Machine diagnosis system

• US 6,144,923 A
• Filed: 02/06/1998
• Issued: 11/07/2000
• Est. Priority Date: 02/22/1997
• Status: Expired due to Term

• Current Assignee: Voith Hydro GmbH & Co. KG (Germany) (Voith GmbH)
• Original Assignee: Voith Hydro GmbH & Co. KG (Germany) (Voith GmbH)
• Inventors: Grosse, Gilbert
• Primary Examiner(s): Hua, Ly V.
• Assistant Examiner(s): Crockett, Robert G.
• Application Number:
  US09/019,619
• Time in Patent Office: 1,005 Days
• Field of Search: 714/25, 700/159, 700/169, 702/179, 702/185, 702/56, 73/659
• US Class Current: 702/56
• CPC Class Codes: G01H 1/003 of rotating machines G01H1/10 takes precedence

Abstract

The invention, that relates to a machine diagnosis system for the state-oriented operation monitoring of a machine, comprises a characteristic value module and a cause module. The characteristic value module, proceeding from machine-state referred measurement values, establishes diagnosis-relevant characteristic values, and the cause module diagnoses from the characteristic values a cause for the detected measurement values. There the cause module calculates for at least one possible cause in each case a cause probability which indicates with what probability the corresponding cause is responsible for the presence of the detected oscillation measurement values.

Claims

1. Machine diagnosis system for state-oriented operation monitoring of a machine, comprising:

a characteristic value module which, proceeding from machine state-related measurement values, forms diagnosis-relevant characteristic values that are linked to operating parameters; and

a cause module which, from the characteristic values, diagnoses a cause for the detected measurement values,

wherein the cause module, for at least one possible cause in each case, calculates a cause probability with the aid of at least one probability-characteristic value relation described by a polygonal course, which indicates with what probability the corresponding cause is responsible for the presence of the detected measurement values.

2. Machine diagnosis system according to claim 1, wherein the cause module for the calculation of the cause probability, calls upon, besides individual characteristic values, at least one characteristic value linkage described by a function of a t least two of the characteristic values that have cross-linked causal interconnections between them.

3. Machine diagnosis system according to claim 1, wherein the cause module calculates the cause probability as a weighted arithmetic mean of individual probabilities, in which the individual probabilities are allocated to individual characteristic values or to characteristic value linkages and, for the mean value formation, are weighted with individual weighting factors.

4. Machine diagnosis system according to claim 3, wherein the individual weighting factors are influenced by characteristic values.

5. Machine diagnosis system according to claim 1, wherein the characteristic value module forms characteristic values dependent on operating parameters and that the cause module draws upon the dependence of the characteristic values or of characteristic value linkages of the operating parameters for the calculation of the cause probability.

6. Machine diagnosis system according to claim 5, wherein the characteristic values or characteristic value linkages are time-dependent, and wherein the characteristic value module assigns the characteristic values or characteristic value linkages to certain operating parameter intervals in order to form characteristic value-operating parameter histograms.

7. Machine diagnosis system according to claim 1, wherein the machine is a rotating hydraulic flow machine, the oscillations of which are to be diagnosed.

8. Machine diagnosis system according to claim 1, wherein the at least one polygonal course is established by a plurality of number pairs stored in a knowledge basis.

9. Machine diagnosis system according to claim 7, wherein the spectral constituents of at least one of shaft oscillations, orbit-characteristics of the shaft oscillations, and phase angles of the shaft oscillations are characteristic values.

10. Machine diagnosis system according to claim 5, wherein an operating parameter is processed, which clearly indicates the momentary operation type, namely whether one of a pump operation, a turbine operation, a phase shifter operation, a starting of the machine, and a running-down of the machine is present.

11. Machine diagnosis system according to claim 5, wherein at least one of momentary performance of the machine, suction tube pressure, tube track pressure, guide wheel position, and drop height of the hydraulic flow machine, is an operating parameter.

12. Machine diagnosis system according to claim 1, wherein a warning module warns the machine operator on presence of a certain cause probability or of a certain distribution of cause probabilities.

13. Machine diagnosis system according to claim 1, wherein a reaction module on presence of a certain cause probability or of a certain distribution of cause probabilities can interfere into the operation of the machine.

14. Machine diagnosis system according to claim 1, wherein the calculation of the cause probability by the cause module proceeds from a knowledge-basis which comprises machine-individual experience knowledge and a theoretical machine model.

15. Machine diagnosis system according to claim 1, wherein the characteristic value module for the formation of the characteristic values proceeds from a knowledge-basis, which comprises machine-individual experience knowledge and a theoretical machine model.

16. Machine diagnosis system according to claim 1, wherein the knowledge-basis is extensible or modifiable by the experience of the machine operator.

24. Process according to claim 1, wherein the at least one polygonal course is established by a plurality of number pairs stored in a knowledge basis.

17. Process for diagnosing the state-oriented operation monitoring of a machine, comprising:

forming diagnosis-relevant characteristic values that are linked to operating parameters from machine state-referenced measurement values; and

diagnosing a cause for the detected measurement values from the characteristic values;

further comprising for at least one possible cause in each case,

calculating a cause probability with the aid of at least one probability-characteristic value relation described by a polygonal course, which indicates with what probability the corresponding cause is responsible for the presence of the detected measurement values.

18. Process according to claim 17, wherein for the calculation of the cause probability, besides individual characteristic values, at least one characteristic value linkage is drawn upon, the characteristic value linkage being described by a function of at least two of the characteristic values that have cross-linked causal interconnections between them.

19. Process according to claim 17, wherein the cause probability is calculated as weighted arithmetic mean of individual probabilities, in which the individual probabilities allocated to individual characteristic values or characteristic value linkages are weighted for mean value formation with individual knowledge-based weighting factors.

20. Process according to claim 19, wherein the mean value formation occurs with characteristic value-influenced weighting factors.

21. Process according to claim 17, further comprising:

forming characteristic values dependent on operating parameters, in which operating parameters are operation type-referred measurement values, and

drawing upon the dependence of the characteristic values or characteristic value linkages by the operating parameters for the calculation of the cause probability.

22. Process according to claim 21, wherein the characteristic values or characteristic value linkages are time-dependent, and wherein the characteristic values or characteristic value linkages are allocated to certain operating parameter intervals in order to form characteristic value-operating parameter histograms.

23. Process according to claim 17, wherein the machine is a rotating hydraulic flow machine, the oscillations of which are to be diagnosed.

25. Process according to claim 23, wherein at least one of the spectral constituents of shaft oscillations, orbit characteristics of the shaft oscillations, and phase angles of the shaft oscillations are characteristic values.

26. Process according to claim 21, wherein an operating parameter is used which indicates whether one of a pump operation, a turbine operation, a phase shifter operation, a starting of the machine, and a running-down of the machine is present.

27. Process according to claim 21, wherein at least one of momentary performance of the machine, suction tube pressure, pipe track pressure, guide wheel position, and fall height of the hydraulic flow machine, is used as operating parameter.

28. Process according to claim 17, wherein on presence of a certain cause probability or of a certain distribution of cause probabilities, the machine operator is warned.

29. Process according to claim 17, wherein on presence of a certain cause probability or of a certain distribution of cause probabilities, it is possible to intervene in the operation of the machine.

30. Process according to claim 17, wherein the calculation of the cause probability proceeds from a knowledge-basis which comprises machine-individual experience knowledge and a theoretical machine model.

31. Process according to claim 17, wherein the formation of the characteristic values proceeds from a knowledge basis which comprises machine-individual experience knowledge and a theoretical machine model.

32. Process according to claim 30, wherein the machine operator can extend or modify the knowledge basis with his rules.