Method and apparatus for determining a state of a measuring transducer integrated in a process container

Patent metadata

Field	Value
Publication number	US-9823100-B2
Application number	US-201314652479-A
Country	US
Kind code	B2
Filing date	Nov 25, 2013
Priority date	Dec 20, 2012
Publication date	Nov 21, 2017
Grant date	Nov 21, 2017

Abstract

A method for determining a state of a measuring transducer integrated in a process container, wherein in the process container one or more processes are being performed, and the measuring transducer registers at least one physical or chemical process parameter within the process container, includes steps as follows: identifying a process currently being performed in the process container; and ascertaining a deviation value as measure of a deviation of a measured value progression registered by the measuring transducer during the process currently being performed in the process container from a measured value progression expected for the identified process, wherein the state of the measuring transducer and/or of the process is determined utilizing the ascertained deviation value.

First claim

The invention claimed is:

1. A method for determining a state of a measuring transducer integrated in a process container, comprising the steps of:
identifying a process currently being performed in the process container, wherein one or more processes are being performed in the process container; and
ascertaining a deviation value as measure of a deviation of a measured value progression registered by the measuring transducer during the process currently being performed in the process container from a measured value progression expected for the identified process, wherein the measuring transducer registers at least one physical or chemical process parameter within the process container,
wherein the state of the measuring transducer and/or of the process is determined utilizing the deviation value, and
wherein the process currently being performed in the process container is identified using a classifier such that the classifier associates the currently registered, measured value progression or at least one or more features of the measured value progression with a predetermined process class and determines as the deviation value a probability, with which the current measured value progression represents a process of the process class.

2. The method as claimed in claim 1 , wherein:
the process currently being performed in the process container is identified based on the measured value progression of the process parameter currently registered by means of the measuring transducer.

3. The method as claimed in claim 1 , further comprising the step of:
comparing the currently registered, measured value progression with at least one expected measured value progression, which represents a predetermined process class, wherein:
in the case of sufficiently small deviation of the currently registered, measured value progression from the expected measured value progression, the process currently being performed in the process container and providing the basis for the currently registered, measured value progression is identified as a process of the predetermined process class.

4. The method as claimed in claim 3 , further comprising the step of:
comparing the currently registered, measured value progression and the expected measured value progression based on one or more features of the measured value progressions, especially based on the measured value progression, a gradient of the measured value progression, an exceeding of a threshold value, a subceeding of a threshold value, a local or absolute extreme of the measured value progression, a noise superimposed on the measured value progression or a plateau of the measured value progression.

5. The method as claimed in claim 1 , wherein:
the classifier is presented in a learning phase with one or more expected measured value progressions of the at least one process parameter, which measured value progressions represent the process class.

6. The method as claimed in claim 5 , wherein:
the classifier extracts by means of a machine learning method features of the one or more expected measured value progressions and associates the features with the process class.

7. The method as claimed in claim 1 , wherein:
the one or more expected measured value progressions are predetermined based on pretrials using a measuring transducer different from the measuring transducer but comparable as regards accuracy of measurement, dynamic range and measuring range, integrated at the location of use of the measuring transducer in the process container.

8. The method as claimed in claim 1 , wherein:
the process currently being performed in the process container is identified based on information from a process control controlling the process currently being performed in the process container.

9. The method as claimed in claim 1 , wherein:
the process currently being performed in the process container is identified based on information input by a user of a measuring and evaluation circuit of the measuring transducer.

10. The method as claimed in claim 1 , wherein:
the one or more processes are performed repeatedly during the lifetime of the measuring transducer in the process container, a change of the respectively ascertained deviation values is determined, and the state of the measuring transducer is determined based on the change.

11. The method as claimed in claim 10 , wherein:
a time development of the repeatedly ascertained deviation values is used for determining the state of the measuring transducer, especially for predicting a remaining life of the measuring transducer.

12. The method as claimed in claim 1 , wherein:
by means of a predetermined quality criterion, the currently ascertained deviation value is classified either as tolerable or as not tolerable; and
in the case, in which the deviation value is classified as not tolerable, an alarm or a warning is triggered.

13. The method as claimed in claim 1 , wherein:
by means of the measuring transducer itself or by means of an additional, auxiliary, measuring transducer likewise integrated in the process container, at least one supplemental parameter is registered, which enters into determining the state of the measuring transducer and, in given cases, into an additional determining of a state of the process currently being performed in the process container.

14. The method as claimed in claim 1 , wherein:
the measuring transducer registers measured value progressions of at least two different process parameters, the process currently being performed in the process container is identified, for example, based on the currently registered, measured value progressions; and
relative to a process parameter, preferably relative to all process parameters, in each case, a deviation value as measure of a deviation of a measured value progression registered by the measuring transducer during the process currently being performed in the process container from a measured value progression expected for the identified process is ascertained.

15. The method as claimed in claim 1 , wherein:
the ascertaining of the deviation value and the determining of the state of the measuring transducer based on the ascertained deviation value is performed by an evaluation circuit, which is at least partially a component of the measuring transducer and/or at least partially a component of a superordinated data processing unit connected with the measuring transducer, especially a measurement transmitter, a computer or a programmable logic controller.

16. An apparatus for performing a method for determining a state of a measuring transducer integrated in a process container, wherein in the process container one or more processes are being performed, and a measuring transducer registers at least one physical or chemical process parameter within the process container, the method comprising the steps of identifying a process currently being performed in the process container, and ascertaining a deviation value as measure of a deviation of a measured value progression registered by the measuring transducer during the process currently being performed in the process container from a measured value progression expected for the identified process, wherein the state of the measuring transducer and/or of the process is determined utilizing the deviation value, and wherein the process currently being performed in the process container is identified using a classifier such that the classifier associates the currently registered, measured value progression or at least one or more features of the measured value progression with a predetermin

Assignees

• Endress + Hauser Conducta Ges Für Mess- Und Regeltechnik Mbh + Co Kg
• Endress & Hauser Conducta Gmbh & Co Kg

Inventors

• Wittmer Detlev
• Steckenreiter Thomas

Classifications

• G01D18/00 Primary

  Testing or calibrating apparatus or arrangements provided for in groups G01D1/00 - G01D15/00 · CPC title

• G01N27/4165 Primary

  for pH meters · CPC title