Method for the computer-assisted evaluation of characteristic values of a technical system

• US 7,630,952 B1
• Filed: 05/04/2005
• Issued: 12/08/2009
• Est. Priority Date: 05/05/2004
• Status: Expired due to Fees

• Current Assignee: Siemens AG
• Original Assignee: Siemens AG
• Inventors: Greiner, Michael, Elbel, Benedikte, Dunkel, Veronika, Meintrup, David
• Primary Examiner(s): Vincent; David R
• Assistant Examiner(s): Fernandez Rivas; Omar F
• Application Number:
  US11/579,631
• Time in Patent Office: 1,679 Days
• Field of Search: 706/12, 706/14, 706 45- 47, 706/60, 706/62, 702/1, 702/59, 702/108, 702/127, 702179-186, 717/104, 717124-135, 717151-156
• US Class Current: 706/62
• CPC Class Codes: G05B 23/0221 Preprocessing measurements, e.g. data collection rate adjustment; Standardization of measurements; Time series or signal analysis, e.g. frequency analysis or wavelets; Trustworthiness of measurements; Indexes therefor; Measurements using easily measured parameters to estimate parameters difficult to measure; Virtual sensor creation; De-noising; Sensor fusion; Unconventional preprocessing inherently present in specific fault detection methods like PCA-based methods

Abstract

During a total temporal interval a plurality of characteristic values of the technical system are determined. The total temporal interval is divided into a plurality of partial intervals, each partial interval being between a partial interval starting point and a partial interval end point in the total interval. According to said method, for each partial interval, at least one parameter of a pre-determined predictive model is adapted to the characteristic values determined in the partial interval in such a way as to obtain a partial adaptation with which a partial interval and the end point of the partial interval are associated. Respectively one modification measure is determined for the partial adaptations in such a way as to indicate the modifications of the parameters of the respective partial adaptation in relation to the parameters of at least one adjacent partial adaptation. The interval between the partial interval end point of a respective partial adaptation and the partial interval end point of the at least one adjacent partial adaptation is marked as an inhomogeneous interval, if the modification measure is higher than a predetermined value.

Claims

1. A method for computer-assisted evaluation of characteristic values of a technical system, comprising:

dividing a total temporal interval into a plurality of partial intervals, the total temporal interval having a start point and an end point between which a plurality of characteristic values of the technical system are determined, each partial interval occurring between a partial interval start point and a partial interval end point within the total temporal interval, each partial interval having at least one characteristic value determined therein;

adapting, for each partial interval, one or more parameters of a predetermined predictive model, the parameters being adapted to the at least one characteristic value determined in the partial interval, the parameters being adapted to provide partial adaptations, each partial adaptation having a corresponding partial interval and a corresponding partial interval end point;

allocating the adjacent partial intervals with respective adjacent partial adaptations, the partial interval end point of each partial interval being adjacent to the partial interval end point of the respective partial adaptation;

determining a modification measure for each of the partial adaptations specifying the modification of the parameters of a respective partial adaptation with reference to the parameters of at least one adjacent partial adaptation;

determining if each modification measure is greater than a predefined value; and

for each modification measure that is greater than the predefined value, flagging as an inhomogeneous interval the time interval between the partial interval end point of the respective partial adaptation and the partial interval end point of the adjacent partial adaptation.

2. The method as claimed in claim 1, wherein

at least a portion of the partial intervals have first and second adjacent partial intervals, the first adjacent partial interval being associated with an earlier partial interval endpoint and the second adjacent partial interval being associated with a later partial interval endpoint,

at least one of the modification measures is determined to modify the parameters to both the first and second adjacent partial intervals, and

the inhomogeneous interval is defined as a time period from the earlier partial interval endpoint to the later partial interval endpoint.

3. The method as claimed in claim 1, wherein the modification measure specifies the amount the parameters of the relevant partial adaptation were modified to the parameters of the at least one adjacent partial adaptation.

4. The method as claimed in claim 1, wherein each partial interval begins at the start point of the total temporal interval.

5. The method as claimed in claim 1, wherein the one or more parameters of the predetermined predictive model are adapted to the at least one characteristic value using a maximum likelihood method and/or a least squares method.

6. The method as claimed in claim 1, wherein the predetermined predictive model is a reliability prediction model, and the characteristic values are values which represent reliability of the technical system.

7. The method as claimed in claim 6, wherein

the characteristic values are each determined at a respective instant in time, and

each characteristic value represents a number of total failures of the technical system at the instant the characteristic value was determined and/or

each characteristic value represent an average time before the technical system will fail at the instant the characteristic value was determined.

8. The method as claimed in claim 1, wherein

the technical system has a processor on which a computer program is executed, and

the predetermined predictive model is a reliability growth model for the computer program.

9. The method as claimed in claim 1, wherein

the total temporal interval is defined as the time period during which a test and correction phase of the technical system occurs, and

during the test and correction phase, the technical system is continuously adapted to improve its reliability.

10. The method as claimed in claim 2, wherein the modification measure specifies the amount the parameters of the relevant partial adaptation were modified to the parameters of the at least one adjacent partial adaptation.

11. The method as claimed in claim 10, wherein each partial interval begins at the start point of the total temporal interval.

12. The method as claimed in claim 11, wherein the one or more parameters of the predetermined predictive model are adapted to the at least one characteristic value using a maximum likelihood method and/or a least squares method.

13. The method as claimed in claim 12, wherein

the predetermined predictive model is a reliability prediction model, and

the characteristic values are values which represent reliability of the technical system.

14. The method as claimed in claim 13, wherein

the characteristic values are each determined at a respective instant in time, and

each characteristic value represents a number of total failures of the technical system at the instant the characteristic value was determined and/or

each characteristic value represent an average time before the technical system will fail at the instant the characteristic value was determined.

15. The method as claimed in claim 14, wherein

the technical system has a processor on which a computer program is executed, and

the predetermined predictive model is a reliability growth model for the computer program.

16. The method as claimed in claim 15, wherein

the total temporal interval is defined as the time period during which a test and correction phase of the technical system occurs, and

during the test and correction phase, the technical system is continuously adapted to improve its reliability.

17. A computer readable medium having stored thereon a computer program which when executed by a computer causes the computer to perform a method for evaluating characteristic values of a technical system, the method comprising:

dividing a total temporal interval into a plurality of partial intervals, the total temporal interval having a start point and an end point between which a plurality of characteristic values of the technical system are determined, each partial interval occurring between a partial interval start point and a partial interval end point within the total temporal interval, each partial interval having at least one characteristic value determined therein;

adapting, for each partial interval, one or more parameters of a predetermined predictive model, the parameters being adapted to the at least one characteristic value determined in the partial interval, the parameters being adapted to provide partial adaptations, each partial adaptation having a corresponding partial interval and a corresponding partial interval end point;

allocating the adjacent partial intervals with respective adjacent partial adaptations, the partial interval end point of each partial interval being adjacent to the partial interval end point of the respective partial adaptation;

determining a modification measure for each of the partial adaptations specifying the modification of the parameters of a respective partial adaptation with reference to the parameters of at least one adjacent partial adaptation;

determining if each modification measure is greater than a predefined value; and

for each modification measure that is greater than the predefined value, flagging as an inhomogeneous interval the time interval between the partial interval end point of the respective partial adaptation and the partial interval end point of the adjacent partial adaptation.

18. A computer-assisted system for evaluating characteristic values of a technical system, comprising:

a division unit to divide a total temporal interval into a plurality of partial intervals, the total temporal interval having a start point and an end point between which a plurality of characteristic values of the technical system are determined, each partial interval occurring between a partial interval start point and a partial interval end point within the total temporal interval, each partial interval having at least one characteristic value determined therein;

an adaptation unit to adapt, for each partial interval, one or more parameters of a predetermined predictive model, the parameters being adapted to the at least one characteristic value determined in the partial interval, the parameters being adapted to provide partial adaptations, each partial adaptation having a corresponding partial interval and a corresponding partial interval end point;

an identification unit to allocate the adjacent partial intervals with respective adjacent partial adaptations, the partial interval end point of each partial interval being adjacent to the partial interval end point of the respective partial adaptation;

a first determination unit to determine a modification measure for each of the partial adaptations specifying the modification of the parameters of a respective partial adaptation with reference to the parameters of at least one adjacent partial adaptation;

a second determination unit to determine if each modification measure is greater than a predefined value; and

a flag unit, for each modification measure that is greater than the predefined value, to flag as an inhomogeneous interval the interval between the partial interval end point of the respective partial adaptation and the partial interval end point of the adjacent partial adaptation.