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

Patent metadata

Field	Value
Publication number	US-11656115-B2
Application number	US-201916654694-A
Country	US
Kind code	B2
Filing date	Oct 16, 2019
Priority date	Oct 18, 2018
Publication date	May 23, 2023
Grant date	May 23, 2023

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.

First claim

The invention claimed is:

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.

Assignees

• Endress Hauser Conducta Gmbh Co Kg

Inventors

• Lopatin Sergey
• Waldhauser Dieter
• Alber Thomas
• Leufke Philipp
• Kilian Markus
• Brengartner Tobias
• Page Rebecca

Classifications

• G01D21/00

  Measuring or testing not otherwise provided for · CPC title

• G06F18/241 Primary

  relating to the classification model, e.g. parametric or non-parametric approaches · CPC title

• G01N21/55

  Specular reflectivity · CPC title

• G01F23/2962

  Measuring transit time of reflected waves · CPC title

• G06V10/751

  Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching · CPC title