Diagnostics data collection and analysis method and apparatus to diagnose vehicle component failures

• US 20080004764A1
• Filed: 06/30/2006
• Published: 01/03/2008
• Est. Priority Date: 06/30/2006
• Status: Active Grant

• Current Assignee: SPX Corporation
• Original Assignee: SPX Corporation
• Inventors: Underdal, Olav M., Liebl, Troy, Chinnadurai, Manokar, Gilbert, Harry M., Lipscomb, Edward
• Granted Patent: US 7,751,955 B2
• Time in Patent Office: Days
• US Class Current: 701/29
• CPC Class Codes: G05B 23/0235 based on a comparison with predetermined threshold or range, e.g. "classical methods", carried out during normal operation; threshold adaptation or choice; when or how to compare with the threshold
• CPC Class Codes: G07C 5/008 communicating information to a remotely located station transmission systems for measured values G08C
• CPC Class Codes: G07C 5/085 using electronic data carriers

Abstract

A vehicle diagnostic data collector/analyzer compiles historical vehicle diagnostic data, including measured operating parameters from a number of different vehicles operating under a variety of normal vehicle operating conditions and vehicle component failure conditions, and performs statistical analyses on various vehicle type/operating condition combinations to establish operating parameter ranges corresponding to normal operating conditions and various failure conditions. The diagnostic data collector/analyzer also measures real-time operating parameters on specific test-subject vehicles and evaluates similarities and differences between the measured operating parameters and the established ranges, and correlates the test data with known operating conditions in order to diagnose potential failure modes of vehicle components.

Claims

1. A computer-implemented method of analyzing vehicle test data to diagnose a failure mode of a vehicle component, comprising:

compiling a collection of historical test data points each of which includes at least one operational parameter measurement of an individual vehicle; and

defining a range corresponding to an operating condition of a vehicle type based on the collection of test data points.

2. The computer-implemented method of claim 1, wherein the test data points correspond to a plurality of vehicles under a plurality of discrete operating conditions.

3. The computer-implemented method of claim 1, wherein the operating condition corresponds to a normal operation of one or more individual vehicles of the vehicle type.

4. The computer-implemented method of claim 1, wherein the operating condition corresponds to the failure mode existing in one or more individual vehicles of the vehicle type.

5. The computer-implemented method of claim 1, wherein each of the test data points comprises a single set of operational parameter measurements recorded at a moment in time.

6. The computer-implemented method of claim 1, wherein each of the test data points comprises a sequence of operational parameter measurements recorded over a period of time.

7. The computer-implemented method of claim 1, wherein the step of defining comprises a method of automated reasoning.

8. The computer-implemented method of claim 7, wherein the automated reasoning comprises at least one method selected from the following:

a method of manifold learning, a method based on a neural network, and a method based on a genetic algorithm.

9. The computer-implemented method of claim 1, further comprising:

representing each of the test data points as a point in a multidimensional vector space; and

statistically analyzing a set of the test data points corresponding to the operating condition to define a parameter space corresponding to the operating condition in the multidimensional vector space, wherein the parameter space comprises the range.

10. The computer-implemented method of claim 9, wherein the step of statistically analyzing further comprises:

associating with the set a multidimensional probability distribution having a mean value and a multidimensional variable variance vector; and

optimizing the parameter space by identifying an optimal variance vector based on the set.

11. The computer-implemented method of claim 9, wherein the step of statistically analyzing further comprises mapping the failure mode to the parameter space.

12. The computer-implemented method of claim 9, further comprising performing a dimensionality reduction procedure on the set of test data points and a correlated set of diagnoses corresponding to at least some of the operating conditions.

13. The computer-implemented method of claim 9, further comprising performing a dimensionality reduction procedure on a plurality of failure modes and a correlated plurality of parameter spaces.

14. The computer-implemented method of claim 1, further comprising:

measuring at least one of a plurality of operational parameters of a test-subject vehicle;

comparing at least one of the measured parameters to the range; and

determining an operating condition of the vehicle based on one or more of the compared parameters lying within the range.

15. The computer-implemented method of claim 14, further comprising apprising a user of the operating condition.

16. The computer-implemented method of claim 14, further comprising identifying the operating condition as normal based on the range corresponding to a normal operating condition.

17. The computer-implemented method of claim 14, further comprising diagnosing the failure mode based on the range corresponding to a failure condition.

18. The computer-implemented method of claim 14, further comprising:

representing each of the measured parameters as a point in a multidimensional vector space; and

diagnosing the failure mode based on the point lying within a predefined first parameter space corresponding to the failure mode existing in one or more individual vehicles of the vehicle type, wherein the parameter space comprises the range and the step of comparing further comprises verifying whether the point lies within the parameter space.

19. The computer-implemented method of claim 18, further comprising statistically analyzing the point with respect to the first parameter space and a predefined second parameter space to determine a probability that the failure mode exists when the point lies within at least the first parameter space and the second parameter space, wherein the first parameter space is associated with the failure mode and the second parameter space is associated with a different failure mode, and the first parameter space at least partially overlaps the second parameter space.

20. A computer program product for analyzing vehicle test data to diagnose a failure mode of a vehicle component, comprising a computer-readable medium encoded with instructions configured to be executed by a processor in order to perform predetermined operations comprising:

compiling a collection of historical test data points each of which includes at least one operational parameter measurement of an individual vehicle; and

defining a range corresponding to an operating condition of a vehicle type based on the collection of test data points.

21. The computer program product of claim 20, wherein the step of defining comprises a method of automated reasoning.

22. The computer program product of claim 20, wherein the predetermined operations further comprise:

representing each of the test data points as a point in a multidimensional vector space; and

statistically analyzing a set of the test data points corresponding to the operating condition to define a parameter space corresponding to the operating condition in the multidimensional vector space, wherein the parameter space comprises the range.

23. The computer program product of claim 22, wherein the step of statistically analyzing further comprises mapping the failure mode to the parameter space.

24. The computer program product of claim 20, wherein the predetermined operations further comprise:

measuring at least one of a plurality of operational parameters of a test-subject vehicle;

comparing at least one of the measured parameters to the range; and

determining an operating condition of the vehicle based on one or more of the compared parameters lying within the range.

25. The computer program product of claim 24, wherein the predetermined operations further comprise diagnosing the failure mode based on the range corresponding to a failure condition.

26. The computer program product of claim 24, wherein the predetermined operations further comprise

representing each of the measured parameters as a point in a multidimensional vector space;

and

diagnosing the failure mode based on the point lying within a predefined first parameter space corresponding to the failure mode existing in one or more individual vehicles of the vehicle type, wherein the parameter space comprises the range and the step of comparing further comprises verifying whether the point lies within the parameter space.

27. A diagnostic tool for analyzing vehicle test data to diagnose a failure mode of a vehicle component, comprising:

a data compiler configured to compile a collection of historical test data points each of which includes at least one operational parameter measurement of an individual vehicle; and

a data analyzer configured to define a range corresponding to an operating condition of a vehicle type based on the collection of test data points.

28. The diagnostic tool of claim 27, wherein the data analyzer is further configured to define the range using a method of automated reasoning.

29. The diagnostic tool of claim 27, wherein the data analyzer is further configured to represent each of the test data points as a point in a multidimensional vector space and to statistically analyze a set of the test data points corresponding to the operating condition to define a parameter space corresponding to the operating condition in the multidimensional vector space, wherein the parameter space comprises the range.

30. The diagnostic tool of claim 29, wherein the data analyzer is further configured to map the failure mode to the parameter space.

31. The diagnostic tool of claim 27, further comprising:

a parameter reader configured to measure at least one of a plurality of operational parameters of a test-subject vehicle;

a comparator configured to compare at least one of the measured parameters to the range; and

a condition identifier configured to determine an operating condition of the vehicle based on one or more of the compared parameters lying within the range.

32. The diagnostic tool of claim 31, further comprising a virtual diagnostician configured to diagnose the failure mode based on the range corresponding to a failure condition.

33. The diagnostic tool of claim 31, wherein the data analyzer is further configured to represent each of the measured parameters as a point in a multidimensional vector space, and the virtual diagnostician is further configured to diagnose the failure mode based on the point lying within a predefined first parameter space corresponding to the failure mode existing in one or more individual vehicles of the vehicle type, wherein the parameter space comprises the range and the step of comparing further comprises verifying whether the point lies within the parameter space.

34. A diagnostic tool for analyzing vehicle test data to diagnose a failure mode of a vehicle component, comprising:

means for compiling a collection of historical test data points each of which includes at least one operational parameter measurement of an individual vehicle; and

means for defining a range corresponding to an operating condition of a vehicle type based on the collection of test data points.

35. The diagnostic tool of claim 34, wherein the step of defining comprises a method of automated reasoning.

36. The diagnostic tool of claim 34, further comprising:

means for representing each of the test data points as a point in a multidimensional vector space; and

means for statistically analyzing a set of the test data points corresponding to the operating condition to define a parameter space corresponding to the operating condition in the multidimensional vector space, wherein the parameter space comprises the range.

37. The diagnostic tool of claim 36, wherein the step of statistically analyzing further comprises mapping the failure mode to the parameter space.

38. The diagnostic tool of claim 34, further comprising:

means for measuring at least one of a plurality of operational parameters of a test-subject vehicle;

means for comparing at least one of the measured parameters to the range; and

means for determining an operating condition of the vehicle based on one or more of the compared parameters lying within the range.

39. The diagnostic tool of claim 38, further comprising means for diagnosing the failure mode based on the range corresponding to a failure condition.

40. The diagnostic tool of claim 38, further comprising:

means for representing each of the measured parameters as a point in a multidimensional vector space; and

means for diagnosing the failure mode based on the point lying within a predefined first parameter space corresponding to the failure mode existing in one or more individual vehicles of the vehicle type, wherein the parameter space comprises the range and the step of comparing further comprises verifying whether the point lies within the parameter space.