Statistical and trend analysis of railroad bearing temperatures

• US 6,813,581 B1
• Filed: 03/26/2003
• Issued: 11/02/2004
• Est. Priority Date: 03/26/2003
• Status: Active Grant

• Current Assignee: Union Pacific Railroad Company (Union Pacific Corporation)
• Original Assignee: Union Pacific Railroad Company (Union Pacific Corporation)
• Inventors: Snyder, Todd W.
• Primary Examiner(s): Barlow, John
• Assistant Examiner(s): Walling, Meagan S
• Application Number:
  US10/400,589
• Time in Patent Office: 587 Days
• Field of Search: 702/130, 702/99, 702/136, 702/179, 374/4, 374/45, 374/101, 374/141, 374/153, 246/169.A
• US Class Current: 702/130
• CPC Class Codes: G01K 1/022 for recording
• CPC Class Codes: G01K 7/42 Circuits effecting compensation of thermal inertia; Circuits for predicting the stationary value of a temperature

Abstract

A system is provided for tracking the wheel bearing temperatures of a train and selecting those that are statistically determined to be abnormally high, while eliminating from consideration, or otherwise categorizing those high temperature readings which are the result of other anomalous and non-anomalous conditions. The system determines the overall distribution of the bearing temperatures for both sides of a train at each wayside location. The bearing temperatures are ranked for comparison with the overall train distribution. Outliers are detected and characterized based on temperature patterns and previous history.

Claims

1. A system for analyzing temperature data received from the wheel bearings of trains, comprising:

a processor;

software operative on said processor for sorting the temperature data into at least a first grouping of temperature data, a second grouping of temperature data, and a third grouping of temperature data;

said first group of temperature data comprising temperatures lower than temperatures grouped in said second grouping of temperature data and said third grouping of temperature data;

said third grouping of temperature data comprising temperatures higher than the temperatures in said second grouping of temperature data;

said software being further operative on said processor for determining the highest temperature within said first grouping of temperature data and determining the lowest temperature within said third grouping of temperature data;

said software being further operative on said processor for detecting temperature outliers within said temperature data;

and means operatively coupled to said processor for displaying information relating to said temperature outliers.

2. The system of claim 1 wherein said first grouping of temperature data and said third grouping of temperature data are comprised of a relatively equal number of the temperature data.

3. The system of claim 1 wherein said first grouping of temperature data is comprised of approximately the lowest 25 percent of the temperature data.

4. The system of claim 2 wherein said third grouping of temperature data is comprised of approximately the highest 25 percent of the temperature data.

5. The system of claim 1 wherein said software is further operative on said processor for calculating an outlier quantifier for said temperature outliers detected within said temperature data.

6. The system of claim 5 wherein said outlier quantifier is calculated by dividing the temperature difference between at least one of said temperature outliers in said third grouping of temperature data and said lowest temperature within said third grouping of temperature data by the difference between said lowest temperature within said third grouping of temperature data and said highest temperature within said first grouping of temperature data.

7. The system of claim 5 wherein said outlier quantifier is calculated by dividing the temperature difference between at least one of said temperature outliers in said first grouping of temperature data and said highest temperature within said first grouping of temperature data by the difference between said lowest temperature within said third grouping of temperature data and said highest temperature within said first grouping of temperature data.

8. The system of claim 1 wherein said software is further operative on said processor for determining an identity of the train.

9. The system of claim 1 wherein said software is further operative on said processor for identifying a location on the train from which said temperature outliers are detected.

10. The system of claim 1 further comprising memory means operatively connected to said processor for storing data relating to said temperature outliers as historical data.

11. The system of claim 10 wherein said software is further operative on said processor for storing the temperature data within said memory means as historical temperature data.

12. The system of claim 10 wherein said software is further operative on said processor for comparing the temperature data to said historical temperature data.

13. The system of claim 1 wherein said software is further operative on said processor for determining a cause of said temperature outliers.

14. The system of claim 1 wherein said software is further operative on said processor for determining a level of severity for said temperature outliers.

15. A method of analyzing temperature data received from the wheel bearings of trains, comprising the steps of:

providing a processor, software that is operative on said processor, and display means operatively coupled to said processor for displaying information;

sorting the temperature data into at least a first grouping of temperature data, a second grouping of temperature data, and a third grouping of temperature data;

said first grouping of temperature data comprising temperatures lower than temperatures grouped in said second grouping of temperature data and said third grouping of temperature data;

said third grouping of temperature data comprising temperatures higher than the temperatures in said second grouping of temperature data;

determining the highest temperature within said first grouping of temperature data and the lowest temperature within said third grouping of temperature data;

detecting temperature outliers within said temperature data; and

displaying information relating to said temperature outliers on said display means.

16. The method of claim 15 further comprising the step of providing memory means operatively coupled to said processor for storing the temperature data and said temperature outliers as historical data.

17. The method of claim 15 further comprising the steps of calculating an outlier quantifier for said temperature outliers detected within said temperature data.

18. The method of claim 17 wherein said outlier quantifier is calculated by dividing the temperature difference between at least one of said temperature outliers and said lowest temperature within said third grouping of temperature data by the difference between said lowest temperature within said third grouping of temperature data and said highest temperature within said first grouping of temperature data.

19. The method of claim 17 wherein said outlier quantifier is calculated by dividing the temperature difference between at least one of said temperature outliers in said first grouping of temperature data and said highest temperature within said first grouping of temperature data by the difference between said lowest temperature within said third grouping of temperature data and said highest temperature within said first grouping of temperature data.

20. The method of claim 15 further comprising the step of determining an identify of the train.

21. The method of claim 15 further comprising the step of identifying a location on the train from which said temperature outliers are detected.

22. The method of claim 15 further comprising the step of determining a cause for said temperature outliers.

23. The method of claim 15 further comprising the step of determining a level of severity for said temperature outliers.