Fill level measuring device

Abstract

The invention relates to a method for a radar-based determination of a fill level (L) in a container (2). In the method, an evaluation curve (C) for a defined measuring range (h) is created in periodic measuring cycles on the basis of the received radar signal (RHF). According to the invention, a selection region (ROI) is defined at the evaluation curve (C) end region which extends upwards from the container base. Within the evaluation curve (C), a respective current fill level signal maximum (mc) is determined for each measuring cycle or is tracked in the subsequent measuring cycles. Normally, the fill level value (L) corresponding to the fill level signal maximum (mc) is output for each measuring cycle. According to the invention, however, the fill level (L) is defined as empty if the fill level signal maximum (mc) has been lost within the selection region (ROI) in relation to one of the preceding measuring cycles. The method is therefore based on the observation that, in particular at high radar frequencies, the emitted radar signal (SHF) is not necessarily reliably reflected in the selection region (ROI) even though the level of the filling material surface actually corresponds to an empty fill level (L).

Description

[0001] Level gauge

[0002] The invention relates to a radar-based level measuring device.

[0003] In process automation technology, suitable field devices are used to acquire relevant process parameters. These field devices incorporate appropriate measurement principles to acquire process parameters such as fill level, flow rate, pressure, temperature, pH value, redox potential, or conductivity. The Endress+Hauser Group manufactures and distributes a wide variety of field device types and variants.

[0004] Non-contact measurement methods have become established for measuring the fill level of contents in containers due to their robustness and low maintenance requirements. A further advantage of non-contact measurement methods is their ability to measure the fill level almost continuously. Therefore, radar-based measurement methods are predominantly used in the field of continuous level measurement (in the context of this patent application, the term "radar" refers to signals or electromagnetic waves with frequencies between 0.03 GHz and 300 GHz). The pulse transit-time method and FMCW ("Frequency Modulated Continuous Wave") have become established as measurement methods. Radar-based level measurement is described in more detail, for example, in "Radar Level Detection," Peter Devine, 2000.

[0005] Typical frequency bands approved for radar-based level measurement are 26 GHz, 60 GHz, 80 GHz, and 120 GHz, and increasingly also 180 GHz. Frequency bands at higher frequencies are advantageous for many applications because, given antenna dimensions, a more focused beam is achieved and more bandwidth is generally available, which can be used for greater distance resolution.

[0006] In principle, radar-based methods determine the fill level by sending a radar signal towards the contents in each measurement cycle, receiving a corresponding received signal after reflection of the radar signal inside the container, creating an evaluation curve based on the received signal, and determining the fill level based on the evaluation curve.

[0007] Within subsequent measurement cycles, these process steps are repeated cyclically. The evaluation curve represents the signal strength of the reflected received signal as a function of the distance to the level sensor in the container. The fill level is determined from the evaluation curve by identifying the signal maximum that corresponds to the reflection of the radar signal at the surface of the contents.

[0008] In addition to this signal maximum, further signal maxima appear in the evaluation curve in practice, which are due to multiple echoes or reflections from peripheral objects such as the container wall, the container bottom, and internal components. Therefore, a specific "7argef / ng" algorithm is applied to the evaluation curve in the level gauge to locate, among all signal maxima, the one that results from reflection at the surface of the contents. As a constraint within a targeting algorithm, it can be implemented, among other things, that the level signal maximum can only be located within a defined measuring range between the container bottom and the level gauge.

[0009] One possible embodiment of a targeting algorithm is described, for example, in publication DE 102018124606 A1. After the correct signal maximum has been found within an initial measurement cycle, it is tracked within the evaluation curve in subsequent measurement cycles. This means that, based on logically comprehensible parameters, the evaluation unit of the level gauge uses an algorithm to check whether the signal maximum identified in the current measurement cycle is plausible in light of the signal maximum determined in the previous measurement cycle. Logically comprehensible parameters could include, for example, that the amplitude of the signal maximum and / or its position within the evaluation curve may only change by a certain maximum amount with each successive measurement cycle.

[0010] In general, the application of targeting and echo tracking algorithms requires that at least one signal maximum is identified within the evaluation curve. Otherwise, in the case of "echo loss," the level gauge becomes inoperable. This scenario is particularly likely near the bottom of a container if the level gauge transmits or receives the radar signal at a frequency above 70 GHz, as the radar beam becomes more focused with increasing frequency. This risk is further amplified if the container bottom is not flat but tapers downwards, for example, conically. These factors increase the risk that the radar signal will not be reflected back to the level gauge, for instance, due to a wavy or uneven surface, resulting in no signal being received and, consequently, at least a temporary echo loss.

[0011] The invention is therefore based on the objective of providing a radar-based level measuring device that can handle this scenario.

[0012] The invention solves this problem by means of a corresponding method for radar-based determination of the fill level of a substance in a container. For this purpose, the method according to the invention comprises the following process steps, which are repeated periodically as a measurement cycle:

[0013] Emitting a radar signal towards the contents, receiving a corresponding signal after reflection of the radar signal inside the container,

[0014] Creation of an evaluation curve based on at least the received signal for a defined measuring range, o wherein a selection range is defined at the end of the evaluation curve which encompasses the bottom of the container,

[0015] Determination of a level signal maximum within the evaluation curve or its tracking in subsequent measurement cycles, and output of the level value corresponding to the level signal maximum. According to the invention, the level is defined as empty if the level signal maximum has been lost in relation to one of the previous measurement cycles within the selection range.

[0016] This handles the scenario of echo loss in the selection area near the bottom of the container by equating this scenario to an empty container or a fill level of zero, at least for a certain number of subsequent measurement cycles. An advantage of this method according to the invention is that it allows the level measuring device to resume echo tracking from the selection area as soon as the fill level signal maximum reappears there in one of the subsequent measurement cycles. For this purpose, the method according to the invention is designed such that the fill level signal maximum in the current measurement cycle is determined exclusively within the selection area, provided that the fill level is defined as empty, at least from the previous measurement cycle.However, if the level signal maximum does not reappear within a defined initial number of measurement cycles after its loss, the level measuring device can output that no level value can be determined, according to the inventive method. This information, that no level can be determined, can optionally also be output by the level measuring device within the inventive method if the level signal maximum is lost in the evaluation curve outside the selection range in relation to one of the previous measurement cycles. If the level signal maximum cannot be found again within the selection range after the initial number of measurement cycles, the inventive method can be designed such that the level signal maximum is then searched for again within the entire evaluation curve, i.e., within the entire measurement range.

[0017] To ensure that, after an echo loss, the level signal maximum is actually detected in subsequent measurement cycles within the measuring range, the method according to the invention can be extended such that the level measuring device outputs that no level can be determined, even though a level signal maximum can be determined within the evaluation curve after a measurement cycle in the current measurement cycle, and provided that the level signal maximum was lost less than a defined second number of measurement cycles ago.Similarly, a procedure can be followed after an echo loss within the selection range: In this case, the level gauge can output an empty level, even though a level signal maximum can be determined within the evaluation curve after a measurement cycle in the current cycle, provided that the level signal maximum was lost within the selection range less than a defined second number of measurement cycles ago. Thus, the output of the determined level value after an echo loss is effectively delayed by this second number of measurement cycles.

[0018] Depending on the application of the level gauge, it is also advantageous to implement the inventive method in the level gauge in such a way that the fill level is defined as empty even if no fill level signal maximum can be determined within the evaluation curve from the time the level gauge is put into operation, i.e., from the first measurement cycle. This is advantageous because the container is typically empty when the level gauge is put into operation, and therefore a valid fill level signal maximum cannot be determined by the level gauge.

[0019] A level measuring device suitable for carrying out the procedure according to one of the previous embodiments shall comprise the following components:

[0020] A high-frequency unit designed to: generate radar signals in periodic measurement cycles and transmit them to the contents; receive corresponding reception signals after reflection of the radar signals inside the container; an evaluation unit designed to: periodically generate an evaluation curve based on the received signal for each measurement cycle; output a level value if a corresponding level signal maximum can be determined within the evaluation curve; and define the level as empty if the level signal maximum has been lost in relation to one of the previous measurement cycles within the selection range.

[0021] Within the scope of the invention, the term "unit" is generally understood to mean any configuration or encapsulation of the electronic circuits required for the specific application, e.g., for high-frequency signal processing or as an interface. Depending on the application, the unit may therefore comprise analog circuits for generating or processing analog signals. However, the unit may also include digital circuits, such as FPGAs, microcontrollers, or storage media, in conjunction with appropriate programs. The program is designed to execute the necessary process steps or perform the required arithmetic operations. In this context, various electronic circuits within the unit, as defined by the invention, can potentially access a common physical memory or be operated using the same physical digital circuit.It is irrelevant whether different electronic circuits within the unit are distributed on a common circuit board or on several interconnected circuit boards.

[0022] The invention is explained in more detail with reference to the following figures. They show:

[0023] Fig. 1: A radar-based level gauge on a container, and

[0024] Fig. 2. A schematic diagram of the method according to the invention.

[0025] For a basic understanding of the invention, Fig. 1 shows a container 3 with a substance 2, the fill level L of which is to be determined. Depending on the type of substance 2 and the application, the container 3 can be more than 100 m high. The conditions inside the container also depend on the type of substance 2 and the application. For example, exothermic reactions can lead to high temperature and pressure stresses. For dusty or flammable substances, appropriate explosion protection measures must be observed inside the container.

[0026] To determine the fill level L independently of the prevailing conditions, a radar-based level gauge 1 is mounted above the contents 2 at a known installation height h above the bottom of the container s. The level gauge 1 is attached to a corresponding opening in the container 3 in such a pressure- and media-tight manner that only an antenna array 11 of the level gauge 1 is directed vertically downwards into the container s towards the contents 2, while the other components of the level gauge 1 are arranged outside the container 3 in a separate housing.

[0027] The level sensor 1 is usually connected via a separate interface unit, which uses a protocol such as "4-20 mA", "PROFIBUS", or "HART". 1 , or “Ethernet 1The system is implemented and connected to a higher-level unit 4, such as a local process control system or a decentralized server system. The measured fill level value L can be transmitted via this connection, for example, to control the inflow or outflow of the container 3. Other information about the general operating status of the level gauge 1 can also be communicated.

[0028] Radar signals SHF are transmitted via the antenna array 11 of the level gauge 1 towards the surface of the material 2 within a predefined frequency band. After reflection from the surface of the material, the level gauge 1 receives the reflected reception signals RHF again via the antenna array 11. The signal propagation time t between transmission and reception of the respective radar signal SHF, RHF is given by proportional to the distance d between the level gauge 1 and the contents 2, where the variable "c" represents the radar propagation speed, which is at least roughly known. The signal travel time t can be determined by the level gauge 1, for example, using the FMCW or pulse travel time method. Based on appropriate calibration, the level gauge 1 can then assign the measured travel time t to the respective distance d. Furthermore, the level gauge 1 can determine the fill level L according to d = h - L, provided the installation height "h" is stored in the level gauge 1. The installation height h of the level gauge 1 above the bottom of the container also corresponds to the measuring range relevant for the fill level measurement.

[0029] To determine the signal propagation time t or the corresponding level value L based on the incoming received signal RHF, the level gauge 1 includes a suitably designed evaluation unit in which the corresponding measurement principle (FMCW or the pulse propagation time method) is implemented. A corresponding radio frequency unit in the level gauge 1 is used to generate the transmitted radar signal SHF and to preprocess the received signal RHF. If the FMCW method is implemented, the transmitting radio frequency unit can, for example, be based on a PLL (phase-locked loop).

[0030] On the receiving side, the high-frequency unit can include a mixer to combine the received signal RHF with the instantaneously generated radar signal SHF ZU, thereby obtaining the intermediate frequency signal characteristic of FMCW. The frequency or frequency band in which the high-frequency unit 10 of the level gauge 1 generates the radar signal SHF is selected primarily depending on the properties of the material being measured: In the case of high-precision level measurement, a comparatively high frequency band, for example 180 GHz, is implemented due to the inherent requirements.

[0031] Based on the received signal RHF or the intermediate frequency signal IF, an evaluation curve C is created in the evaluation unit, which reflects the signal strength of the received signal RHF as a function of the signal propagation time t or the distance d.

[0032] For the fill level L shown in Fig. 1, a corresponding evaluation curve C is shown schematically to the right of the container s.

[0033] This shows that in the evaluation curve C, at the level of the fill material surface, a fill level signal maximum m is observed due to the reflection of the radar signal SHF. c appears. This fill level signal maximum m cThe evaluation unit of the level gauge 1 can determine the corresponding distance value d using a targeting algorithm. In practice, the implementation of a targeting algorithm is necessary to avoid erroneously determining the level value L based on peripheral signal maxima. As shown in Fig. 1, a peripheral signal maximum can occur, for example, due to multiple reflections within the container, virtually outside the measuring range h or below the container bottom in the evaluation curve C. The level gauge 1 periodically repeats the generation of an updated evaluation curve C and the subsequent recalculation of the level value L in defined measurement cycles of, for example, 10 ms. The level signal maximum m identified by the targeting algorithm is used in this process. sThe measurement cycles are tracked using echo tracking. Since the container s shown in Fig. 1 does not have a flat bottom, but rather a conically tapered one, a decreasing fill level L can lead to an echo loss, i.e., a loss of the fill level signal maximum m, with increasing measurement cycles. c lead.

[0034] As illustrated in Fig. 1, the portion of the evaluation curve C that extends across the narrowing is defined as a selection range ROI and accordingly stored in the evaluation unit of the level gauge 1. The definition of the selection range ROI at this end of the evaluation curve C, which connects to the lower end of the container 3, forms the basis of a method according to the invention. This method, which can be implemented in the level gauge 1, is explained in more detail below with reference to Fig. 2.

[0035] The core of the procedure is the determination, based on the evaluation curve C of the current measurement cycle, of whether the fill level signal maximum m c has been lost compared to one of the previous measurement cycles. This can be determined, for example, by applying the targeting algorithm to the evaluation curve C, either successfully or unsuccessfully. In the successful case, i.e., if a plausible level signal maximum m is found within the selection range ROI or within the measurement range h, c If the level is identifiable or traceable, the corresponding fill level value L is output as the last step of the procedure before the next measurement cycle.

[0036] However, if this step of the procedure detects an echo loss, the next step is to determine whether the echo loss occurred within or outside the selection range (ROI). This can be clarified by examining the evaluation curve C of the previous measurement cycle, in which the level signal maximum m cThe process analyzes whether the echo loss has not yet been lost: If the level sensor was outside the selection range (ROI) in the measurement cycle before the echo loss, the level sensor 1 then outputs that no level value L can be determined. Within the framework of the method, this output has no further consequences until a new evaluation curve C is recorded in the next measurement cycle. However, if the echo loss occurred within the selection range (ROI), the method outputs, at least for a defined initial number x of measurement cycles, that the container 3 is empty or the level value L = 0. According to the invention, the method is therefore based on the assumption that despite the loss of the level signal maximum m, the level sensor 1 can still be determined. c Within the selection range ROI, the fill level L must likely lie within this selection range ROI.

[0037] This assumption implies the conclusion that any signal maxima appearing outside the selection range (ROI) in the evaluation curve C of the measurement cycle following the echo loss are not the level signal maximum m. c can act. Accordingly, within the framework of the inventive method, it is saved for each subsequent measurement cycle whether the container 3 was defined as empty or the fill level = 0 in the current measurement cycle.

[0038] As shown in Fig. 2, the corresponding saved result from the previous measurement cycle is retrieved after the new evaluation curve C is recorded. Depending on the query, the procedure determines whether the fill level signal maximum m c The search is conducted only within the selection range ROI, or within the entire measurement range or the entire evaluation curve C:

[0039] If container 3 was defined as empty or the fill level = 0 in the previous measurement cycle, the fill level signal maximum m c Using the targeting algorithm, the search is conducted only within the selection range (ROI) of the current evaluation curve C. In the other two cases (either a fill level value L could be determined normally in the previous measurement cycle; or no fill level L could be determined because an echo loss occurred outside the selection range ROI), the fill level signal maximum m is used. c The targeting algorithm was used to search within the entire current evaluation curve C.

[0040] This case distinction ("Was L defined as empty in the previous measurement cycle?") within the procedure thus implements the conclusion that any signal maxima appearing outside the selection range ROI in the evaluation curve C of the measurement cycle following the echo loss are not the level signal maximum m.c can act.

[0041] The possible case that the fill level signal maximum m c Even after more than the defined initial number x of measurement cycles, the level signal can no longer be found within the selection range (ROI) after its loss, which can lead to an endless loop within the process. As shown in Fig. 2, the inventive method can be extended accordingly to avoid such an endless loop, so that after more than x measurement cycles the level signal maximum m c The search is then conducted again within the entire measuring range. For this purpose, in the measurement cycles, after determining that the fill level signal maximum m c The measurement cycle was lost and was within the ROI selection range before its loss, and the system consecutively queried whether the fill level signal maximum m chas been lost or cannot be located for more than x measurement cycles. As soon as the first number x of measurement cycles in which the level signal maximum m c If the fill level L is still not detectable within the selection range (ROI), or if the ROI is exceeded, the procedure will then output that the fill level L cannot be determined. As shown in Fig. 2, this results in the fill level signal maximum m being displayed in the subsequent measurement cycle. c The search is again conducted within the entire evaluation curve C, since the query "Was L defined as empty in the previous measurement cycle?" must be answered with "no". This creates a potential endless loop of measurement cycles in which the level signal maximum m cThe search is interrupted exclusively within the ROI selection area. The number x in the inventive method must be selected such that it is ensured that the surface of the contents is no longer within the ROI selection area as soon as the fill level signal maximum m is reached. c The entire evaluation curve C is searched for again. In practice, it is preferable to choose x > 10 measurement cycles.

[0042] The case-dependent result of the method according to the invention, i.e., either the determined fill level value L, the information that no fill level value L can be determined, or that the container 3 is empty or the fill level L = 0, can be output in various ways within the scope of the invention: For example, a corresponding output to the higher-level unit 4 is conceivable. However, it is also conceivable that a corresponding output is made directly on a display of the level measuring device 1.

[0043] Reference symbol list

[0044] 1 level gauge

[0045] 2 Filling material 3 Containers

[0046] 4. Higher-level unit

[0047] 11 Antenna arrangement

[0048] C Evaluation curve d Distance h Installation height / measuring range

[0049] L level m c Level signal maximum RHF receive signals

[0050] SHF radar signals ROI selection range x number of measurement cycles

Claims

Patent claims 1. Method for radar-based determination of a fill level (L) of a fill material (2) in a container (3), comprising the following method steps which are repeated in periodic measurement cycles: Emitting a radar signal (SHF) towards the fill material (2), Reception of a corresponding receive signal (RHF) after reflection of the radar signal (SHF) inside the container, Creation of an evaluation curve (C) based on at least the received signal (RHF) for a defined measuring range (h), o wherein a selection range (ROI) is defined at the end region of the evaluation curve (C) that encompasses the bottom of the container, determination of a level signal maximum (m c ) within the evaluation curve (C) or its tracking in the subsequent measurement cycles, and Output of the signal maximum at the fill level (m c) correspond to the fill level value (L), where the fill level (L) is defined as empty if the fill level signal maximum (m) c ) in relation to one of the previous measurement cycles within the selection range (ROI).

2. Method according to claim 1, wherein the level signal maximum (m c ) is determined exclusively in the selection range (ROI) in the current measurement cycle, provided that the fill level (L) is defined as empty at least from the previous measurement cycle.

3. Method according to claim 1 or 2, wherein it is output that no level value (L) can be determined, provided that the level signal maximum (m) c ) in relation to one of the previous measurement cycles outside the range of interest (ROI), and / or if the level signal maximum (m c ) was lost more than a defined initial number (x) of previous measurement cycles within the selection range (ROI).

4. Method according to at least one of the preceding claims, wherein no level value (L) is output or the level (L) is defined as empty if the level signal maximum (m) occurs less than a defined second number (y) of measurement cycles ago. c ) has been lost, although a level signal maximum (m) is present within the evaluation curve (C) in the current measurement cycle. c ) can be determined.

5. Method according to at least one of the preceding claims, wherein the fill level (L) is defined as empty if no fill level signal maximum (m) occurs in a first measurement cycle. c ) can be determined within the evaluation curve (C).

6. Level measuring device for carrying out the method according to one of the preceding claims, comprising: A high-frequency unit designed to: generate a radar signal (SHF) and transmit it to the contents (2), and to receive a corresponding received signal (RHF) after reflection of the radar signal (SHF) inside the container; an evaluation unit designed to: generate an evaluation curve (C) based on the received signal (RHF), and output a level value (L) if a corresponding level signal maximum (m) is present within the evaluation curve (C). c ) can be determined, and o to define the fill level (L) as empty, provided that the fill level signal maximum (m c ) in relation to one of the previous measurement cycles within the selection range (ROI).

7. Level measuring device according to claim 6, wherein the high-frequency unit is designed to generate or receive the radar signal (S, RHF) with a frequency of at least 70 GHz, in particular 100 GHz.

Images