JP2014092417A - Liquid level measurement instrument - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an accurate FMCW radar type liquid level measurement instrument by calculating a specific permittivity of a radio wave propagation space and maintaining linearity of an F-V curve of a VCO and correcting frequency characteristics of radio propagation in a circular waveguide.SOLUTION: The FMCW radar type liquid level measurement instrument includes: a V-T table for sweeping a frequency of a signal generated by a VCO; a reflecting member located at a known distance to correct linearity of F-T characteristics of the VCO; a correction V-T curve table generated by analyzing errors from a fixed sine wave signal of a beat waveform generated by a transmission signal according with F-T characteristics before correction and a reflected signal from the reflecting member; a corrected V-T table obtained by correcting the V-T table by the correction V-T curve table; a predistortion table generation unit for furthermore correcting the corrected V-T table; and a predistortion V-T curve table generation unit which applies a predistortion table to the corrected V-T table to generate a predistortion V-T curve table.

Description

  The present invention relates to an FMCW (Frequency Modulated Continuous Wave) radar level measuring device that can further increase measurement accuracy.

One of the measurement methods in the liquid level measurement device is a radar method, and one of the radar methods is an FMCW radar method. As shown in FIG. 21, the FMCW radar system sweeps a predetermined frequency (F) linearly with respect to a time axis at a predetermined fixed time (this time is called a sweep time (T)). However, radio waves are transmitted toward the measurement point. As shown in FIG. 22, during the round-trip time t until the radio wave transmitted at the transmission point is reflected at the measurement target point (the distance from the transmission point is L) and returns to the transmission point, the transmission frequency is Swept by F · t / T (Hz). The swept frequency is the difference of the reception frequency F _R and the transmission frequency F _T of the moment of receiving the reflected wave (hereinafter referred to as "beat frequency F _B").

The round trip time t is t = (T / F) × F _B
Since it is, if it is possible to measure the beat frequency F _B, it can be radio waves from the transmission point to the measurement point measures the time t required for the round trip. Since the propagation speed of the radio wave in free space is the same as the speed of light C, the distance L from the transmission point to the measurement point can be expressed by Equation 1.
Formula 1
L = C × t / 2 = C × T × F _B / 2F

  FIG. 20 shows a conventional example of an FMCW radar type distance measuring device. In FIG. 20, the FMCW radar type distance measuring device includes a DSP (Digital Signal Processor) 101, a digital / analog converter 102, a VCO (Voltage Controlled Oscillator) 103, a coupling circuit 104, a transducer 106, and a mixer 107. , An AGC (automatic gain control circuit) 108, and an analog / digital converter 109. The components including the digital / analog converter 102 and the VCO 103 constitute a transmission system, and the components from the coupling circuit 104 to the analog / digital converter 109 constitute a reception system.

  The DSP 101 has a built-in memory 111. The memory 111 determines the relationship between the voltage applied to the VCO 103 that determines the oscillation frequency of the VCO 103 with respect to the sweep time T, that is, a voltage-time curve (hereinafter referred to as “VT curve”) as a voltage-time table (hereinafter referred to as “VT table”). I remember).

  The DSP 101 reads the VT table from the memory 111, converts the voltage value that continuously changes with the passage of time into an analog signal by the digital / analog converter 102, and sets it as the control voltage of the VCO 103. The oscillation frequency of the VCO 103 changes continuously according to the control voltage. This oscillation signal is transmitted from the transducer 106 to the measurement target point (for example, the liquid level) via the coupling circuit 104. The substance of the transducer 106 is, for example, a planar antenna.

A measured system 31 is interposed between the transducer 106 and the measurement point. The radio wave reflected at the measurement target point returns to the measured system 31 and is captured by the transducer 106 and is guided to the receiving system via the coupling circuit 104. In the reception system, the oscillation signal upon reception the received signal in the mixer 107 is mixed, the difference between the reception frequency and the oscillation frequency of the moment, that is, the beat frequency F _B is taken out.

Beat frequency _{F B} signal, after being controlled to an appropriate amplitude value AGC108, is converted into a digital signal by an analog-to-digital converter 109 is input to the DSP 101. In the DSP 101, a process of reading out VT data from the VT table and taking in a beat signal is executed during the sweep time (T). A filtering process is performed on the beat signal group, which is time axis data captured during the sweep time (T), to remove unnecessary noise components. Thereafter, an FFT (Fast Fourier Transform) processing, to extract the beat frequency _{F B.} Further, the DSP 101 obtains the distance L to the measurement point by performing calculation processing by applying the formula 1.

  FIG. 19 shows an example in which the above-described FMCW radar type distance measuring device is used for liquid level measurement of liquid stored in a tank. In FIG. 19, the measuring device 10 is installed on the top plate of the tank 40. The tank 40 stores the liquid 41. On the upper part of the tank 40, a transducer 106 as an antenna is installed from the top plate of the tank 40 toward the liquid surface of the liquid 41. If the internal space of the tank 40 is free space, the radio wave transmitted from the transducer 106 toward the liquid 41 propagates through the tank 40 at the speed of light C, and is reflected by the liquid surface of the liquid 41. Received.

  When radio waves are transmitted toward the free space in the tank 40, when the dielectric constant of the liquid 41 is low, the radio wave reflection level at the liquid level is low, and the measurement accuracy of the liquid level is lowered. In order to solve this problem, as shown in FIG. 19, the measuring apparatus 10 is installed on the top plate of the tank 40, and the cylindrical circular waveguide 30 is installed from the top plate of the tank 40 toward the bottom. . A radio wave is transmitted from the transducer 106 toward the inside of the circular waveguide 30, and the reflected wave reflected by the liquid surface of the liquid 40 is received by the transducer 106. The liquid 41 can enter the circular waveguide 30 so that the liquid level in the tank 40 and the liquid level in the circular waveguide 30 coincide with each other.

  The measuring apparatus 10 is an FMCW radar type measuring apparatus configured by blocks shown in FIG. 20, and a transducer 106 is interposed between the measuring apparatus 10 and the upper end of the circular waveguide 30. The transducer 106 is a transmission antenna that converts the high-frequency signal output from the measurement apparatus 10 into a radio wave and radiates it into the circular waveguide 30, and a reception antenna that receives the radio wave reflected by the liquid surface of the liquid 41. Function.

The measuring device 10 uses, for example, a 10 GHz band radio wave. The distance from the top plate surface to the surface of the liquid 41 is L, and the inner diameter of the circular waveguide 30 is d. F _T is the frequency of the radio waves radiated to the circular waveguide 30 from the measuring device 10 toward the liquid surface, F _R is towards the measuring device 10 a circular waveguide 30 above wave is reflected by the liquid surface This shows the frequency of the radio wave that returns.

In order to calculate the distance L by the measuring device 10, the speed of the transmitted radio wave and the reflected wave must be clear. It is known that the propagation speed of radio waves in free space is the same as the speed of light C. On the other hand, as shown in FIG. 19, when the circular waveguide 30 is used, if the wavelength of the radio wave in free space is λ ₀ , the wavelength λ _gε of the radio wave in the circular waveguide 30 is
λ _gε = ε _r × (λ ₀ / (ε _r − (λ ₀ / (K _mn × d)) ² ) ^1/2 )
It is known that it is required. Here, Kmn is a propagation coefficient in the circular waveguide 30 and εr is a relative dielectric constant in the circular waveguide 30 (propagation path).

The radio wave propagation coefficient Kmn in the circular waveguide 30 is generally known. When the inside of the circular waveguide is filled with air, the relative permittivity in the air is 1. Therefore, by determining the inner diameter d of the circular waveguide 30, the wavelength λ in the circular waveguide 30 is determined. _gε is calculated. By using this, the velocity v _g of the radio waves,
v _g = λ ₀ / λ _gε × C
Is calculated by

However, the liquid 41 is generated by evaporation in the space from the liquid level of the liquid 41 in the tank 40 to the ceiling surface of the tank 40, that is, in the radio wave propagation space (above the liquid level) from the measuring device 10 to the measurement target point. When the gas stays, the relative dielectric constant εr of the space changes under the influence of the staying gas component and the temperature and pressure of the propagation space. When the specific dielectric constant εr is changed, the wavelength lambda _Jiipushiron in the circular waveguide 30, and even changes the distance measurement also changes the speed v _g of the radio waves. Therefore, a correct distance measurement value cannot be obtained unless the relative dielectric constant of the gas staying in the radio wave propagation space in the tank 40 is obtained.

  If the component, temperature, and pressure of the gas in the tank 40 are known in advance, a dielectric constant corresponding to the gas component can be preset in the measuring device, and the distance L can be calculated using the accurate radio wave velocity. However, the components of the stagnant gas, the temperature and pressure at the time of measurement vary depending on the situation, and unless these are measured accurately, the distance L based on the accurate dielectric constant cannot be calculated.

  In response to the above problems, a liquid level measurement system that can accurately measure the liquid level even if there is a stagnant gas above the liquid level, assuming that the dielectric constant in the circular waveguide is within a predetermined numerical range. Is known (see, for example, Patent Document 1).

  In the liquid level measurement system described in Patent Document 1, even if there is no information such as accurate gas components or pressure on the liquid level, the measurement result is within the allowable range of measurement error due to the assumed relative dielectric constant variation range. It is effective if it exists. However, the distance to the liquid level cannot be measured with higher measurement accuracy. In order to accurately measure the distance to the object to be measured in an FMCW radar type measuring device, the relative dielectric constant of the radio wave propagation space is accurately estimated at the time of measurement, and the measured value is calibrated using the estimated dielectric constant. Good.

  Factors that reduce the measurement accuracy of the liquid level measuring device include the deterioration of the linearity of the VCO that generates the sweep frequency signal in addition to the change in the relative permittivity in the radio wave propagation space as described above. There are also various factors of deterioration of sex.

  The VCO 103 used in the transmission system of the FMCW radar shown in FIG. 20 is generally used in a PLL (Phase Locked Loop) circuit that can oscillate a signal with a stable frequency. However, since the frequency sweep speed is fast in the FMCW radar, the frequency to be transmitted is controlled by directly changing the control voltage of the VCO without using the PLL circuit. The linearity of the frequency-voltage characteristic (FV characteristic) output from the VCO 103 is insufficient to obtain the measurement accuracy required by the FMCW radar. In the FMCW radar, in order to guarantee the linearity of the FT characteristic output from the VCO 103, a corrected voltage-time table (hereinafter referred to as a "corrected VT curve table") in which the FT characteristic is corrected. ) And the oscillation voltage of the VCO 103 is controlled using this table.

  However, even if the linearity of the FV curve output from the VCO is guaranteed, in the liquid level measuring device using the waveguide in the transmission path, the frequency of the radio wave propagating in the waveguide as described above. Measurement accuracy deteriorates depending on the characteristics. In addition, even if the linearity of the FV curve output from the VCO is guaranteed, the FV characteristics fluctuate over time due to changes in usage conditions, and the linearity of the FV curve is maintained. Measurement accuracy is degraded.

Further, as described above, velocity v _g of the electromagnetic wave of the circular waveguide is because it is a function of frequency, the propagation velocity has a frequency characteristic. This changes the beat frequency of the FMCW radar with the frequency sweep, and as a result, decreases the measurement accuracy of the liquid level measuring device. Therefore, in order to increase the measurement accuracy of the liquid level measuring device, it is necessary to consider the frequency characteristics of the velocity of the radio wave propagating in the waveguide.

Japanese Patent No. 4695394

  The present invention further corrects the frequency characteristics of radio wave propagation in a circular waveguide by calculating the relative permittivity of the space in which the radio wave propagates, and further maintaining the linearity of the FV curve of the VCO. Accordingly, an object of the present invention is to provide a liquid level measuring apparatus that enables measurement with higher accuracy.

The liquid level measuring device according to the present invention is:
Through a circular waveguide, a transmitter that transmits radio waves toward the liquid surface while sweeping the frequency for a predetermined sweep time, a receiver that receives the radio waves reflected by the liquid surface, and the radio waves received An FMCW radar-type liquid level measurement device that has an extraction unit that extracts a beat frequency that is a difference between a transmission frequency and a reception frequency at the time of measurement, and measures a distance to the liquid level based on the beat frequency. And
A VCO that generates a sweep frequency signal;
A voltage-time table (hereinafter referred to as “V-T table”) for sweeping the frequency of a signal generated by the VCO;
A reflecting member disposed at a known distance position to correct the linearity of the frequency-time characteristic (hereinafter referred to as “FT characteristic”) of the VCO;
When the frequency sweep signal generated by the VCO is transmitted toward the reflection member while the linearity of the FT characteristic is maintained, the beat generated by the transmission signal and the reflection signal from the reflection member A fixed sine wave signal generation unit for obtaining a fixed sine wave signal having the same waveform;
When the frequency sweep signal generated by the VCO is transmitted toward the reflecting member with the FT characteristic before correction, the fixed sine of the beat waveform generated by the transmission signal and the reflected signal from the reflecting member A correction voltage-time curve table (hereinafter referred to as “correction VT curve table”) generated by analyzing an error with respect to a wave signal;
A corrected VT table obtained by correcting the VT table with the corrected VT curve table,
The corrected VT table is used for the operation of the VCO to generate FT characteristics having linearity,
A spectrum analyzer for analyzing the difference between the beat frequency and the sweep frequency;
A predistortion table generating unit for generating a predistortion table for further correcting the corrected VT table in accordance with the degree of change in the difference of the beat frequency from the analysis result by the spectrum analyzing unit;
A predistortion VT curve table generating unit that generates the predistortion VT curve table by applying the predistortion table to the corrected VT table;
A memory for storing the predistortion VT curve table and serving as a control signal for generating the sweep frequency signal from the predistortion VT curve table to the VCO.

  According to the liquid level measurement device of the present invention, in the FMCW radar type liquid level measurement, the liquid level can be calculated by applying the calculated relative dielectric constant in order to calculate the relative dielectric constant of the system to be measured. In addition, calibration is performed to correct the VT table applied to the VCO according to the characteristics of the VCO constituting the liquid level measuring device, and to obtain the FT characteristics having linearity. By executing the VCO predistortion, the liquid level can be measured with higher accuracy without causing an error in the frequency of the beat signal.

It is a block diagram which shows one Example of the liquid level measuring apparatus which concerns on this invention. It is a functional block diagram of the dielectric constant calculation program execution part in the said Example. It is a flowchart which shows the flow of the process by the said dielectric constant calculation program. It is a block diagram which shows another Example of the liquid level measuring apparatus which concerns on this invention. It is a functional block diagram of the dielectric constant calculation program execution part in the said another Example. It is a flowchart which shows the flow of the process by the said dielectric constant calculation program. It is a front view which shows the example of the planar transducer corresponding to 2 transmission modes suitable for the liquid level measuring apparatus which concerns on this invention. It is a center part longitudinal cross-sectional view of the said transducer. It is a connection diagram which shows roughly the electric power feeding circuit example of the said transducer. It is a model figure which shows distribution of the electric field and magnetic field in TE11 mode in the circular waveguide by the said transducer. It is a model figure which shows distribution of the electric field and magnetic field in TE01 mode in the circular waveguide by the said transducer. It is a model figure which shows the relationship between an electric current and a magnetic field when there is no excitation pin in the side of a slot in TE01 mode. FIG. 5 is a model diagram showing a relationship between a current and a magnetic field when an excitation pin is provided on one side of a slot in the TE01 mode. It is a graph which shows the process of obtaining the FT characteristic before correction | amendment in the said Example. It is a graph which shows the process of obtaining the FT characteristic after correction | amendment in the said Example. (A) shows the frequency sweep when the linearity of the FT characteristic is maintained, (b) shows the frequency sweep when the linearity of the FT characteristic is not maintained (c) ) Is a graph showing the difference in beat waveform depending on the presence or absence of linearity of the FT characteristic. It is a flowchart which shows the process of the calibration process performed in the said Example in order. It is a flowchart which shows in order the process of the VCO predistortion process performed in the said Example. It is a schematic diagram which shows notionally the mode of measurement by the liquid level measuring apparatus which concerns on this invention. It is a block diagram which shows the example of the conventional liquid level measuring apparatus. It is a graph which shows the example of the frequency sweep by FMCW radar. It is a graph which shows the distance measurement principle by the frequency sweep of FMCW radar. It is a perspective view which shows the example of a waveguide whose cross section is a square. It is a perspective view which shows the example of a waveguide with a circular cross section. It is a graph which shows the example of the propagation speed with respect to the frequency of the electromagnetic wave in a waveguide. It is a graph which shows the relationship between the sweep frequency and beat frequency of the liquid level measuring apparatus using a waveguide. It is a graph which compares and shows the spectrum waveform obtained by carrying out FFT processing of the beat signal by free space reflection, and the beat signal by reflection in a circular waveguide.

  Embodiments of a liquid level measuring device according to the present invention will be described below with reference to the drawings.

  In FIG. 1, a liquid level measuring device 1 includes a digital signal processing device (hereinafter referred to as “DSP”) 11, a digital / analog converter (hereinafter referred to as “DAC”) 12, a VCO 13, a coupling circuit 14, a switch 22, and the like. , A transducer 16, a mixer 17, an AGC 18, and an analog / digital converter (hereinafter referred to as “ADC”) 19. The part including the DAC 12 and the VCO 13 constitutes a transmission system, and the part from the mixer 17 to the ADC 19 constitutes a reception system. The switch 22 switches the coupling circuit 14 to the near-end reflecting member 26 side at the time of VCO calibration, and switches to the transducer 16 side at the time of liquid level measurement.

  The DSP 11 has a built-in memory 20. The memory 20 stores a plurality of VT tables. The VT table is obtained by converting a plurality of VT curves that define the relationship between the applied voltage V to the VCO 13 and the sweep time T, and each VT table has a data with respect to the sweep time T. This is a data group for determining the oscillation frequency of the VCO 13. In addition, the memory 20 stores a program for executing a calculation process described later.

The configuration from the DAC 12 to the ADC 19 is substantially the same as the configuration of the FMCW radar type distance measuring apparatus shown in FIG. Thus, the process of extracting the beat frequency F _B is also the same as the process in the measurement unit of the conventional FMCW radar system. That is, the DSP 11 reads a predetermined VT table based on a program stored in the memory 20, and the read VT table is a voltage that continuously changes over time by the DAC 12. Converted to a value (analog signal).

  The analog signal becomes the control voltage of the VCO 13. When this control voltage is applied to the VCO 13, the frequency of the oscillated signal (oscillation frequency) changes continuously. The oscillation signal whose frequency changes continuously is converted into a radio wave by the transducer 16 which is an antenna through the coupling circuit 14. This radio wave is transmitted toward an object to be measured (for example, a liquid level of liquid fuel stored in a liquid tank).

A measured system 31 is interposed between the transducer 16 and the measurement target point. The radio wave reflected at the measurement target point is captured by the transducer 16 through the system under measurement 31. The reflected wave captured by the transducer 16 is converted into a reception signal and guided to the reception system via the coupling circuit 14. In the receiving system, the mixer 17 mixes the received signal and the oscillation signal at the moment when the received signal is received, and generates a beat signal based on the difference between the received frequency and the oscillation frequency. The frequency of the beat signal that beat frequency F _B.

The beat signal is controlled to an appropriate amplitude value by the AGC 18, converted to a digital signal by the ADC 19, and input to the DSP 11. In the DSP 11, the process from the reading process of the VT data to the process of taking in the beat signal is executed during the sweep time (T). A filtering process is performed on the beat signal group, which is the time axis data captured during the sweep time (T). It performs FFT processing on unwanted beat signal group from which the noise components have been removed by the filtering process to extract the beat frequency F _B. Signal representative of the extracted beat frequency F _B is stored in the memory 20 associated with the V-T table became its original.

  In the memory 20 of the DSP 11, in addition to the program for executing the above processing, a program for outputting a plurality of high-frequency signals from the VCO 13 under different conditions, and a system under test based on the beat frequency extracted for each different condition A program for calculating the relative dielectric constant of 31, that is, a relative dielectric constant calculation program is stored. FIG. 2 shows a relative permittivity calculation program 200 stored in the memory 20.

[Relative permittivity calculation program 1]
As shown in FIG. 2, the relative dielectric constant calculation program 200 includes a VT table selection instruction unit 201, a beat frequency storage unit 202, a relative dielectric constant calculation unit 203, a transmission frequency instruction unit 211, and an AGC control unit. 212, a filtering processing unit 213, and a beat frequency extracting unit 214.

  The VT table selection instruction unit 201 selects a specific VT table from a plurality of VT tables stored in the memory 20. By this VT table selection process, radio waves can be transmitted under different transmission conditions.

  Based on the VT table selected by the VT table selection instruction unit 201, the transmission frequency instruction unit 211 outputs to the DAC 12 data based on a voltage value that continuously changes over time.

  The AGC control unit 212 adjusts the amplitude of the beat signal of the AGC 18 based on the control data stored in the memory 20 in order to adjust the amplitude of the beat signal input from the mixer 17 to the AGC 18 to an appropriate value. Control.

The filtering processing unit 213 removes unnecessary noise components from the beat signal captured by the ADC 19. Beat frequency extraction section 214 performs FFT processing on the beat signal, converts the beat signal group to the data of the frequency axis, and extracts the beat frequency F _B from the transformed data.

Beat frequency storage unit 202, a signal representative of the beat frequency F _B extracted in beat frequency extraction section 214, in association with the V-T table selected in V-T table selection instruction unit 201, and stores in the memory 20 .

Dielectric constant calculating unit 203, using a signal representative of the beat frequency F _B stored in the memory 20 in the beat frequency storage unit 202, performs a relative dielectric constant calculated based on the equations 2 below. The relative dielectric constant calculating unit 203, the dielectric constant calculation processing until ready signal representative of the beat frequency F _B required, select a different V-T table by V-T table selection instruction unit 201, the selected V also performs overall control of processing to extract beat frequency F _B on the basis of the -T table.

Formula 2

Each variable in Equation 2 is as follows.
λ ₀₁ : Wavelength F _B1 related to the first sweep frequency transmitted by the first VT table: First beat frequency λ ₀₂ based on the VT table related to the first sweep frequency λ ₀₂ : Second V- Wavelength F _B2 related to the second sweep frequency transmitted by the T table: Second beat frequency K _mn based on the _VT table related to the second sweep frequency: Propagation constant d of the circular waveguide 30 d: Circular waveguide The inner diameter of the tube 30 The numerical values related to λ ₀₁ , λ ₀₂ , d, and K _mn are stored in the memory 20 in advance.

[Relative permittivity calculation processing flow 1]
Next, an example of processing by the relative permittivity calculation program will be described with reference to FIG. Each processing step is expressed as S11, S12, S13,.

  First, the VT table selection instruction unit 201 selects a specific VT table from the VT tables stored in the memory 20. Since a plurality of VT tables are stored in the memory 20, a VT table that meets the first condition is selected from the VT tables.

  Next, the control voltage data stored in the selected VT table (first table) is continuously read out by the transmission frequency instruction unit 211 and input to the DAC 12. The control voltage data is converted into an analog control voltage signal by the DAC 12.

  The VCO 13 outputs a high-frequency signal (microwave signal) related to the first sweep frequency in accordance with the input control voltage signal. The high-frequency signal having the first sweep frequency output from the VCO 13 is converted into a radio wave by the transducer 16 via the coupling circuit 14 and transmitted to the system under test 31 (S11).

The system under test 31 includes an object to be measured (for example, the liquid surface of the liquid 41 that is a measurement target point), and the transmitted radio wave is reflected by the object to be measured. The reflected radio wave passes through the system to be measured 31, is then captured by the transducer 16, converted into a reception signal, and guided to the reception system via the coupling circuit 14. In the reception system, the reception signal is mixed with the transmission signal at that time in the mixer 17, and a beat signal that is the difference in frequency between the two signals is output. The beat signal is controlled to a predetermined amplitude by the AGC 18, converted to a digital signal by the ADC 19, and input to the DSP 11. In DSP 11, and FFT processing a beat signal, extracting a first beat frequency _{F B1} (S12).

The extracted signal representing the first beat frequency F _B1 is stored in the memory 20 in association with the first table by the beat frequency storage unit 202.

  Next, the VT table selection instruction unit 201 selects a VT table (second table) that matches the second condition from the VT tables stored in the memory 20. Next, the control voltage data stored in the selected VT table (second table) is continuously read out by the transmission frequency instruction unit 211 and converted into an analog signal by the DAC 12. This analog signal becomes a control voltage signal for the VCO 13.

  The VCO 13 outputs a second sweep frequency signal, which is a high frequency signal, according to the input control voltage signal. The second sweep frequency signal passes through the coupling circuit 14 and is then converted into a radio wave by the transducer 16 and transmitted to the system under test 31 (S13).

Thereafter, as in the case of the first sweep frequency signal, the radio wave is reflected by the object to be measured in the system under measurement 31, is captured by the transducer 16 through the system under measurement 31, and is converted into a received signal. The reception signal is guided to the reception system through the coupling circuit 14, and a beat signal with the transmission signal at that time is output by the mixer 17 of the reception system. This beat signal is input to the DSP 11 via the AGC 18 and ADC 19. In DSP 11, the beat signal according to the second sweep frequency to FFT processing, to extract the second beat frequency _{F B2} (S14). The extracted signal representing the second beat frequency _FB2 is stored in the memory 20 in association with the second table by the beat frequency storage unit 202.

Next, the relative permittivity calculating unit 203 stores the wavelength λ ₀₁ of the oscillation frequency related to the first sweep frequency, the first beat frequency F _B1 related to the first sweep frequency, and the second sweep stored in the memory 20. The wavelength λ _{02 of the} oscillation frequency related to the frequency, the propagation constant K _{mn of} the circular waveguide, the inner diameter d of the circular waveguide 30, and the second beat frequency F _B2 related to the second sweep frequency are read out. Using the read out λ ₀₁ , λ ₀₂ , d, K _mn , F _B1 , and F _B2 , the calculation process based on Equation 2 is performed. By this calculation process, the relative dielectric constant ε _r is calculated (S15).

  As described above, the relative dielectric constant can be calculated by actual measurement in an actual measurement environment using the FMCW radar type measurement device. The liquid level can be measured with higher accuracy by applying the calculated dielectric constant to the FMCW radar type measuring device, calculating the velocity of the radio wave corresponding to the dielectric constant, and correcting the measurement distance L. it can.

[Relative permittivity calculation program 2]
Next, another example of a relative permittivity calculation program applicable to the present invention will be described. In FIG. 4, the liquid level measuring apparatus 1a includes a DSP 11a, a DAC 12, a VCO 13, a coupling circuit 14, a switch 22, a switch 15, a transducer 16a, a transducer 16b, a mixer 17, an AGC 18, and an ADC 19. The part including the DAC 12 and the VCO 13 constitutes a transmission system, and the part from the mixer 17 to the ADC 19 constitutes a reception system. The switch 22 switches the coupling circuit 14 to the near-end reflecting member 26 side at the time of VCO calibration, and switches to the switch 15 side at the time of liquid level measurement.

  The DSP 11a has a built-in memory 20a. In the memory 20a, a VT table is stored as in the above example. Further, the memory 20a stores a program for executing a calculation process described later.

  The configuration from the DAC 12 to the ADC 19 is almost the same as the configuration of the above-described embodiment. The different configurations are DSP 11a, switch 15, transducers 16a and 16b. The transducer 16a and the transducer 16b are antennas having different transmission modes. For example, the transducer 16a is an antenna corresponding to the TE01 mode, and the transducer 16b is an antenna corresponding to the TE11 mode.

  The switch 15 switches to either the transducer 16a or the transducer 16b according to a control signal from the DSP 11a. Accordingly, the switch 15 can transmit and receive radio waves by alternately switching the transmission mode of the liquid level measuring device 1a to two different transmission modes.

  In addition to the program already described, the memory 11a of the DSP 11a calculates the relative dielectric constant of the system under test 31 based on the program for selectively switching the transmission mode and the beat frequency extracted for each different transmission mode. A relative permittivity calculation program comprising a program to store is stored. FIG. 5 is a functional block diagram of the relative permittivity calculation program 200a stored in the memory 20a.

[Relative permittivity calculation program 2]
As shown in FIG. 5, the relative permittivity calculation program 200a includes a transmission mode instruction unit 204, a transmission frequency instruction unit 211, an AGC control unit 212, a filtering processing unit 213, a beat frequency extraction unit 214, and a beat frequency. A storage unit 202 and a relative dielectric constant calculation unit 203a are provided. The relative dielectric constant calculation program 200a is different from the above-described example shown in FIG. 2 in that a transmission mode instruction unit 204 instead of the VT table selection instruction unit 201 is provided and the relative dielectric constant calculation unit 203a is different. It is that you are.

  The transmission mode instruction unit 204 includes a program for controlling the operation of the switch 15, and the switch 15 selects one of the transducer 16a and the transducer 16b by this program. Radio waves are transmitted under different transmission conditions depending on whether the switch 15 selects the transducer 16a or the transducer 16b. The different transmission conditions are different radio wave transmission modes. In the relative permittivity calculation program 2, the radio wave transmission mode in the circular waveguide 30 is switched to the TE01 mode or the TE11 mode.

  The transmission frequency instruction unit 211, the AGC control unit 212, the filtering processing unit 213, and the beat frequency extraction unit 214 perform the same processing as the corresponding portions in the above example.

Beat frequency storage unit 202, a signal representative of the beat frequency F _B extracted in beat frequency extraction section 214, and stored in the memory 20a in the transmission mode each time are switched in the transmission mode instruction unit 204.

Dielectric constant calculating unit 203a, by using a signal representative of the beat frequency F _B stored in the memory 20a in the beat frequency storage unit 202, performs a relative dielectric constant calculated based on the equations 3 below. The relative dielectric constant calculating unit 203a until ready signal representative of the beat frequency F _B required dielectric constant calculation processing, switching the switch 15 by the transmission mode instruction unit 204, the beat frequency F _B based on different transmission modes Let it be extracted.

Formula 3

Each variable in Equation 3 is as follows.
lambda _0: wavelength radio wave _{F B0:} first transmission beat frequency _K according to Mode _mn0: propagation constant _{F B1} of the circular waveguide 30 of the first transmission mode: beat frequency _{K mn1} of the second transmission mode : Propagation constant of the circular waveguide 30 according to the second transmission mode d: Inner diameter of the circular waveguide 30 Note that numerical values related to λ ₀ , K _mn0 , K _mn1 , d are stored in the memory 20a in advance.

[Relative permittivity calculation processing flow 2]
Next, the process flow of the relative dielectric constant calculation program 2 will be described with reference to the flowchart of FIG. First, a control signal is sent to the switch 15 by the transmission mode instruction unit 204, and the switch 15 selects a transmission mode (first transmission mode) corresponding to the first condition. It is assumed that radio waves are transmitted and received by the transducer 16a under the first condition.

  Next, the control voltage data is continuously read from the VT table stored in the memory 20 a by the transmission frequency instruction unit 211 and input to the DAC 12. The DAC 12 converts the control voltage data into an analog signal.

  A high frequency signal having a frequency corresponding to the input control voltage signal is output from the VCO 13. The high-frequency signal output from the VCO 13 is converted into a radio wave by the transducer 16a after passing through the coupling circuit 14, and transmitted to the system under test 31 (S11a).

The transmitted radio wave in the first transmission mode is reflected by the object to be measured, and is guided to the receiving system via the measured system 31 via the transducer 16a and the coupling circuit 14. In the reception system, as already described, predetermined processing is performed by the mixer 17, the AGC 18, and the ADC 19, and the beat signal of the received signal and the oscillation signal at that time is input to the DSP 11a. The beat signal to FFT processing in DSP11a, extracts the beat frequency _{F B0} according to the first transmission mode (S12a). Signal representative of the beat frequency F _B0 which is extracted, the beat frequency storage unit 202 of the memory 20a, and stored as a signal representative of the beat frequency F _B0 according to the first transmission mode.

  Next, a control signal is sent from the transmission mode instructing unit 204 to the switch 15, and the transmission mode (second transmission mode) corresponding to the second condition is selected by the switch 15. Subsequently, control voltage data is continuously read from the VT table stored in the memory 20 a by the transmission frequency instruction unit 211 and input to the DAC 12.

  Thereafter, similarly to the operation in the first transmission mode, a high-frequency signal is output from the VCO 13 and input to the transducer 16b via the coupling circuit 14. From the transducer 16b, the signal is transmitted into the circular waveguide 30 toward the system under test 31 in a second transmission mode different from the first transmission mode.

The radio wave is reflected by the object to be measured, received by the transducer 16b via the system to be measured 31, and a beat signal that is a frequency difference between the received signal and the transmission signal at that time is extracted by the mixer 17 in the receiving system. . Beat signal signal representative of the beat frequency _{F B1} according to the second transmission mode is FFT processed by DSP11a is extracted (S14a). Signal representative of the beat frequency F _B1, which is extracted, the beat frequency storage unit 202 of the memory 20a, and stored as a signal representative of the beat frequency F _B1 according to the second transmission mode.

Next, the relative permittivity calculating unit 203a transmits the wavelength λ ₀ of the radio wave stored in the memory 20, the beat frequency F _B0 related to the first transmission mode, and the propagation of the circular waveguide 30 related to the first transmission mode. The constant K _mn0 , the propagation constant K _{mn1 of} the circular waveguide 30 related to the second transmission mode, the inner diameter d of the circular waveguide 30 and the beat frequency F _B1 related to the second transmission mode are read out. Using the read out λ ₀ , F _B0 , K _mn0 , K _mn1 , d, and F _B1 , the calculation process based on Equation 3 is performed. By this calculation process, the relative dielectric constant ε _r is calculated (S15a).

  As described above, the relative dielectric constant can be calculated by actual measurement in an actual measurement environment using the FMCW radar type measurement device. The liquid level can be measured with higher accuracy by applying the calculated dielectric constant to the FMCW radar type measuring device, calculating the velocity of the radio wave corresponding to the dielectric constant, and correcting the measurement distance L. it can.

[2-transmission mode compatible planar transducer]
In order to estimate the relative permittivity of the radio wave propagation space using the two transmission modes as described above, a transducer corresponding to the two transmission modes is required. In addition, the transducer corresponding to the two transmission modes is preferably a planar transducer that can transmit and receive at least in the same plane. An example of a novel two-transmission mode-compatible planar transducer that satisfies such conditions will be described below.

  7 and 8, a radial waveguide in which two circular conductor plates 57 and 58 are placed in parallel constitutes the main body of a two-mode flat transducer. A cavity 59 is formed between the two circular conductor plates 57 and 58. As shown in FIG. 8, a hole is formed in the center of the lower conductor plate 57 to connect the outer conductor 64 of the coaxial line to the conductor plate 57 and connect the inner conductor 63 of the coaxial line to the upper conductor plate 58. Is a common form of radial waveguide. In order to widen the frequency bandwidth, the end portion of the inner conductor 63 is a tapered end portion 631 that is widened in a tapered shape. Although there is a gap between the tapered end 631 and the upper conductor plate 58, there is no problem in transmitting a high-frequency signal.

  Further, the upper and lower conductor plates 57 and 58 constituting the radial waveguide are short-circuited on the circumference of the end portion. A plurality of slots 61 are formed at equal intervals in the circumferential direction on the upper conductor plate 58 of the radial waveguide, and excitation pins 62 for short-circuiting the upper and lower conductor plates 57 and 58 are provided on one side of each slot 61. As a result, a TE01 mode transducer 51 is formed, and the TE01 mode is excited in the circular waveguide.

  As the transducer 32 in the usage example of the liquid level measuring device shown in FIG. 19, the above-described two-transmission mode-compatible planar transducer is used. The radial waveguide shown in FIGS. 7 and 8 is coupled upside down to the upper end of the circular waveguide 30 shown in FIG. 19, and the upper end of the circular waveguide 30 is short-circuited by the radial waveguide. In other words, the upper end of the circular waveguide 30 is concentrically covered with the radial waveguide shown in FIGS. Each slot 61 shown in FIG. 7 is set so as to be within the inner diameter d of the circular waveguide 30.

  7 corresponds to the case where the hollow long circular waveguide 30 is placed vertically on the paper surface of FIG. In FIG. 19, the upper end of the circular waveguide 30 is short-circuited by a conductor plate 58. Hereinafter, the conductor plate 58 is referred to as a “short-circuit conductor plate 58”. In this short-circuit conductor plate 58, only eight slots 61 for exciting the TE01 mode are arranged at equal intervals in the circumferential direction of the short-circuit conductor plate 58.

  FIG. 11 schematically shows the state of excitation in the TE01 mode in the circular waveguide 30. A solid line with an arrow represents a line of electric force, and a broken line with an arrow represents a line of magnetic force. As shown in FIG. 11, the electric field is distributed concentrically with the circular waveguide 30 in the TE01 mode. Dotted lines that are perpendicular to the electric field at each point indicate the magnetic field. Both the electric field and the magnetic field at the center of the circular waveguide 30 are reduced. This corresponds to the fact that the central portion of the short-circuit conductor plate 58 shown in FIG. 7 is a wide space for the TE01 mode.

  A loop 71 for exciting the TE11 mode is provided in the circular waveguide 30 as shown in FIGS. 7 and 8 in a wide area at the center of the short-circuit conductor plate 58. A microstrip line 73 is provided in the radial direction of the short-circuit conductor plate 58 along the surface of the short-circuit conductor plate 58 as a line for feeding power to the loop 71. The loop 71 forms a TE11 mode transducer 52, and the tip of the loop 71 is grounded to the short-circuit conductor plate 58 via a termination resistor 74. As described above, the loop 21 is manufactured as a part of the planar circuit on the short-circuit conductor plate 58. The short-circuit conductor plate 58 is also a grounding plate of the radial waveguide shown in FIG. When a high frequency current is fed to the loop 71 using the microstrip line 73, a magnetic field is generated inside the circular waveguide 30. This magnetic field generates an electric field from Faraday's law. These electromagnetic fields are in the TE11 mode.

  The state of the TE11 mode electromagnetic field is modeled in FIG. 10 in accordance with FIG. 11 showing the state of the TE01 mode electromagnetic field. A solid line with an arrow represents a line of electric force, and a broken line represents a line of magnetic force. As shown in FIG. 10, the electric lines of force in the TE11 mode are bisected in an antisymmetrical manner by the center line in the horizontal direction and are directed from the lower side to the upper side, and are perpendicular to the inner wall surface of the circular waveguide 30. It is distributed so that it goes in the direction and enters at right angles. In the TE11 mode, the magnetic field lines are oriented in the direction perpendicular to the electric field lines at each point of the electric field lines as in the TE01 mode.

  Explained in detail with reference to FIGS. 12 and 13 is excitation in the TE01 mode by providing the short-circuit conductor plate 58 with the slot 61 and the excitation pin 62 on one side of the slot 61. FIG. 12 and 13 both show one of the plurality of slots 61 shown in FIG. A high frequency signal is supplied to the short-circuit conductor plate 58 from the inner conductor 63 of the coaxial line to the center of the short-circuit conductor plate 58. The high-frequency signal current flows radially through the short-circuit conductor plate 58 from the center to the outside in the radial direction. If the wavelength of the high frequency signal to be excited is λ, the length of the slot 61 in the longitudinal direction is about λ / 2. The width dimension of the slot 61 is considerably smaller than a fraction of the longitudinal dimension λ / 2 of the slot 61.

  FIG. 12 shows an example in which no excitation pin is provided on the side of the slot 61, a current J flows in the length direction of the slot 61, and a magnetic field M is generated in the width direction of the slot 61, which is a direction perpendicular to the current J. Show. Since the dimension in the width direction of the slot 61 is too small with respect to the wavelength λ of the high-frequency signal to be excited, the magnetic field lines cannot pass through the slot 61 and no electromagnetic wave is radiated from the slot 61.

FIG. 13 shows a case where an excitation pin 62 is provided on one side of the slot 61 and a current J flows in the longitudinal direction of the slot 61. As described above, the magnetic field M is generated in a direction orthogonal to the direction of the current J. When the magnetic field M attempts to cross the slot 61 in the width direction, the magnetic field M is excited on one side of the slot 61. It is disturbed by 62 and crosses the slot 61 while meandering around the excitation pin 62. Therefore, the magnetic field M across the edge of the slot 61 on the excitation pin 62 side is inclined with respect to the edge of the slot 61. This oblique magnetic field M can be divided into a component M _{1 in} the width direction of the slot 61 and a component M ₂ in the length direction of the slot 61. A magnetic field line in the width direction of the slot 61 is generated by the component M ₁ of the magnetic field M, and a magnetic field line in the length direction of the slot 61 is generated by the component M ₂ of the magnetic field M.

The magnetic field component M ₁ is the same as the magnetic field M, and no electromagnetic wave is radiated from the slot 61 as already described. In contrast the magnetic field component M ₂ generated in the longitudinal direction of the slot 61, since the length dimension of the slot 61 with respect to the wavelength lambda is lambda / 2, the magnetic field lines representing the magnetic field component M ₂ Can pass through the slot 61, and electromagnetic waves are radiated from the slot 61. As shown in FIG. 7, when a plurality of slots 61 are provided in the short-circuit conductor plate 8 and an excitation pin 62 is provided on one side of each slot 61, an electromagnetic wave having the same frequency is generated in each slot 61, and this electromagnetic wave is radiated. Is done.

  As described above, the transducers shown in FIGS. 7 and 8 constitute a planar transducer corresponding to the two transmission modes. This transducer uses a circular waveguide 30 as a transmission path, and can use two types of electromagnetic fields shown in FIGS. 10 and 11 propagating in the circular waveguide 30. That is, there are a TE11 mode in which the loop 71 shown in FIGS. 7 and 8 excites, and a TE01 mode in which the high-frequency current flowing through the inner conductor 63 of the coaxial line shown in FIG.

  In the example shown in FIG. 7, although the widthwise dimension of each slot 61 is considerably smaller than the longitudinal dimension of each slot 61, it has a relatively wide slot shape. As a result, the frequency bandwidth of the excitable high-frequency signal can be widened, so that a planar transducer suitable for an FMCW radar type liquid level measuring device that measures the distance to the liquid level while sweeping the frequency is obtained. Can do.

  In FIG. 9, reference numeral 16a denotes a TE01 mode transducer, and 16b denotes a TE11 mode transducer. The TE01 mode transducer 16a and the TE11 mode transducer 16b are supplied with power from the common transceiver 3. An input / output line drawn from one transceiver 3 is configured to be selected and connected by a switch 4 to one of two branched lines. One branch line is connected to the transducer 16 a via the connector 5, and the other branch line is connected to the transducer 16 b via the connector 6.

  As described above, the signals supplied to the two transducers 16a and 16b are the same, but the propagation modes of the radio waves converted and radiated by the two transducers 16a and 16b are different. By switching the switch 4, one transducer corresponding to the TE01 mode or the TE11 mode is selected. The switch 4 corresponds to the switch 15 in the embodiment of FIG. The transceiver 3 corresponds to the components including the DSP 11a, the transmission system, and the reception system in the embodiment of FIG.

  If the FMCW radar type liquid level measuring device is constructed by attaching the above-mentioned two-transmission mode-compatible planar transducer to the upper end of the circular waveguide 30 having an inner diameter d, it is convenient for estimating the relative permittivity of the electromagnetic wave transmission space. is there. The transmission characteristics of electromagnetic waves differ depending on the relative permittivity of the transmission space, and the beat frequency differs. Therefore, the relative permittivity in the waveguide through which the electromagnetic waves propagate is obtained, and the measurement results are corrected according to the relative permittivity. Therefore, a highly accurate measurement can be performed. Since the two-transmission-mode-compatible planar transducer can be used as a TE01 mode and TE11 mode transducer, the relative dielectric constant in the waveguide can be estimated from the difference between the measured values in the two modes. .

  In addition to the fact that the measured value fluctuates depending on the relative permittivity of the electromagnetic wave transmission space, there are the following factors as factors that lower the measurement accuracy of the FMCW radar type liquid level measuring device. The first factor for reducing accuracy is that the characteristics of the VCO included in the liquid level measuring device fluctuate with temperature and that the characteristics of the VCO itself change over time. The second factor for reducing accuracy is that the velocity of the radio wave propagating in the waveguide has frequency characteristics when the waveguide is used in the transmission path.

  In the embodiment described below, VCO calibration is executed for the first accuracy reduction factor, and measurement accuracy is improved by executing VCO predistortion for the second accuracy reduction factor. I am trying.

[VCO calibration]
1 and 4, the switch 22 connects the coupling circuit 14 to the near-end reflecting member 26 side. The near-end reflecting member 26 is made of a member that can reflect radio waves radiated from the liquid level measuring device, and is installed at a known length position from the reference plane position. The near-end reflecting member 26 is a member necessary for performing VCO calibration, which is a feature of the present embodiment. Next, the necessity of VCO calibration and the VCO calibration itself will be described.

  The relationship of the oscillation frequency to the control voltage in the VCO does not necessarily have linearity. In general, the characteristics change due to temperature changes or secular changes. FIG. 14A shows an example of a general FV characteristic of a VCO, and the linearity is broken. FIG. 14B shows a VT table before correction stored in the memory 20, and the control voltage changes linearly with respect to the time axis.

  Assume that the DSP 11 reads the VT table from the memory 20, converts a control voltage value that changes over time into an analog signal by the DAC 12, and applies it to the VCO 13. Even if the VT curve is a straight line, if the linearity of the FV characteristic of the VCO 13 is broken as described above, the FT characteristic representing the relationship of the oscillation frequency with respect to the time axis output from the VCO 13 is: As shown in FIG. 14C, the line is non-linear. Therefore, even if the liquid level is measured using this FT characteristic, it is not possible to perform measurement with good accuracy.

  Therefore, the VT curve stored in the memory 20 is corrected corresponding to the FV characteristic of the VCO 13 so that the FT characteristic of the signal output from the VCO 13 becomes a straight line. FIG. 15 shows this correction. FIG. 15 (a) shows the same non-linear FV characteristic of the VCO 13 as FIG. 14 (a). FIG. 15B shows a VT curve based on the corrected VT table stored in the memory 20. The corrected VT curve has a non-linearity opposite to the non-linear FV characteristic of the VCO 13.

  When the corrected VT curve table is read, and the control voltage value that changes over time is converted into an analog signal by the DAC 12 and applied to the VCO 13, linearity as shown in FIG. 15C is secured. The corrected FT characteristic can be obtained. When the liquid level is measured using the corrected FT characteristic, it is possible to perform measurement with good accuracy.

  The process of generating a corrected VT table corresponding to the FV characteristic of the VCO 13 as shown in FIG. 15B is referred to as VCO calibration (calibration) here. By performing VCO calibration, it is possible to obtain FT characteristics that maintain linearity, and to perform liquid level measurement with high accuracy.

The configuration for performing VCO calibration and the method of VCO calibration will be described in more detail. VCO calibration, as shown in FIG. 22, FMCW radar, when the linearity of the F-T characteristics are ensured through the sweep time, the beat frequency F _B signal caused by the reflection of radio waves from a known distance, Take advantage of having the same frequency throughout the sweep frequency band.

  In the embodiment shown in FIGS. 1 and 4, the VCO calibration is performed by the DSP 11 selecting the near-end reflecting member 26 side by controlling the operation of the switch 22 through the control signal line CAL. As described above, the near-end reflecting member 26 is installed at a known length position from the reference plane position. Therefore, if the linearity of the FT characteristic is ensured, the reflected signal from the near-end reflecting member 26 is obtained. The beat signal waveform obtained by receiving the signal becomes a fixed sine wave signal. On the other hand, in the case of a non-linear curve like the FT characteristic before correction shown in FIG. 14C, the beat frequency fluctuates within the sweep time T, so the beat signal waveform is accurate. Not a sine wave signal. FIG. 16 shows the difference in operation between the case where the linearity of the FT characteristic is maintained and the case where it is not maintained.

Figure 16 (a) shows a case where the beat frequency F _B in the sweep time T does not vary. Since the sweep frequency of the transmitted and received signals in the sweep time are both kept linearity, frequency F _B of the beat signal of the transmission signal and the received signal is constant everywhere within a sweep time. Accordingly, the waveform of the beat signal at this time is as indicated by the waveform A in FIG. 16 (c), the beat frequency F _B are accurate sine wave has no collapse of waveforms at a constant.

Figure 16 contrast (b), the beat frequency F _B in the sweep time T is varied, F-T characteristic shows a case made of a non-linearity. As the waveform of the beat signal at this time is shown in waveform B of FIG. 16 (c), the now waveform collapse from the waveform A is an accurate sine wave, the beat frequency F _B is deviated from the original beat frequency. Therefore, even seek distance L to the measurement point by applying the beat frequency F _B signals to the formula 1, it can not be obtained a highly accurate distance L, causing errors in the measurement results.

  FIG. 17 illustrates a VCO calibration procedure for correcting the VT table so that the FT characteristic of the output signal of the VCO 13 is a straight line in the embodiment shown in FIGS. The processing steps are shown in the vertical direction in the center of FIG. Each processing step is provided with a symbol such as S21, S22,.

  As a premise for performing the VCO calibration, the switch 22 is switched to the near-end reflecting member 26 side while the linearity of the FT characteristic is secured as described above, and the reflected signal from the near-end reflecting member 26 is changed. Receive. The near-end reflecting member 26 is a fixed-length reflecting point whose distance from the measuring device unit is accurately measured in advance. Since the distance to the near-end reflecting member 26 is determined in advance with high accuracy, a beat signal waveform of a fixed sine wave signal can be obtained by ensuring the linearity of the FT characteristic.

  The beat signal waveform of the fixed sine wave signal is referred to as a “fixed sine wave”, and the portion that generates the fixed sine wave signal is referred to as a “fixed sine wave signal generation unit”. The fixed sine wave signal generation unit exists in the DSP 11 as a part having one function. The FT characteristic in which linearity is ensured is set in advance so that the sweep frequency changes linearly with respect to the sweep time, and is stored as a VT table in the memory 20 of the DSP 11. .

Next, as shown in step S21 in FIG. 17, it performs sweeping the initial V-T table or the uncorrected V-T table, the frequency F _T and the radio wave of the radio wave radiated is reflected back by the liquid surface obtaining a beat waveform of a radio wave frequency _{F R} (S22). The sinusoidal waveform in the upper right of FIG. 17 shows an example of the beat waveform acquired as described above. Next, a shift of the beat waveform with respect to the fixed sine wave is analyzed while comparing the beat waveform acquired in step S22 with the fixed sine wave on the time axis (S23). The waveform diagram in the second stage on the right side of FIG. 17 shows an image of the beat waveform analysis, and it can be seen that the acquired beat waveform deviates on the time axis with respect to the fixed sine wave.

  In the beat waveform analysis step, a waveform error function: e (x) is generated from the result of analyzing the error of the beat waveform with respect to the fixed sine wave on the sweep time axis (S24). The waveform diagram in the third stage on the right side of FIG. 17 shows an example of the waveform error function: e (x). On the time axis, the waveform error continuously occurs from the top to the bottom with respect to the reference value. I can see the situation.

  Next, in order to prevent the waveform error from occurring, a corrected VT curve table for correcting the VT table is generated based on the waveform error function: e (x) (S25). The waveform chart at the bottom right side of FIG. 17 shows a correction VT curve for correcting the VT curve, which is a waveform corresponding to the waveform error function: e (x). A portion for generating a corrected VT curve table for correcting the VT curve exists as a portion that functions in the DSP 11.

  Next, a corrected VT table is generated by applying the corrected VT curve table (S26). Of the two graphs shown in the lower left of FIG. 17, the left graph shows a VT curve, the curve a1 shows a VT curve before correction, and the curve a2 shows a VT curve after correction. . The right graph of the two graphs shows the FT characteristic output from the VCO 13, the curve b1 shows the FT characteristic before correction, and the curve b2 shows the FT characteristic after correction. . The corrected VT curve indicated by a2 indicates an example of the corrected VT table generated in step S26. By applying this post-correction VT table to the VCO 13, the post-correction FT characteristic maintaining the linearity indicated by b2 can be obtained from the VCO 13. The corrected VT curve table and the part that generates the VT curve table exist in the DSP 13 as a function.

The calibration of the VCO 13 is performed as described above, and a signal having an FT characteristic whose frequency linearly changes with respect to the sweep time can be output from the VCO 13 after the calibration. The signal of the linearly varying F-T characteristics toward the liquid surface to be measured emits a signal representative of the beat frequency F _B obtained by receiving a reflected signal from the liquid surface is distorted in waveform In addition, the distance to the liquid level can be accurately measured.

  Since the calibration of the VCO 13 can be performed in real time on a fixed sine wave signal, it can quickly respond to changes in the environmental conditions in which the liquid level measurement device is installed, changes in the characteristics of the VCO over time, etc. The liquid level can always be measured with high accuracy.

  As in the above-described embodiment, the FMCW radar type liquid level measuring device including the means for calculating the relative permittivity of the radio wave propagation space is provided with means for performing VCO calibration, thereby enabling the liquid level measurement. The accuracy can be further increased.

[VCO predistortion]
Next, a VCO predistortion that copes with the second cause of reduced accuracy will be described. The FMCW radar type liquid level measuring device capable of executing VCO predistortion may have the same configuration as the block diagrams shown in FIGS. 1 and 4. However, a program for executing VCO predistortion is stored in the memory 20 of the DSP 11. The VCO predistortion method executed according to the above program will be described below.

The wavelength λg of the radio wave propagating in the waveguide is expressed by the following formula 4.
Formula 4



Here, λo is the wavelength of the propagating frequency, and λc is the cutoff frequency, that is, the wavelength of the lowest frequency that can be transmitted through the waveguide.

In the case of a waveguide having a rectangular cross section as shown in FIG. 23 and having a long side dimension “a” and a short side dimension “b” on the inner surface side, the wavelength λc of the cutoff frequency is It can be expressed by Equation 5.
Formula 5

Here, m and n are the number of modes in the waveguide, respectively. Therefore, in the basic mode, that is, the TE10 mode, since m = 1 and n = 0, λc = 2a.

In the case of a waveguide having a circular cross section and an inner diameter d as shown in FIG. 24, the wavelength λc of the cutoff frequency can be expressed by the following equation (6).
Equation 6
λc = Knm · d
Here, Knm is a unique value for each transmission mode. That is, Knm = 1.706 in the TE11 mode and Knm = 0.82 in the TE01 mode.

The velocity vg of the radio wave propagating in the waveguide can be expressed by the following formula 7.
Equation 7


As can be seen from Equation 7, the velocity vg of the radio wave in the waveguide is a function of the wavelength λo of the propagating frequency, and the velocity vg changes as the frequency changes. The graph of FIG. 25 represents the frequency characteristic of the in-pipe speed, that is, the change in the in-pipe speed (vg) with respect to the change in the frequency (f), and the in-pipe speed changes nonlinearly with respect to the frequency.

Substituting Equation 7 into Equation 1 yields Equation 8.
Equation 8

Beat frequency F _B from Equation 8 can be obtained by the following equation 9.
Equation 9

26, when the reflecting surface is fixed, shows a graph showing a change in the beat frequency F _B with respect to the sweep frequency, the beat frequency F _B is not linearly change with respect to change of the swept frequency, non It is changing linearly.

  FIG. 27 shows a spectrum waveform obtained by subjecting a beat signal to FFT processing. Waveform A shows a spectrum waveform in the case of free space reflection, and waveform B shows a spectrum waveform in the case of reflection in a circular waveguide. Show. The reflection in the circular waveguide has a wider spectrum waveform than in the case of free space reflection. This indicates that the beat frequency is changed by the frequency sweep of the FMCW radar in the circular waveguide even if the measurement point, that is, the reflection point of the radio wave does not change. In the case of frequency sweeping in a circular waveguide, it can be seen that the spectrum waveform obtained by subjecting the beat signal to the FFT calculation process is widened, so that the accuracy of peak detection deteriorates and the measurement accuracy deteriorates.

  As described above, one of the factors that degrade the measurement accuracy is that the beat frequency changes due to the frequency sweep of the FMCW radar. Therefore, in anticipation of a change in the beat frequency due to the frequency sweep, the VT curve characteristic of the control voltage applied to the VCO 13 is reversed with respect to the change in the beat frequency to suppress the change in the beat frequency. When the change in the beat frequency is suppressed, the spread of the spectrum shown by the waveform B in FIG. 27 is suppressed, and the deterioration of measurement accuracy can be suppressed by sharpening the spectrum waveform. As described above, generating a VT curve in which the control voltage applied to the VCO 13 has reverse characteristics is referred to as “VCO predistortion”.

  FIG. 18 shows the procedure of VCO predistortion. The processing steps are shown in the central vertical direction of FIG. Hereinafter, the procedure of VCO predistortion will be described.

  First, the control voltage of the VCO 13 is read from the VT curve table prepared in advance in accordance with the lapse of the sweep time, and this control voltage is applied to the VCO 13 to generate a signal having a frequency corresponding to the control voltage. . The VT curve table should originally maintain the linearity, but the VCO so that the linearity of the FT curve output from the VCO 13 (see the graph on the upper left and right in FIG. 18) is maintained. A calibration is being performed. The VT curve shown on the upper left side of FIG. 18 is a non-linear VT curve in which VCO calibration has been executed in advance according to the characteristics of the VCO 13 or the like.

  Frequency sweeping is performed in accordance with a non-linear VT curve table in which VCO calibration is performed (S31), and radio waves generated by the sweep are radiated toward a measurement point through a circular waveguide. The radio wave reflected at the measurement point is received, and as described above, the beat waveform of the reception frequency and the transmission frequency at a certain time point within the sweep time is acquired (S32). The upper right waveform in FIG. 18 shows an image of the beat waveform.

  Next, beat spectrum analysis is performed by performing FFT calculation processing on the beat waveform (S33). The second waveform diagram from the upper right of FIG. 18 shows an image of spectrum analysis of the beat waveform. As shown in the waveform diagram, the beat frequency difference with respect to the center frequency of the sweep frequency is obtained, and the spectrum analysis of the beat waveform is performed by analyzing the degree of change in the beat frequency difference with respect to the sweep frequency. This spectrum analysis is executed as a part of the function of the DSP 11 by a program incorporated in the DSP 11.

  Next, a spectrum correction function: D (x) is generated from the spectrum analysis result of the beat waveform (S34). The waveform diagram at the third stage on the right side of FIG. 18 shows an example of the spectrum correction function: D (x), and the spectrum correction function: D (x) corresponds to the degree of change in the difference in the beat frequency. Yes. The generation of the spectrum correction function: D (x) is executed as a part of the function of the DSP 11 by a program incorporated in the DSP 11.

  Next, a predistortion table is generated based on the spectrum correction function: D (x) (S35). The predistortion table is a table for correcting the VT curve applied to the VCO 13, and the waveform diagram at the bottom right side of FIG. 18 shows the VT correction curve. The VT correction curve has a reverse characteristic with respect to the waveform representing the spectrum correction function: D (x). A predistortion table generation function is also provided in the DSP 11 as a part of the function.

  Next, a predistortion VT curve table is generated by applying the VT correction curve to the VT curve (S36). A predistortion VT curve table generation function is also provided in the DSP 11 as part of the function. Of the two graphs shown in the lower left of FIG. 18, the left graph shows a VT curve, the curve a1 shows an initial VT curve before correction, and the curve a2 shows a predistortion VT curve. . The initial VT curve a1 is a VT curve after the VCO calibration is performed.

  The right graph of the above two graphs shows the FT curve output from the VCO 13, the curve b1 shows the FT curve before the predistortion, and the curve b2 shows the FT curve after the predistortion. ing. The FT curve b1 before pre-distortion maintains linearity, but when a sweep frequency signal is generated according to the FT curve b1, a change in the propagation speed of radio waves in the waveguide, that is, a change in frequency as described above. Becomes a factor of deterioration of measurement accuracy. Therefore, a sweep frequency signal is generated according to the FT curve b2 after the predistortion. Further, the pre-distortion FT curve b2 is stored as a pre-distortion VT curve table in a memory built in or attached to the DSP 11.

  A predistortion VT curve table is used for the subsequent liquid level measurement. By controlling the oscillation frequency of the VCO 13 according to the predistortion VT curve table, a frequency signal corresponding to a change in the propagation speed of the radio wave in the waveguide is output, so that the beat frequency changes within the sweep time. Less. As a result, the spectrum waveform of the beat frequency signal is sharpened, and the measurement accuracy is increased.

  Thus, if the VCO predistortion is executed in the liquid level measuring device according to the present invention, the oscillation frequency of the VCO is controlled according to the VT curve table after the predistortion. Therefore, the oscillation frequency of the VCO can be controlled in accordance with the frequency characteristics of the radio wave propagating in the waveguide, and the spread of the beat frequency between the oscillation frequency and the reception frequency is suppressed, and the spectrum analysis result of the beat frequency is sharp. The degree of measurement increases and the measurement accuracy can be increased.

  As in the above-described embodiment, the FMCW radar type liquid level measuring device including the means for calculating the relative dielectric constant of the radio wave propagation space is provided with means for executing VCO predistortion, so that the liquid level measurement can be performed. The accuracy can be further increased. If the FMCW radar type liquid level measuring device having means for calculating the relative permittivity of the radio wave propagation space is provided with means for executing VCO calibration and VCO predistortion, the accuracy of liquid level measurement can be further improved. it can.

  The object to be measured by the liquid level measuring apparatus according to the present invention may be anything as long as it is a liquid, but is particularly suitable for an application for measuring a highly accurate liquid level.

11 DSP
13 VCO
16 transducer 16a transducer 16b transducer 17 mixer 20 memory 31 system to be measured

Claims (6)

Through a circular waveguide, a transmitter that transmits radio waves toward the liquid surface while sweeping the frequency for a predetermined sweep time, a receiver that receives the radio waves reflected by the liquid surface, and the radio waves received An FMCW radar-type liquid level measurement device that has an extraction unit that extracts a beat frequency that is a difference between a transmission frequency and a reception frequency at the time of measurement, and measures a distance to the liquid level based on the beat frequency. And
A VCO that generates a sweep frequency signal;
A voltage-time table (hereinafter referred to as “V-T table”) for sweeping the frequency of a signal generated by the VCO;
A reflecting member disposed at a known distance position to correct the linearity of the frequency-time characteristic (hereinafter referred to as “FT characteristic”) of the VCO;
When the frequency sweep signal generated by the VCO is transmitted toward the reflection member while the linearity of the FT characteristic is maintained, the beat generated by the transmission signal and the reflection signal from the reflection member A fixed sine wave signal generation unit for obtaining a fixed sine wave signal having the same waveform;
When the frequency sweep signal generated by the VCO is transmitted toward the reflecting member with the FT characteristic before correction, the fixed sine of the beat waveform generated by the transmission signal and the reflected signal from the reflecting member A correction voltage-time curve table (hereinafter referred to as “correction VT curve table”) generated by analyzing an error with respect to a wave signal;
A corrected VT table obtained by correcting the VT table with the corrected VT curve table,
The corrected VT table is used for the operation of the VCO to generate FT characteristics having linearity,
A spectrum analyzer for analyzing the difference between the beat frequency and the sweep frequency;
A predistortion table generating unit for generating a predistortion table for further correcting the corrected VT table in accordance with the degree of change in the difference of the beat frequency from the analysis result by the spectrum analyzing unit;
A predistortion VT curve table generating unit that generates the predistortion VT curve table by applying the predistortion table to the corrected VT table;
A memory for storing the predistortion VT curve table and serving as a control signal for generating the sweep frequency signal from the predistortion VT curve table to the VCO;
The extraction unit extracts the beat frequency under different transmission conditions,
A liquid level measurement apparatus further comprising a calculating unit that calculates a relative permittivity in the propagation space based on the beat frequency extracted for each of the different transmission conditions.   The liquid level measurement device according to claim 1, wherein the transmission condition is two different sweep frequencies of the transmission radio wave.   The liquid level measurement device according to claim 1, wherein the transmission conditions are two different transmission modes of transmission radio waves.   The liquid level measuring device according to claim 3, wherein the two different transmission modes are a TE01 mode and a TE11 mode.   The liquid level measuring device according to claim 1, further comprising an antenna or a conversion device that guides a sweep frequency signal generated by the VCO toward the liquid surface. The antenna or conversion device
A disc-shaped radial waveguide;
A plurality of slots radially formed in the radial direction of the radial waveguide at equal intervals around the entire radial direction of the radial waveguide;
A plurality of excitation pins provided on one side of each slot;
A radial loop material provided at the center of the radial waveguide inside each slot forming region, and
5. The liquid level measurement according to claim 4, wherein the plurality of slots and the plurality of excitation pins constitute a TE01 mode transducer, and the radiating loop material is a two-transmission mode compatible planar transducer constituting a TE11 mode transducer. apparatus.