Apparatus and method for identifying a defect and/or operating characteristic of a system

• US 8,160,832 B2
• Filed: 02/19/2010
• Issued: 04/17/2012
• Est. Priority Date: 06/06/2007
• Status: Active Grant

• Current Assignee: Progress Rail Services Corporation (Caterpillar Incorporated)
• Original Assignee: Progress Rail Services Corporation (Caterpillar Incorporated)
• Inventors: Luo, Huageng, Durling, Michael Richard, Ruggiero, Andrew Lawrence, Sun, Zongqi, Mathews, Harry Kirk Jr., Tait, Robert William, Church, Benjamin Paul, Tougas, Thomas Anthony
• Primary Examiner(s): Nghiem, Michael
• Assistant Examiner(s): Le, Toan M
• Application Number:
  US12/708,921
• Publication Number:
  US 20100141441A1
• Time in Patent Office: 788 Days
• Field of Search: 702/99, 702 33- 36, 702/39, 702 41- 43, 702/56, 702/66, 702 70- 73, 702 75- 76, 702/81, 702 84- 85, 702/101, 702103-106, 702/127, 702130-132, 702/136, 702142-143, 702145-148, 702173-176, 702182-185, 104/18, 104/41, 104/43, 104/46, 105/1.4, 105/26.05, 105 53- 54, 105/96, 105154-155, 105/218.1, 246/2.F, 246/169.A, 246/169.R, 246/169.S, 701 19- 20, 374/4, 374/100, 374/117, 374/119, 374/121, 374/124, 374/132, 374/137, 374141-142, 374/159, 374186-187
• US Class Current: 702/99
• CPC Class Codes: B61K 9/04 Detectors for indicating the overheating of axle bearings and the like, e.g. associated with the brake system for applying the brakes in case of a fault
• CPC Class Codes: B61L 25/021 Measuring and recording of train speed
• CPC Class Codes: B61L 25/026 Relative localisation, e.g. using odometer
• CPC Class Codes: G01K 13/08 in rotary movement
• CPC Class Codes: G01K 15/00 Testing or calibrating of thermometers

Abstract

An apparatus and methods for identifying a defect and/or an operating characteristic of a system being monitored (and/or one or more of the system'"'"'s components) are described. In an embodiment, orthogonally related data monitored by two or more detectors may be fused to determine whether a component of a system is defective and/or malfunctioning. Additionally or alternatively, data from a first detector may be determined to be accurate using non-orthogonally related data outputted by a second detector. Both types of determinations may be made with minimal or no false indications, which lowers the cost of operating the system being monitored. Embodiments of the invention may also be configured to forecast and/or prevent accidents and/or damage to the system being monitored by predicting whether a defect and/or a malfunction will occur.

Claims

1. A method, comprising:

collecting acoustical data produced by a system passing in proximity to an acoustic detector having a microphone;

analyzing the acoustical data to identify an operating characteristic of the system;

performing signal-processing on the acoustical data, wherein the signal-processing at least includes an instant frequency analysis method; and

fusing the acoustical data with orthogonally-related data produced by the system passing in proximity to an orthogonally-related detector to improve the accuracy of operating characteristic identification.

2. The method of claim 1, further comprising:

alerting an operator of the system of an identified defect and/or an identified operating characteristic of the system.

3. The method of claim 1, wherein the instant frequency analysis method comprises:

receiving a short duration signal produced by the system;

filtering the short duration signal;

applying a Hilbert transform to the short duration signal;

extracting phase information from the result of the Hilbert transform step;

calculating a first derivative of the phase with respect to time and obtaining an instant frequency.

4. The method of claim 1, wherein the step of analyzing the acoustical data further comprises:

analyzing the instant frequency to identify a defect of the system and to identify the operating characteristic of the system; and

wherein the step of fusing the acoustical data further comprises;

fusing the instant frequency with the orthogonally-related data collected by at least the orthogonally-related detector to reduce the false indication rate.

5. The method of claim 1, wherein the system is selected from the group consisting of a piece of railroad rolling stock, a turbine, manufacturing equipment, and machining equipment.

6. The method of claim 1, wherein the orthogonally-related detector is a temperature detector.

7. The method of claim 1, further comprising:

predicting whether at least one of a defect and a malfunction will occur.

8. The method of claim 1, wherein the operating characteristic of the system includes defects in the system.