Granted Patent B2

US 12,411,483 · App. 17/913,872 · Granted Sep 9, 2025

Inventors: Thomas Bierweiler (Karlsruhe, DE); Jean Pascal John (Munich, DE); Daniel Labisch (Karlsruhe, DE)

Assignee: Siemens Aktiengesellschaft

G05B23/024

Abstract

Claims (59)

1. A method for diagnosing messages generated during operation of a technical plant, a process engineering process running within the plant being controlled via a process control system including a planning and engineering tool, an operator control and monitoring device and a plurality of automation devices, which are interconnected via a bus system for data communication, data sets, characterizing the operation of the technical plant, with values of process variables being recorded and stored in a data memory, diagnosis statements about the operation of the technical plant being determined via the data sets and at least one predetermined, self-organizing map, and n-tuples with predetermined values of process variables for an error-free operation of the technical plant corresponding to the data sets being stored at nodes of the self-organizing map, the method comprising:

determining, during operation of the technical plant, a smallest distance between data sets of a current time interval and data sets of nodes of the self-organizing map for a current time interval, within which a message has occurred following occurrence of the message, and selecting a corresponding node following occurrence of the message;

determining at least one current symptom to form differences from the data sets of the current time interval for a previously determined node during operation of the technical plant, only the differences of process variables which exceed a predetermined threshold value being considered;

issuing process variables involved in at least one previously determined symptom during operation of the technical plant; and

controlling the process engineering process via the process control system and implementing changes to the at least one predetermined, self-organizing map or predetermined threshold value during operation of the technical plant to ensure the error-free operation of the technical plant.

2. The method as claimed in claim 1 , following occurrence of the message, the method further comprising:

determining how often the message has previously occurred;

determining smallest distances between data sets of historical time intervals and the data sets of the nodes of the self-organizing map for time intervals in the past, within which the message has occurred, selecting corresponding nodes, wherein i>=1;

determining historical symptoms to form differences from the data sets of the historical time intervals for previously determined nodes, only the differences of process variables which exceed the predetermined threshold value being considered; and

issuing process variables involved in the previously determined historical symptoms.

3. The method as claimed in claim 2 , following the occurrence of the message, the method further comprising:

determining whether the message forms part of a message chain and how often the message chain has previously occurred;

determining the smallest distances between the data sets of the time intervals and the data sets of the nodes of the self-organizing map for all time intervals within which the message chain has occurred, and selecting corresponding nodes, wherein i>=1;

determining current and historical symptoms to form differences from the data sets of the time intervals for the previously determined nodes, only differences of process variables which exceed the predetermined threshold value being considered; and

issuing process variables which are involved in previously determined symptoms.

4. The method as claimed in claim 2 , further comprising:

performing a clustering of the symptoms;

selecting a cluster containing at least one current symptom; and

issuing the process variables which are involved in the at least one current symptom.

5. The method as claimed in claim 3 , further comprising:

performing a clustering of the symptoms;

selecting a cluster containing at least one current symptom; and

issuing the process variables which are involved in the at least one current symptom.

6. The method according to claim 1 , wherein the time intervals are one of (i) determinable variably and (ii) set in accordance with process dynamics.

7. The method according to claim 1 , wherein symptoms are determined for a random selection of time points of the time intervals.

8. The method according to claim 1 , wherein the threshold value is variably determinable for determining the symptom.

9. The method according to claim 1 , wherein for symptoms of a cluster, historical process variables of the historical time intervals which are involved in the symptoms in this cluster are also issued.

10. The method according to claim 1 , wherein information from a shift log of the technical plant with the symptoms is further also issued; and wherein information in a short time interval prior to and/or after the symptoms is also considered.

11. A system for diagnosing messages of a process control system of a technical plant, the process control system including a planning and engineering tool, an operator control and monitoring device and a plurality of automation devices, which are interconnected via a bus system for data communication, the system comprising:

a data memory in which at least one data set characterizing the operation of the technical plant with values of process variables is storable; and

an evaluation facility;

wherein the evaluation facility is configured to:

determine diagnosis statements about the operation of the technical plant via the data sets and at least one predetermined self-organizing map, n-tuples corresponding to the data sets with predetermined values of process variables for an error-free operation of the technical plant being are stored at nodes of the self-organizing map; and

wherein following occurrence of a message the evaluation facility is further configured to:

determine a smallest distance between data sets of a current time interval and data sets of nodes of the self-organizing map for a current time interval, within which a message has occurred following occurrence of the message, and select a corresponding node following occurrence of the message;

determine at least one current symptom to form differences from the data sets of the current time interval for a previously determined node, only the differences of process variables which exceed a predetermined threshold value being considered; and

issue process variables involved in at least one previously determined symptom; and

wherein a process engineering process is controlled via the process control system and changes are implemented to the at least one predetermined, self-organizing map or predetermined threshold value during operation of the technical plant to ensure the error-free operation of the technical plant.

12. The system as claimed in claim 11 , further comprising:

an analysis facility connected to evaluation facility, said analysis facility being configured to calculate and analyze message chains evaluation facility and being configured to:

determine whether the message forms part of a message chain and how often the message chain has previously occurred;

determine the smallest distances between the data sets of the time intervals and the data sets of the nodes of the self-organizing map for all time intervals within which the message chain has occurred, and selecting corresponding nodes, wherein i>=1;

determine current and historical symptoms to form differences from the data sets of the time intervals for the previously determined nodes, only differences of process variables which exceed a predetermined threshold value being considered; and

issue the process variables which are involved in the previously determined symptoms.

13. The system according to claim 11 , wherein the evaluation facility is further configured to perform a clustering of previously calculated symptoms, select a cluster which contains at least one current symptom, and issue the process variables which are involved in at least one current symptom.

14. The system according to claim 12 , wherein the evaluation facility is further configured to perform a clustering of previously calculated symptoms, select a cluster which contains at least one current symptom, and issue the process variables which are involved in at least one current symptom.

15. The system according to claim 11 , wherein the evaluation facility is further configured to one of (i) variably determine the time intervals and (ii) set the time intervals in accordance with process dynamics.

16. The system according to claim 11 , wherein the evaluation facility is further configured to determine symptoms for a random selection of points in time of the time intervals.

17. The system according to claim 11 , wherein the evaluation facility is further configured such that a threshold value is variably determinable for determining the symptom.

18. The system according to claim 11 , wherein the evaluation facility is further configured to also issue historical process variables of historical time intervals for symptoms of a cluster.

19. The system according to claim 11 , wherein the evaluation facility is connected with a component which comprises a shift log and is connected with an analysis facility which is configured to evaluate information of the shift log of the technical plant in connection with the symptoms and to issue these to the evaluation facility.

20. The system according claim 11 , wherein the evaluation facility is further connected with a display device, which is configured to display a selection of at least one of (i) a message or calculated message chain, (ii) a frequency of occurrence of the message or message chain, (iii) current and historical time intervals with a selection of process variables, (iv) process variables that are involved in historical symptoms and (v) process variables that are involved in current symptoms.

21. A non-transitory computer-readable product medium encoded with a computer program which, when executed by a processor of a computer causes messages of a process control system of a technical plant to be diagnosed, the process control system including a planning and engineering tool, an operator control and monitoring device and a plurality of automation devices, which are interconnected via a bus system for data communication, the computer program comprising:

program code for determining, during operation of the technical plant, a smallest distance between data sets of a current time interval and data sets of nodes of the self-organizing map for a current time interval, within which a message has occurred following occurrence of the message, and selecting a corresponding node following occurrence of the message;

determining at least one current symptom to form differences from the data sets of the current time interval for a previously determined node during operation of the technical plant, only the differences of process variables which exceed a predetermined threshold value being considered;

program code for issuing process variables involved in at least one previously determined symptom during operation of the technical plant; and

program code for controlling the process engineering process via the process control system and for implementing changes to the at least one predetermined, self-organizing map or predetermined threshold value during operation of the technical plant to ensure the error-free operation of the technical plant.

22. The non-transitory computer-readable medium according to claim 21 , wherein the computer-readable medium comprises a data carrier or storage medium.

23. A user interface, which is displayed on a display unit, and which is configured to display analysis results of the system as claimed in 11 .

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (20)