Method for determining a process variable with a classifier for selecting a measuring method

Abstract

The present disclosure relates to a method for determining at least one process variable of a medium. The method includes steps of recording a first value for the process variable by means of a first method for determining the process variable and recording a second value for the process variable by means of a second method for determining the process variable. The method also includes steps of selecting at least one of the detected values for the process variable by means of a classifier and outputting the selected value for the process variable. The present disclosure further relates to a computer program designed for executing a method according to the present disclosure, and to a computer program product having a computer program according to the present disclosure.

Claims

1. A method for determining at least one process variable of a medium, including the following method steps: recording a first value for the process variable by means of a first method for determining the process variable; recording a second value for the process variable by means of a second method for determining the process variable; selecting at least one of the recorded values for the process variable using a classifier, wherein the classifier selects the at least one of the recorded values for the process variable based on process parameters, medium parameters and/or environmental parameters; outputting the selected value for the process variable; determining a feature vector; designing the classifier to select the selected value for the process variable based on the feature vector; and further including using a first and a second classifier, wherein the first classifier performs a feature extraction and/or creates the feature vector, and wherein the second classifier selects the selected value for the process variable based on the feature vector.

2. The method of claim 1, wherein the classifier is designed to learn the selection of at least one of the recorded values for the process variable.

3. The method of claim 2, wherein the classifier is trained online or offline.

4. The method of claim 1, wherein the classifier selects the selected value for the process variable based in part on at least one influencing variable.

5. The method of claim 1, further including creating a map used by the classifier to determine the selected value, wherein the map is created based on a data set, wherein the data set includes at least one input variable and an output variable associated with the input variable.

6. The method of claim 1, further including determining at least one value for the process variable using a soft sensor.

7. The method of claim 1, further including determining at least one value for the process variable using a device for determining or monitoring the process variable.

8. The method of claim 1, further including determining a classification quality in relation to the selection of the selected value for the process variable based on a probability in relation to the selection of the selected value for the process variable.

9. The method of claim 1, further including comparing at least a first and a second value for the process variable or a first and a second value derived from the first and second values for the process variable with one another.

10. The method of claim 1, further including performing a status monitoring of at least one soft sensor or one device for determining or monitoring the process variable.

11. The method of claim 1, wherein the process variable is a fill level of the medium in a container, a flow rate of the medium through a pipe, a conductivity of the medium, or a concentration of a substance contained in the medium.

12. A method for determining at least one process variable of a medium, including the following method steps: recording a first value for the process variable by means of a first method for determining the process variable; recording a second value for the process variable by means of a second method for determining the process variable; selecting at least one of the recorded values for the process variable using a classifier; outputting the selected value for the process variable; determining a feature vector; designing the classifier to select the selected value for the process variable based on the feature vector; and further including using a first and a second classifier, wherein the first classifier performs a feature extraction and/or creates the feature vector, and wherein the second classifier selects the selected value for the process variable based on the feature vector.

13. A non-transitory computer-readable medium storing a program including instructions that, when executed by a processor, cause a computer to determine at least one process variable of a field device including steps of: recording a first value for the process variable by means of a first method for determining the process variable; recording a second value for the process variable by means of a second method for determining the process variable; selecting at least one of the recorded values for the process variable using a classifier wherein the classifier selects the at least one of the recorded values for the process variable based on process parameters, medium parameters and/or environmental parameters; outputting the selected value for the process variable; determining a feature vector; designing the classifier to select the selected value for the process variable based on the feature vector; and further including using a first and a second classifier, wherein the first classifier performs a feature extraction and/or creates the feature vector, and wherein the second classifier selects the selected value for the process variable based on the feature vector.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The present disclosure is explained in greater detail with reference to the following Figures.

(2) FIG. 1a shows a block diagram for illustrating the method according to the present disclosure;

(3) FIG. 1b shows a block diagram for illustrating the method according to the present disclosure;

(4) FIG. 2a illustrates a possible application in connection with the determination of a fill level of a medium;

(5) FIG. 2b illustrates a possible application in connection with the determination of a fill level of a medium;

(6) FIG. 2c illustrates a possible application in connection with the determination of a fill level of a medium;

(7) FIG. 3a illustrates a possible application in connection with determining a conductivity of a medium;

(8) FIG. 3b illustrates a possible application in connection with determining a conductivity of a medium; and

(9) FIG. 4 shows a possible application in connection with determining the pH value of a medium.

DETAILED DESCRIPTION

(10) In Figures, identical elements are respectively given the same reference characters.

(11) The method according to the present disclosure is schematically illustrated in FIG. 1. The method can, for example, on the one hand be implemented in the electronic unit of a field device M.sub.1-M.sub.n (with M.sub.1 being shown), or also in a higher-level unit. According to the present disclosure, values y.sub.1-y.sub.i (with y.sub.1 and y.sub.2 being shown) for a process variable P are respectively determined by means of the methods based on the sensor signals x.sub.1-x.sub.i (with x.sub.1 and x.sub.2 being shown). x.sub.1 and V.sub.1, or x.sub.2 and V.sub.2, are respectively part of a device M.sub.1 or M.sub.2 for determining the process variable P by means of the method or measuring method V.sub.1 and V.sub.2. Furthermore, in addition to the devices M.sub.1-Mn, a soft sensor S.sub.1 is optionally shown here by way of example, which soft sensor S.sub.1 also determines a value w.sub.1 for the process variable based on the input variable v.sub.1, by means of the measuring method V.sub.n.

(12) Depending on the embodiment, all measuring methods V.sub.1-V.sub.n can be computational methods. All measuring methods V.sub.1-V.sub.n can also be metrological methods. As is shown using the example from FIG. 1, said methods can also be partially computational methods V.sub.n and partially metrological methods M.sub.1, M.sub.2.

(13) According to the present disclosure, the classifier K serves for determining and/or selecting a selected value (here y.sub.2) for the process variable P from the set of the determined values y.sub.1-y.sub.i, w.sub.1 for the process variable. For this purpose, the determined values y.sub.1-y.sub.i, w.sub.1 are supplied to the classifier K. The selection of the classifier is illustrated in FIG. 1 by the switching element S, which is controlled by the classifier K. The selected value y.sub.2 is then output as a value for the process variable P.

(14) Optionally, one or more influencing variables can be made available to the classifier K, as indicated by the dashed arrows. In the present instance these are, for example, the sensor signals x.sub.1-x.sub.i, an input variable v.sub.1 of the soft sensor S.sub.1, as well as the other influencing variables x.sub.j and x.sub.k. These can be provided, for example, by process parameters, media parameters, and/or environmental parameters.

(15) The different methods V.sub.1-V.sub.n are typically differently well suited to different applications, measurement ranges, or the like. For example, the various methods may cover different combinational configurations in the process, or different application fields, or different media. Individual methods may also not be applicable to certain applications, so that a selection of the corresponding value for the process variable P can be excluded in these instances.

(16) In the simplest instance, the respective selected value or the selection of the selected value remains the same for a predefinable duration of a particular process. However, it is also conceivable that process and/or environmental conditions change during continuous operation in such a way that a change in the selection by the classifier K is to be performed continuously, periodically, or selectively. For example, in the instance of FIG. 1b, a change in selection from the value y.sub.2 to the selected value w.sub.1 by the classifier K has taken place, which is then output as the value w.sub.1 for the process variable P.

(17) A possible application of the method according to the present disclosure with regard to the determination of a fill level F of a medium M as a process variable P is illustrated in FIG. 2. The transit time-based fill level measuring method V.sub.1 known per se from the prior art is used as the first measuring method V.sub.1, as is illustrated in FIG. 2a. Corresponding field devices M.sub.1 are manufactured by the applicant in many different embodiments and are sold, for example, under the names Micropilot, Levelflex, or Prosonic.

(18) A transmission signal S is reflected at a surface O of a medium M located in a container 1, and the received echo signal R is then evaluated with respect to the fill level F of the medium M. Different field devices M.sub.1, M.sub.2 operating according to the transit time-based fill level measurement method V.sub.1 typically have different measurement ranges with regard to the determinable fill levels within the container 1. The measuring range here is understood to be a specific height range Δh within the container 1 within which the fill level can be determined on the basis of the two field devices M.sub.1 and M.sub.2. The two measuring devices M.sub.1 and M.sub.2 cover different measurement ranges Δh.sub.1 and Δh.sub.2 within the container.

(19) However, neither of the measuring devices M.sub.1 or M.sub.2 is suitable in order to be able to determine the fill level in the event of an almost completely filled container. A further measuring device M.sub.3 can be used here, for example a vibronic sensor M.sub.3 as shown in FIG. 2b. Such a field device M.sub.3 operates on the vibronic measuring principle V.sub.3 (not specifically shown, but represented as V.sub.n), which is likewise known from the prior art. Corresponding field devices are also manufactured by the applicant in many different embodiments and are sold under the names LIQUIPHANT or SOLIPHANT.

(20) The vibronic sensor M.sub.3 comprises a sensor unit 2 with a mechanically oscillatory unit 3 in the form of a vibration fork. The mechanically oscillatory unit 3 is excited to mechanical oscillation by means of the drive/receiving unit 4. Furthermore, the drive/receiving unit 4 is used to receive mechanical oscillations of the vibration fork 3 by means of which the fill level of the medium can be determined and/or monitored. Furthermore, an electronic unit 6 that follows on a neck tube 5 is shown, by means of which signal evaluation and/or signal feed takes place.

(21) The three measuring devices M.sub.1-M.sub.3 cover different measuring ranges Δh.sub.1-Δh.sub.3. Together, they enable the fill level to be determined over the entire height h of the container 1, as illustrated in FIG. 2c. As can be seen there, for the shown example the different measuring ranges Δh.sub.1-Δh.sub.3 partially overlap. The classifier now respectively serves to choose a selected value for the process variable on the basis of the values y.sub.1-y.sub.3 (with y.sub.3 not explicitly shown) determined by the three measuring devices M.sub.1-M.sub.3 for the process variable P. If the fill level is low, for example, the measuring device M.sub.2 should be selected, and thus the selected value y.sub.2 for the process variable P.

(22) A further exemplary application of the method according to the present disclosure relates to the determination of the process variable of the conductivity σ of a medium M. In this respect, for example, the conductive and inductive measuring methods have become known from the prior art. Corresponding field devices are produced and marketed by the applicant under the name Condumax or Indumax, for example.

(23) A measuring device M.sub.4 operating on the conductive measuring principle V.sub.4 (not specifically shown, but represented as V.sub.n) is shown in FIG. 3a. The measuring device M.sub.4 comprises a sensor element 7 protruding into the container 1, having an inner electrode 8 and an outer electrode 9 which are electrically isolated from one another. In order to determine the conductivity of the medium, such as the electrolytic conductivity, the resistance or conductance of the measured section in the medium M is determined.

(24) A measuring device M.sub.5 operating on the inductive measuring principle V.sub.5 (not specifically shown, but represented as V.sub.n) is in turn drawn in FIG. 3b. The sensor M.sub.5 is introduced via a process connection 10 into the container 1 in which the medium M is located. The sensor M.sub.5 comprises a transmitter coil 11 and a receiver coil 12 that are arranged in a housing 13. In an end region facing toward the medium M, the housing 13 forms a channel 14 into which the medium M can penetrate. The coils 11, 12 and the channel are thereby arranged in such a way that a closed current path 15 passing through both coils can form in the medium M. If the transmitter coil is charged with an alternating voltage, a magnetic field is generated which induces the current path 15 and whose strength depends on the conductivity σ of the medium M.

(25) The typical areas of application for a conductive conductivity sensor M.sub.4 and an inductive conductivity sensor M.sub.5 typically differ depending on the conductivity σ of the medium M. The conductive measuring principle V.sub.4 is suitable for media with a lower conductivity σ, such as in the range of 0.04 μS/cm-200 mS/cm, whereas the inductive measuring principle V.sub.5 is to be preferred for media with a greater conductivity, such as in the range of 2 μS/cm-2000 mS/cm.

(26) Here, the method according to the present disclosure is suitable. The classifier K can respectively suitably select a measured value y.sub.4 or y.sub.5 (with y.sub.4 and y.sub.5 not explicitly shown) for the conductivity σ of the medium M, so that a measured value that is as accurate as possible can be determined as a selected value, independently of the medium in question.

(27) It should be noted that the considerations mentioned in connection with FIG. 3 can also be transferred, mutatis mutandis, to the determination of a fill level of a medium according to the capacitive and conductive measurement principles.

(28) A further possible application of the method according to the present disclosure in connection with a soft sensor relates to the measurement of the pH value of a medium. A widespread measuring method relates to the pH measurement method which is very well known per se from the prior art, by means of a pH sensor designed as a potentiometric combined electrode. Corresponding measuring devices are likewise produced by the applicant in many embodiments and are sold under the names Memosens or Orbisint. A combined electrode of a pH sensor M.sub.1 is schematically illustrated in FIG. 4. The combined electrode M.sub.1 is of a rod-shaped design and comprises a measuring half-cell 16 with an outer chamber 16a and a reference half-cell 17 with an inner chamber 17a, which reference half-cell 17 is arranged coaxial to the measuring half-cell 16 and surrounds it in segments. The reference half-cell 17 comprises a reference electrode 18 and a reference electrolyte 19, whereas the measuring half-cell 16 comprises a measuring electrode 20 and an inner electrolyte 21. Furthermore, the measuring half-cell 16 has a media-contacting glass membrane 22 which is terminally arranged at the combined electrode 1. However, problems can arise given the use of glass electrodes, such as in the event of high-purity water such as, for example, in steam circuits in the power plant sector. As time passes, it may happen that, among other things, ions of the electrolyte pass into the medium, which has a negative effect on the measurement of the pH value.

(29) Another measurement method for determining the pH value is what is known as differential conductivity measurement, also known from the prior art. The conductivity of the medium before and after passing through a cation exchanger is hereby determined. The cation exchanger typically contains a resin regenerated with sulfuric acid. By means of a suitable formula, the pH value can be determined by calculation from the two values for the conductivity. This in turn can be realized by means of a suitable soft sensor, for example. The formula to be used will depend on, among other things, the respective particular chemical reaction taking place in the process or in the cation exchanger.

(30) Whether a determination of the pH value by means of a pH sensor or by means of a differential conductivity measurement is more suitable now depends on various factors, for example on the purity of the medium. The method according to the present disclosure could be used in this context. The classifier K can respectively suitably select a measured value for the pH, so that the highest possible measurement accuracy can be achieved independently of the medium M.