a linearly increasing time delay. The mixer principle is described, for example, in DE 31 07 444 A1 by means of a high-resolution impulse radar method. A generator generates first microwave pulses and emits them via an antenna with a predetermined transmission repetition frequency in the direction of the surface of the medium. Another generator generates reference microwave pulses that correspond to the first microwave pulses, but differ slightly from those in the repetition frequency. The echo signal and the reference signal are mixed, for example, by a frequency converter or mixer, whereby an intermediate frequency signal is produced. The intermediate frequency signal has the same course as the echo signal, but is compared to this stretched by a transformation factor which is equal to a quotient of the pulse repetition frequency and the frequency difference between the pulse repetition frequency of the first microwave pulses and sampling frequency of the reference microwave pulses. With a pulse repetition frequency of a few megahertz, a frequency difference of a few hertz and a microwave frequency of a few gigahertz, the frequency of the intermediate frequency signal is well below 200 kHz. The advantage of the transformation to the intermediate frequency is that relatively slow and thus inexpensive electronic components can be used for signal acquisition and / or signal evaluation. Reference is also made in this context to the German Utility Model DE 29815069 Ul, which describes this known transformation technique in a TDR - level gauge. This sampling circuit has two oscillators, of which at least one is variable in frequency, with one oscillator controlling the transmitting generator and the other oscillator controlling the sampling pulse generator. A frequency mixer forms the difference between the two frequencies which is used to set or regulate the transformation factor, which is as constant as possible, to a desired value via a feedback branch.

The disadvantage of the adjustment of the difference frequency to a desired value according to the prior art is that the control takes a very long time and under certain circumstances comes to a Einregelung on a wrong target value of the difference frequency.

The invention has for its object to ensure a fast, safe and accurate adjustment of the difference frequency of the signals from two oscillators to a predetermined target value of the difference frequency by a control algorithm.

This object is achieved by a method for determining the level on the basis of the duration of a high-frequency measurement signal, which is transformed by means of a transformation process with a certain transformation factor in a low-frequency intermediate frequency signal, wherein the transformation factor of a difference frequency or a difference time of a difference signal between a pulse repetition signal with a pulse repetition frequency and a sampling signal with a sampling frequency is generated and determined, the pulse repetition frequency and / or the sampling frequency is changed by a control with a controlled variable by a corresponding control algorithm / is so that the difference frequency to a desired value of the difference frequency or the Differential time is controlled to a target value of the difference time, wherein between at least two values of the difference frequency or between at least two values of the difference time in Ab dependent on the controlled variable a gradient is determined, wherein based on the gradient and the difference frequency or the difference time at the set control variable, an operating point of the control is determined, and accordingly the control algorithm is adjusted. By determining the gradient, it is possible to determine the position of the operating point of the control in the safe and critical control range, taking into account the measured value of the difference frequency or the difference time at the operating point. If the operating point is in a critical control range, the controlled variable is set to a defined state with a work point in the safe control range, and from there the control algorithm is started and executed. If, after determination of the gradient, the operating point has been localized in the safe control range, then the control algorithm in the control / evaluation unit changes the controlled variable, so that the current position approaching the setpoint of the operating point.

In a particularly preferred embodiment of the invention it is provided that the polarization or the sign of the difference frequency of the difference signal or the difference time is determined by means of the gradient. By determining the polarization or the sign of the difference frequency or difference time, it is possible to detect whether the first oscillator or sampling clock oscillator is clocked slower or faster relative to the second oscillator or Sendetaktoszillator. This polarization detection is necessary because only the difference frequency between the sampling frequency with respect to the pulse repetition frequency will be determined by the structure of the control loop, no matter which oscillator is clocked faster or slower to generate the sampling frequency or the pulse repetition frequency. The sampling clock oscillator with the sampling frequency normally has a slightly slower clock cycle than the transmit clock oscillator with the pulse repetition frequency. The case that the sampling clock oscillator has a faster clock cycle or a higher frequency than the transmit clock oscillator is not desired in the transformation process, thus preventing the differential frequency from being adjusted to a desired value in this critical control region.

An expedient embodiment of the invention is that a control characteristic of the difference signal is recorded by the difference frequency or the difference time is determined and stored in dependence on the changing control variable. The control characteristic can be recorded by going through the entire range of the controlled variable once and determining and storing the corresponding values of the difference frequency or difference time. These data are displayed in a diagram, whereby according to the difference time a pole or according to the difference frequency a zero with a reversal point in the diagram becomes recognizable. Because of this pole or zero point, at least two identical values of the difference frequency or the difference time can be determined for two different control variables. One of these values or operating points lies in a critical control range, in which the sampling clock oscillator has a faster clock cycle compared with the transmit clock oscillator. The other operating point is a safe control range in which the sampling clock oscillator has a slower clock cycle than the transmit clock oscillator. Based on the gradient and the diagram can be made a statement about the Einregelungsvorgang. In the diagram, a suitable setpoint of the operating point is determined, to which the control adjusts with the control algorithm. In addition, the control behavior of the control loop can be monitored, as this may change due to external influences or aging phenomena. Thus, with regard to the predictive maintenance of the device, the control loop is checked for errors and old monitored.

According to an advantageous embodiment of the method according to the invention it is provided that the control algorithm resets the operating point in a defined safe control range by the controlled variable is reset to a minimum value when the operating point is determined in a critical control range of the scheme. The operating point is transferred to a safe control range by the controlled variable, e.g. a control voltage is regulated back to a defined minimum value which lies within the safe control range. Starting from this minimum value, the control algorithm in the control / evaluation unit changes the control variable stepwise, so that the current operating point is constantly approaching the setpoint of the operating point.

A very advantageous variant of the method according to the invention is that

The control algorithm adjusts the operating point to the desired value of the operating point by the control variable is changed continuously when the operating point is determined in a defined safe control range of the scheme. The control algorithm adjusts the actual difference frequency to the setpoint value of the difference frequency or the current difference time to the setpoint value of the difference time by the control variable, e.g. a control voltage is continuously reduced or increased, depending on whether the current operating point is located above or below the setpoint of the operating point.

According to an advantageous embodiment of the method according to the invention it is provided that a fault is detected by the control algorithm, if the operating point is permanently determined in the disturbance range above a maximum limit of the difference frequency or below a minimum limit of the difference time. Since a difference time below this minimum limit value or because of the difference frequency above this maximum limit value can not occur in the closed-loop control, a hardware fault must be present. This detection of a disturbance of the control loop can be represented as a message on a display of the filling level measuring device or transmitted via the field bus to a control center or other field devices. Furthermore, it is possible that by determining a fault in the control loop, the control / evaluation unit or the entire meter re-initializes and thus brings the control in a defined initial state.

In an advantageous embodiment of the method according to the invention it is proposed that at a defined limit of the deviation of the difference frequency of the desired value of the difference frequency or the difference time of the setpoint of the difference time in the control algorithm, an error counter is incremented when this limit is exceeded, or is decremented when this Limit value is maintained. If the measured values of the difference time or the difference frequency deviate to some extent from the predetermined setpoint values of the difference time or the predetermined setpoint values of the difference frequency, eg +1 μs (microsecond) or 1 mHz (milli hertz), an error counter is activated. The error counter is incremented by "one" if the current reading is outside the set limits and decremented by "one" if the current reading meets the set limits.

According to an advantageous embodiment of the method according to the invention it is provided that a maximum count of the error counter is set, in which the error counter is deleted, the control algorithm is interrupted and the operating point in a defined safer control range of the control characteristic by returning the controlled variable to a minimum value is offset. If the error counter generates a so-called overflow, if the current count value exceeds the set maximum count value, for example 100, a regulation in a safe control range is restarted by reducing the controlled variable to a minimum value. The maximum count of the error counter determines the fault tolerance of the control system or control loop.

An advantageous embodiment of the method according to the invention is to be seen in that as a control variable up to a first tolerance value, a large deviation of the difference frequency of the desired value of the difference frequency or the difference time from the target value of the difference time a pulse width modulated control signal is effected by the control algorithm. The first tolerance value determines in which range of the deviations of the current difference time or the current difference frequency from the setpoint values of which a pulse-width-modulated control signal is used for controlling and activating the frequency regulation unit. The pulse width modulated control signal is suitable for large deviations of the current difference time or the current difference frequency of their setpoints, since large changes in the controlled variable or control voltage can be achieved while fine adjustment of the difference frequency or difference time to the setpoint for small deviations only difficult is possible. This first tolerance value of a large control deviation is, for example, 5 ms (millisecond).

A supplementary advantageous variant of the method according to the invention is to be seen in that as a control variable at a second tolerance value of a mean deviation of the difference frequency of the setpoint of the difference frequency or the difference time from the setpoint of the difference time a pulse-modulated control signal is generated by the control algorithm. From a second tolerance value, for the reasons described above, namely that a pulse-width-modulated control signal for finely adjusting the actual value is adjusted to the setpoint value of the difference time or differential value. frequency is not suitable, a pulse train modulated control signal applied. Depending on their operating frequency, the control / evaluation unit or microcontroller generate short pulses which result in the control signal which is modulated by pulse train. The microcontroller generates maximum short pulses which, depending on the required control variable, drive the frequency control unit as a corresponding pulse train. This second tolerance value of a mean control deviation is for example 1 μs (microsecond).

Furthermore, it is proposed that at a third tolerance value of a small or tendency deviation of the difference frequency of the target value of the difference frequency or the difference time from the target value of the difference time is not immediately controlled with a corresponding control variable, but a tendency of the system deviation is determined and a is generated, which is added to the subsequent pulse train modulated control signals or to the pulse width modulated control signals. If the current deviation of the difference time, for example, in the range of a few microseconds microsecond and / or this corresponds to a tendency error, so is not readjusted immediately, but these minimal changes are included in the subsequent pulse train modulated or pulse width modulated control signals. Since such small control deviations below the third tolerance value can only be corrected very difficult by readjustment or tracking of the controlled variable, the computational integration of these tendency deviations into the larger and medium control deviations creates a possibility for making the control more accurate and stable. An overshoot and a subsequent settling of the control can be avoided.

The invention will be explained in more detail with reference to the following drawings. For simplicity, identical parts have been given the same reference numerals in the drawings. It shows:

Fig. 1 shows an embodiment of a device for determining the level in a container;

2 shows an embodiment of the control circuit of the device for generating the intermediate frequency with the transmitting / receiving unit and control / evaluation unit.

3 shows an embodiment of the control characteristic with the safe and critical control areas and the fault area; and

4 shows an exemplary embodiment of a schematic state diagram of the signals in the control circuit shown in FIG. 2.

In Fig. 1, an embodiment of the device 1 according to the invention for determining the distance d or the level 1 is shown on the basis of the transit time t of high-frequency measurement signals S _HF . For this purpose, the device 1 mainly a connected to the transmitter 2 measuring unit 3, by means of which the high-frequency measurement signal S _RF or transmission signal S _{τx is coupled} into a filling material 5 comprehensive process _space 6 of the container 7 and sent out. The measuring unit 3 is inserted through an opening, for example a nozzle, into the process space 6 of the container 7. The measuring unit 3 or the transducer element, as shown in Fig. 1, as an antenna 3a and in particular as a horn antenna, a rod antenna, a parabolic antenna or a planar antenna, be configured, the transmission signal S _τx in the process _space 6 of the container 7 radiates freely. Instead of antennas 3a which radiate freely in the process space 6, a surface waveguide 3b in the device 1 for level measurement, which measures the high-frequency measurement signal S _HF due to the skin effect on the surface thereof, can also be used as measuring unit 3, as shown in FIG along leads.

In the transmitter 2 of the device 1, the transmission signals S _τx in a

Transmitting / receiving unit 17 is generated and transmitted to the emitting measuring unit 2. Sending transmitted signals S _T χ, which have for example been reflected again on the surface 4 of the medium 5, are received by measuring unit 2 and passed back to the transmitting / receiving unit 17, in which the received reflection signals S _{RX are} pre-processed. The high-frequency measurement signals S _HF , consisting of transmission signals S _T χ and reflection signals S _RX are, for example, in the transmitting / receiving unit 17 via a sampling method or a sequential sampling with two frequency-shifted slightly in the frequency high-frequency pulse sequences in a time-stretched, lower-frequency intermediate frequency signal S. _2F converted. The intermediate frequency signal S _ZF mixed down in this way can then be evaluated in the low-frequency range by a control / evaluation unit 15 and the transit time t or the travel x of the emitted high-frequency measurement signal S _HF can be determined.

[0026] Accordingly, in the guided microwave method, the time domain reflectometry or the time domain reflection (TDR) measuring method, a transmission signal _{S.sub.τx is emitted,} for example, along a Sommerfeld or Goubausch surface waveguide 3b or coaxial waveguide, which in the case of a discontinuity of the DK Value (dielectric constant) of the surface waveguide 3b surrounding filling material 5 is reflected back. Due to the impedance jumps caused by the filling material 5 within the process space 6 of the container 7, in particular at the boundary layer 4 between free space and filling material 5 in the container 7, the transmission signal S _{τx is} at least partially reflected back. Due to this, a corresponding, usually weaker reflection signal S _RX in the opposite direction on the surface waveguide 3b to the transmitting / receiving unit 17 runs back.

The free radiated from the antenna 3a and at the lossless Oberflä- Chen chenwellenleiter 3b guided high-frequency measurement signals S _HF propagate in the free process space 6 of the container 7 in an atmosphere of air or inert gas approximately at the speed of light C ₀ . These transmission signals S _τx are partially or completely reflected back to surfaces 4 of media having a higher dielectric constant value, for example the filling material 4, than the air or the protective gas in the radiation cone of the antenna 3a or in the vicinity of the surface waveguide 3b , About the measured transit time t of the transmitted signal S _τx to the reflected reflection signal S _RX is determined by a conversion over the formula of the wave velocity, the distance traveled distance x or distance d. This difference distance or this distance d corresponds to the height h of the container 7 minus the level 1 of the filling material 5 in the container 7. Since the height h of the container 7 or the position of the coupling of the transmission signal S _{τx is} known, thus the level 1 and possibly even the volume, or with knowledge of the density of the filling material 5, the mass of the filling material 5 in the container 7 determine.

The task of the communication / supply unit 14 is to control the communication with, for example, a remote control center or another measuring device or field device via a field bus 9 and to measure the measured values, e.g. the level 1, or to receive configuration data of the device 1 and to send. The fieldbus 9 operates according to the usual communication standards, e.g. Foundation Fieldbus or Profibus PA, and is designed, for example, in a usual in process instrumentation two-wire technology. The supply of the device 1 with energy can be done in addition to the power supply of the device 1 via the field bus 9 according to the two-wire standard by means of a separate power supply line 8. The communication / supply unit 14 may be configured as an integral part of the control / evaluation unit 15.

In Fig. 2 shows an embodiment of a block diagram of the device 1 is shown, which shows the transmitter 2 of the device for generating the intermediate frequency S _ZF with a transmitting / receiving unit 17 and control / evaluation unit 15. The transmitting / receiving unit 17 can basically be converted into an HF circuit part 21 with a transmitting pulse generator 18, sampling circuit 19, and transmitting / receiving dipole 23, in which in principle HF signals are generated and processed, and a LF circuit part 20 with transmitting clock oscillator 13, sampling clock oscillator 12, frequency converter 11, in which in principle LF signals are generated and processed, are divided. The individual circuit elements in the RF circuit part 21 are constructed according to experience in analog circuit technology, ie analog measurement signals are generated and processed. On the other hand, the individual circuit elements in the LF circuit part 20 can be constructed either on the basis of digital circuit technology and / or analog circuit technology. Under the From the point of view of the rapid progress of digital signal processing, it is also conceivable to design the RF circuit part 21 with digital circuit elements. There are also the most diverse variations of the individual circuit elements in digital and analog circuit technology conceivable, which are not explicitly carried out here. Therefore, the following description of the embodiment of Fig. 2 is to be considered as an example of many possible embodiments. The communication / supply unit 14 has not been shown explicitly again in FIG. 2 for reasons of reducing the representation to the essentials.

In the transit time measurement of pulsed high-frequency measurement signals S _HF coupled to the measuring unit 3 transmitter / receiver unit 17 is used to mutually coherent wave packets of predeterminable pulse shape and pulse width, so-called bursts or short wave packets to produce and process. The pulse shape of a single burst or a single short wave packet usually corresponds to needle-shaped or sinusoidal, half-wave pulses of predeterminable pulse width. However, if necessary, other suitable pulse shapes may be used for these bursts. For this purpose, the transmitting / receiving unit 17 comprises a transmitting pulse generator 18 triggered by the transmitting clock oscillator 13 for generating a first burst sequence serving as a transmitting signal S _T χ. The pulses of the transmission signal S _T χ are carried at a high frequency f _{HF of} the transmit pulse generator 18, which is approximately in the range between 0.5 and 78 GHz, and moreover triggered with a pulse repetition frequency f _PRF or shot rate, which has a frequency range of some Megahertz, in particular a frequency range from 1 MHz to 10 MHz is set. This pulse repetition frequency f _PRF for driving the transmit pulse generator 18 is generated by a transmit clock oscillator 13. However, the high frequency f _HF and / or pulse repetition frequency f _PRF can also, if necessary, be outside the respectively specified frequency ranges.

The applied to the signal output of the transmission pulse generator 14 transmission signal S _TX is by means of a transmitting / receiving switch 23, in particular by means of a directional coupler or a hybrid coupler, the transmitting / receiving unit 17 in the at a first signal output of the transmitting / receiving switch 23 connected measuring unit 3, for example, antenna 3a or surface waveguide 3b, coupled. Practically at the same time, the transmission signal S _{τx is} also applied to the second signal output of the transmission / reception _switch 23. As already mentioned, the reflection measuring signals S _RX generated in the measuring volume 6 of the container 5 in the manner described above are again received by the device 1 by means of the measuring unit 3 and coupled out at the second signal output of the transmitting / receiving switch 23. Accordingly, at the second signal output of the transmitting / receiving switch 23, by means of the transmission signal S _T χ and the reflection measurement signal S _RX formed total measurement signal S _TX + S _{RX are} tapped.

A direct evaluation of the voltage applied to the second signal output of the transmitting / receiving switch 8 high-frequency total measurement signals S _TX + S _RX , in particular a direct measurement of the transit time t, is practically no longer or only with a high technical complexity, eg by using high-frequency electronics Components, possible. Because of this, the transmitting / receiving unit 2 further comprises a sampling circuit 19 which temporally expands the high-frequency-carrying total measuring signal S _TX + S _RX in such a way that the high frequency f _HF and the pulse repetition frequency f _{PRF are transformed} into a lower frequency range of a few hundred kilohertz. For temporally stretching the total measurement signal S _TX + S _RX , this is supplied to a first signal input of the sampling circuit 19 connected to the second signal output of the transmitting / receiving switch 23. At the same time, the total measuring signal Sxχ + Si "of the sampling circuit 19 is a sampling signal Ss applied _amp i of the sampling oscillator 12 at a second signal input. A sampling frequency f _Samp i or clock rate at which the sampling signal Ss _amp i is clocked is normally set slightly smaller than the pulse repetition frequency f _{PRF of} the transmission signal S _τx . By means of the sampling circuit 19, the total measurement signal S _TJC + S _I «is mapped to an intermediate frequency signal S _ZF , which is time-stretched by a transformation factor K _τ compared to the total measurement signal S _TJC + S _I «. Due to the frequency offset between the pulse repetition frequency f _PRF and the sampling frequency f _Samp i, this sampling circuit 19 samples the total measurement signal S _TX + S _RX in each phase with different phase angles, resulting in a time-expanded intermediate frequency signal S _ZF having the above-described transformation factor k _τ . The sampling circuit 19 is designed, for example, as an HF frequency converter or RF mixer with a sampling pulse generator having the same phase position and frequency of the burst sequence as the transmission pulse generator 18, or a fast sampling switch. As a sampling switch, for example, RF diodes or fast transistors can be used.

The transformation factor k _τ , with which the total measurement signal Sx _x -I-Si _{Ot is converted} into a lower-frequency intermediate frequency signal S _2F corresponds to a quotient of the pulse repetition frequency f _{PRF of} the transmission signal S _τx divided by a difference of the pulse repetition frequency f _{PRF of} the transmission signal S _τx and the sampling frequency f sam _p i of the sampling signal S _Saπψh

[0034]

An intermediate frequency f _{IF of} the intermediate frequency signal S _IF generated in this way is included such devices 1 for determining the level 1 usually in a frequency range of 50 to 200 kHz; if necessary, the frequency range can be selected higher or lower. Empirically, the intermediate frequency f _{IF is set} to about 160 kHz in the measuring instruments of the applicant. The dependence of the intermediate frequency f 1 on the ratio of the sampling frequency f _Samp i and the pulse repetition frequency fp _R p as shown in the second equation (equation 2) can be derived from the first equation (equation 1). [0036]

(Equation 2)

[0037] If necessary, the intermediate frequency signal S _IF, k over the total measuring signals S _TX + S _RX by a transformation factor is zeitgedehnt _τ, _found in a suitable manner by a filter / amplifier unit 22 as a filtered intermediate frequency signal S zp amplified and filtered, before it is evaluated in the control / evaluation unit 15 or other evaluation circuits as echo curve or envelope.

This difference _frequency f _s " _eeP is determined for two reasons; First, by this control circuit 24, the instantaneous control and triggering of the sampling clock oscillator 17 and possibly also the Sendetaktoszillators 18 by the control / off value unit 15 checks and second, from the quotient of the known or measured pulse repetition frequency f _PRF and the difference _frequency f _s " _eeP a Transformation factor k _τ in the control / evaluation unit 15 determined. In the control / evaluation unit 15, the transit time t of the measurement signals, as well as the fill level 1, can also be determined by echo signal evaluation of the filtered intermediate frequency signal S _ZF and by the knowledge of the transformation factor k _τ .

In this embodiment, in the LF circuit part 20, the transmission

/ Receiving unit 17 by the sequential sampling of the pulse repetition frequency f _PRF at the sampling frequency fs _amp i by means of a frequency converter 11, a difference frequency fs _weep between the sampling clock oscillator 13 and the transmit clock oscillator 12 determined. This difference frequency f _s "ee _P or its difference time t _s " _e e _P is processed and measured in the control / evaluation unit 15. The difference time t _s " _e e _p corresponds to the reciprocal of the difference frequency f _s " _e e _P - via a feedback branch controls the control / evaluation unit 15 the controllable sampling clock oscillator 13 according to the determined difference _frequency fs " _eeP or the difference time t _SweeP again. By this structure, a control loop 24 has been created, the difference _frequency f _s " _eeP largely on the desired setpoint of the difference _frequency fs _WeeP _ _setPomt , eg 21.73913 Hz, sets. The controllable sampling clock oscillator 13 and possibly the Sendetaktoszillators 12 is effected by a frequency control unit 10, the output of voltage controlled oscillators, eg VCO or oscillator with a parallel variable-frequency capacitance diode, a control voltage V _c as the corresponding control variable c_var or the digitally controllable oscillators, eg NCO , outputs a digital control value V _dlg as a control variable c_var. The frequency control unit 10 is _adjusted by means of a so-called three-level control by an up-control signal R _Up and a down-control signal R _Dθwn of the control / evaluation unit 15 so that an applied to the output of the frequency control unit 10 controlled variable c_var controls the controllable sampling oscillator 12 accordingly. The appropriate control of the sampling oscillator 12 takes place in the manner that the generation of the defined target value for the difference frequency f _sweep _ _setpomt between the pulse repetition frequency f _PRF and the sampling frequency f _samp _e i is effected. For example, either a control voltage V _c or a digital control value V _d i _g can be used as controlled variable c_var. The frequency control unit 10 is configured, for example, as an RC element or a charge pump, which stabilizes the control voltage Vc, which regulates the sampling frequency f _Samp i of the sampling oscillator 12. With upregulation signals R _Up , for example, the charge voltage of the charge pump is increased and lowered with down-regulation signals R _Dθwn . The adjusted control variable c_var or control voltage V _c is determined by the control / evaluation unit 15 via a measuring line. This determination is necessary because normally only the difference frequency f _{sweep is} determined and the actually generated control variable c_var or control voltage V _{c is} needed to determine the gradient. In the rule / off value unit 15, for example, a microcontroller 16 is integrated, which controls the control and executes the control algorithm. FIG. 3 shows an exemplary embodiment of the control characteristic con_char of the difference time t _sweep with the safe control range E and critical control ranges A, B, C, F and the fault range D. On the ordinate of the coordinate system, the difference time t _SweeP and on the abscissa of the coordinate system, the control voltage V _c is displayed as a controlled variable c_var. As a control variable c_var an analog control voltage V _{c has been} displayed in this example; regardless, any other controlled variable c_var such as a digital control value V _dlg or a mechanical controlled variable is applicable. The control characteristic con_char of the difference time t swee _P is shown completely in FIG. 3, but it is also possible that only individual properties, such as exceeding the limit value and / or the gradient grad of the control characteristic con_char at a current control voltage V _act , are determined. The control characteristic con_char of the difference time t _sweep or the properties determined in sections can be generated, for example, by that the control voltage V _c or control variable c_var is continuously changed at the sampling oscillator 12 of FIG. 2 and corresponding values of the difference time t _sweep or difference _frequency fs " _{eeP are} determined and stored by the control / evaluation unit 15. From the determination of these values, it is possible to deduce the position of the current operating point OP in the corresponding control ranges A, B, C, D, E, F. If the operating point OP is within a critical control range A, B, C, F or a fault range D, then the control algorithm is interrupted and a renewed adjustment in a safe control range E is started. The initiation of the control in a safe control range E takes place, for example, in that the control voltage V _{c is brought} to a minimum value of the control voltage V _mm and consequently increased continuously until a desired value of the operating point OP_setpoint or a desired value of the difference time t _SweeP _ _setPomt the control / evaluation unit 15 is determined. An operating point OP can not be in the critical control ranges A, B, since the control voltage V _c according to the invention can be set only between a minimum value of the control voltage V _mm and a maximum value of the control voltage V _max . These limiting values of the control voltage V _c were therefore generated because this difference time t _s "ee _p below a minimum limit value of the difference time ts" _e e _P _mm, for example less than 1 ms (millisecond) lie, and here operating points OP and Grad Grad not exactly determined can be.

The position of the setpoint of the operating point OP_setpoint is chosen so that a Einregel the difference time t _SweeP or difference _frequency f _s " _eep by a corresponding control voltage V _{c is} easily possible. If, on the other hand, the position of the reference value of the difference time t _Sw ee _P _set _P omt or of the reference value of the difference frequency fs _W ee _P _set _P omt is selected so that a slight change in the control voltage V _c _{results in} a large change in the differential time t _SweeP or difference _frequency f _s _eeP caused by the operating point OP is at a point of the control characteristic con_char, at which the gradient is very high, so an exact adjustment to the setpoint of the operating point OP_setpoint is difficult is possible. Due to the grad grad is determined with what accuracy the control voltage V _c must be changed so that a certain change in the differential time t _SweeP or difference _frequency f _s " _eeP can be achieved. Consequently, upon detection of a large gradient, the control voltage V _{c is changed} with a higher accuracy by a more exact, slower control algorithm that makes only small changes in the voltage. However, if the gradient grad is small, the control voltage V _{c is changed} in larger change _steps , as here is a change in the difference time t _SweeP not so strong.

The control characteristic con_char or the control characteristic of the differential time t _swe e _P has a pole point at a specific pole position control voltage V _Po i. At this Polstelle the control characteristic con_char are the sampling frequency f _SamP i _{e of} the controlled sampling oscillator 12 and the pulse repetition frequency f _{PRF of} the pulse repetition oscillator 13 exactly the same. Due to the pole of the control characteristic con_char of the difference time ts " _eeP , it is possible to _generate the same difference time t _SweeP by means of a first control voltage V _C i in the safe control range E and by means of a second control voltage V _C2 in the critical control range F. In this case, the first control voltage V _C i in the safe control range E is correspondingly smaller than the pole position control voltage V _Po i and the second control voltage V _C2 in the critical control range F is correspondingly greater than the pole position control voltage V _Po i. Above a maximum limit value of the difference time t _Swe e _P _max, which is above the setpoint of the difference time t _Sw ee _P _set _P omt, the control algorithm is interrupted and the control of the difference time t _SweeP or the corresponding difference _frequency f _s " _eeP in the safe control range E started again. By forming this maximum limit value of the difference time t _SweeP _ _max , rapid control of the difference time t _SweeP to a setpoint value of the difference time t swee _p _set _p oint is only possible. The control in the critical control range C around the pole, in which the difference time t _swe e _P assumes high, inappropriate values and thereby causes long measurement and control times, is thus prevented. The contrary case that the control characteristic con_char or the control characteristic of the difference frequency f _s "ee _P - to a certain extent the reciprocal of the difference time t _swe e _P - has a zero point with a reversal point is not explicit due to the analogy to the above explanations in FIG have been carried out.

In Fig. 4 is a state diagram of the signals of the control loop 24 of FIG. 2 is shown. The state diagram consists of five signal characteristics, all of which follow the same time unit or time scale t _s . The uppermost, first signal characteristic of the supply voltage P indicates the time of initialization of the microcontroller Int_μC or the start time of the voltage supply of the microcontroller 16 or of the entire control / evaluation unit 15. The third and fourth signal characteristics indicate the control signals R _c for up-regulation R _1J p and for down-regulation R _Dθwn , which _indicate signals for driving a frequency regulation unit 10. The fifth signal characteristic represents the control voltage V _c , which is output, for example, from the frequency control unit 10 and controls a voltage-controlled sampling clock oscillator 12. The feedback of the control loop 24 or the control of the sampling clock oscillator 12 via a frequency control unit 10 is achieved by the signal characteristics three, four and five. The second signal characteristic of the difference signal S _s " _eeP with the varying difference time t _SweeP represents the generated control result of the control loop 24.

By switching on the supply voltage P, the microcontroller 16 or the control / evaluation unit 15 is initialized. It happens that due to lying rule process phases and the operating state or transient response of the electrical components in this state a not previously definable current control voltage V _{act applied to} the sampling oscillator 12. For this reason, in the method according to the invention, first a work point detection OP_ID is performed which, in a first control step, determines a first difference time t _SweeP i with the corresponding current control voltage V _act or fourth control voltage V _C4 . Subsequently, in a further control _step, a second control voltage V _{C2 is} set and a corresponding second differential time t _{SweeP2 is} determined. From the quotient of the difference of the first difference time t _SweeP i and second difference time t _sweep2 to the difference of the corresponding control _voltages V _C4 , V _C2 , a gradient grad or a slope is determined. The same can be _done by generating a first control voltage V _cl and a third control voltage V _C3 and determining the corresponding difference _times t _SweeP i, t _s " _eep2 . By determining the grad grad and by determining the exceeding of the limit value of the difference frequency ts "ee _P _set _P omt it can be determined very quickly whether the operating point OP of the control in a critical control range A, B, C, F or a safe control range E is located. In FIG. 4, the example of FIG. 3 has been taken up, that the current control voltage V _{act was located} in the critical control range F during the initialization of the microcontroller Init_μC or the initialization of the control / evaluation unit 15 and a critical control range F corresponding gradient exists. In this case, the control algorithm down-regulates the current control voltage V _act to a minimum value of the control voltage V _mm , whereby a new control phase in the safe control range E is started. At the start of this Einregelungsphase there is a large deviation of the current difference time t _s "ee _P to the setpoint of the difference time t _Sw ee _P _set _P omt before. _{Which is} why the current difference _frequency t _{SweeP is changed} by a plus-width-modulated control C_PWM with pulse-width-modulated control signals R _PWM , which are configured both as up-control signals R _Up and down-control signals R _Dθwn . The up-control signals R _Up and down-control signals R _{Dθwn in} this _case control the frequency control unit 10, which generates a corresponding control voltage V _c . On the other hand, an applied to the frequency control unit 10 up-control signal R _Up causes an increase in the control voltage V _c at the output of the frequency control unit 10. On the other hand, applied to the frequency control unit 10 down-control signal R _{Dθwn causes} a drop in the control voltage V _c Is the deviation of the current difference time t _Svιeep the setpoint of the difference time ts " _e e _P _set _P omt only small, so is in a pulse train modulated control C_Toggle with pulse train modulated control signals Rτ _ogg ie changed, in which the frequency control unit 10 is driven only with short pulse trains of very short pulses. Due to the short pulses of the impulse sequence modulated control signals Rτ _ogg i _e , it is possible that only a slight change in the difference _frequency t _SweeP can be made. By this method, that with small deviations of the difference time t _SweeP from the setpoint of the differential time t _Sw ee _P _set _P omt a pulse train modulated control signal Rτ _ogg ie is used for fine adjustment, only a small transient response of the control occurs. If the difference time t _swe e _{P is} adjusted to the setpoint of the difference time t _Sw ee _P _set _P omt, then the setpoint of the operating point OP_setpoint is set in the safe control range E. By single short pulses of the down-control signal R _Dθwn and the up-control signal _R-up of an overall system deviation can be counteracted at a time t the adjusted difference _sweep to the target value of the difference time ts _weep _ _setPomt. For this purpose, for example, the period is interrupted depending on the tendency of the control deviation, or one or more short pulses are omitted. [0045] List of Reference Numerals

1. Device

2. Transmitter

3. Measuring unit 3a Antenna

3b surface waveguide

1st surface, boundary layer

2. contents; medium

3rd process room

4. container

5. Power supply line

6. Fieldbus

7. Frequency control unit

8. Frequency converter

9. first oscillator, sampling clock oscillator

10. second oscillator, transmit clock oscillator

11. Communication / supply unit

12. Control / evaluation unit

13. microcontroller

14. Transceiver unit

15. Transmit pulse generator

16th sampling circuit

17. LF circuit part

18. RF circuit part

19. Amplifier / filter unit

20. Transceiver 21. Control circuit

S _HF high-frequency measurement signal

S _τx transmission signal

S _RX reflection signal

[0051] Szp Intermediate Frequency Signal

[0052] S _s "ee _p differential signal

[0053] S _PRF Pulse Reception Signal

F _PRF pulse repetition frequency

[0055] _mp Ssa ie sampling signal

[0056] _mp ie sampling frequency fsa

Fs _weep difference _frequency

[0058] t _Sw ee _p difference time

[0059] t _s "ee _p i first difference time

T _s "eep2 second difference time

F sweep_setpoint setpoint of the difference frequency

Ts "eep_setpomt setpoint of the difference time

[0063] t _s "ee _p _ _m a _x maximum limit value of the difference time

[0064] ts "ee _p _ _mm minimum limit value of the difference time

[0065] P supply voltage

Init_μC Initialization of the microcontroller

C_PWM plus-width modulated control

C_Toggle pulse-sequence modulated control

[0069] OP operating point

[0070] OP_setpoint setpoint of the operating point

[0071] OP_ID Work Point Detection

R _c control signals

R _up up-control signals

R _DOWΠ down-regulation signals

R _PWM pulse width modulated control signal

[0076] Rτ _ogg i _e pulse train modulated control signal

C_var controlled variable

C_varl first controlled variable

C_var2 second controlled variable

Degree gradient; of change; pitch

[0081] Vc control voltage

[0082] Vc ₁ first control voltage

[0083] V _C2 second control voltage

Vc ₃ third control voltage [0085] V _C4 fourth control voltage

[0086] V _max Maximum value of the control voltage

[0087] V _mm Minimum value of the control voltage

[0088] V _act current control voltage, actual value of the control voltage

[0089] V _set setpoint of the control voltage

[0090] Vp _o i pole position control voltage

V _dlg digital rule value

[0092] con_char control characteristic

E safe control areas

A, B, C, F are critical control ranges

[0095] D disturbance region

K _τ transformation factor

[0097] d distance

H height

[0099] 1 level

[0100] t running time

[0101] t _s time scale

[0102] x running distance

Claims

claims

[0001] Method for determining the filling level (1) on the basis of the transit time (t) of a high-frequency measuring signal (S _HF ) which is transformed by means of a transformation method with a specific transformation factor (k _τ ) into a lower-frequency intermediate frequency signal (S _ZF ), wherein the transformation factor (k _τ ) of a difference frequency (fs _W ee _P ) or a difference time (t _s " _e e _P ) of a difference signal (Ss" ee _P ) between a pulse repetition signal (S _PRF ) with a pulse repetition frequency (f _PRF ) and a sampling signal (S _S on _P i) with a sampling frequency (f sam _p i) is generated and determined, wherein the pulse repetition frequency (f _PRF ) and / or the sampling frequency (fsam _P i) based on a control with a controlled variable (c_var) by a corresponding control algorithm is / will be changed so that the difference _frequency (f _s " _eeP ) to a setpoint of the difference frequency (f swee _p _set _p omt) or the difference time (t _Swe e _P ) to a setpoint d the difference time (ts _weeP _ _setpomt ) is regulated, wherein a gradient (grad) is determined between at least two values of the difference _frequency (fs " _eep ) or between at least two values of the difference time (ts" _eep ) as a function of the controlled variable (c_var) , wherein on the basis of the gradient (grad) and the difference _frequency (fs " _eeP ) or the difference time (t _sweep ) with set control variable (c_var) an operating point (OP) of the control is determined, and accordingly the control algorithm is adjusted.

2. The method of claim 1, wherein the polarization or the sign of the

Difference _frequency (f _s "ee _p ) of the difference _signal (S _s " ee _P ) or the difference time (ts " _eep ) by means of the gradient (grad) is determined.

3. The method of claim 1, wherein a control characteristic (con_char) of the difference signal (Ss "ee _p ) is recorded by the difference frequency (fs _W ee _P ) or the difference time (t _s " ee _P ) in dependence on the changing control variable (c_var) is determined and stored.

4. The method of claim 1 or 3, wherein the control algorithm resets the operating point (OP) in a defined safe control range (E) by the controlled variable (c_var) is reset to a minimum value when the operating point (OP) in one critical control range (A; B; C; F) of the control is determined.

5. The method of claim 3 or 4, wherein the control algorithm adjusts the operating point (OP) to the setpoint of the operating point (OP_setpoint) by the controlled variable (c_var) is changed continuously when the operating point (OP) in a defined safe Control range (E) of the scheme is determined

6. The method of claim 3, 4 or 5, wherein of the control algorithm a Fault is detected if the operating point (OP) permanently in the fault range (D) above a maximum limit value of the difference _frequency (f _s " _eeP ) or below a minimum limit value of the difference time (t _SweeP ) is determined.

7. The method of claim 1, 3, 4, 5 or 6, wherein at a defined

Limit value of the deviation of the difference _frequency (f _s " _eeP ) from the setpoint value of the difference frequency (fs" ee _p _set _p o _m t) or the difference time (t _Swe e _P ) from the setpoint value of the difference time (ts "ee _P _set _P o _m t) in the control algorithm, an error counter is incremented if this limit is exceeded or decentred if this limit is met.

8. The method of claim 7, wherein a maximum count of the error counter is set, in which the error counter is deleted, the control algorithm is interrupted and the operating point (OP) in the defined safer control areas (E) of the control characteristic (con_char) by Returning the controlled variable (c_var) to a minimum value.

9. The method of claim 1 or 7, wherein as a controlled variable (c_var) up to a first tolerance value, a large deviation of the difference frequency (fs _W ee _P ) of the desired value of the difference frequency (fs _W ee _P _set _P o _m t) or the difference time (t swee _P ) from the target value of the difference time (t _s "ee _P _set _P omt) a pulse width modulated control signal (R _PWM ) is effected by the control algorithm.

10. The method of claim 1, 7 or 9, wherein as a controlled variable (c_var) up to a second tolerance value of a mean deviation of the difference frequency (f _s "ee _P ) of the desired value of the difference frequency (fs _W ee _P _set _P o _m t) or the difference time (t _Swe e _P ) from the setpoint of the difference time (t _s "ee _P _set _P o _m t) a pulse train modulated control signal (Rτ _ogg i _e ) is effected by the control algorithm.

11. The method of claim 1, 7, 9 or 10, wherein up to a third tolerance value of a tendency deviation of the difference _frequency (f _s " _eeP ) of the desired value of the difference frequency (fs _W ee _P _set _P o _m t) or the difference time (t swee _P ) from the target value of the difference time (t _s "ee _P _set _P o _m t) is not immediately countered with a corresponding control variable (c_var), but a tendency of the control deviations is determined and a tendency control signal (S _Dπft ), which is added to the pulse sequence modulated control signal (R i _Togg _e) or to the pulse-width control signal tenmodulierten (R _PWM) is generated.