Data processing: measuring – calibrating – or testing – Measurement system in a specific environment – Mechanical measurement system
Reexamination Certificate
2000-10-30
2002-12-17
Hilten, John S. (Department: 2863)
Data processing: measuring, calibrating, or testing
Measurement system in a specific environment
Mechanical measurement system
C702S034000
Reexamination Certificate
active
06496782
ABSTRACT:
BACKGROUND
The invention relates generally to electric machine monitoring and more particularly to gear and bearing monitoring.
Locomotive traction systems include traction motors, gears, axles, wheel-sets, and bearings. The gears are often lubricated with oil, and the pinion gear is sometimes fitted onto the traction motor shaft using an interference or shrink fit. In some situations, the oil in a gear case also lubricates motor bearings through a passage from the gear case to the bearings. There are several failures that can result in gear problems, including, for example, cracks in gear teeth due to excessive loading and loss of lubrication resulting in gear teeth wear. Loss of lubrication and/or gear problems can lead to gear damage, slipping of pinion gear on the motor shaft, and/or damage to bearings from vibrations that result in motor failures, and ultimately road failures. Prevention of serious gear, bearing, motor and road failures through incipient failure detection would therefore be desirable.
Haynes et al., U.S. Pat. No. 4,965,513, describes a motor current signature analysis method for diagnosing motor operated devices such as motor-operated valves (MOVs). Frequency domain signal analysis techniques are applied to a conditioned motor current signal to identify various operating parameters of the motor-driven device from the motor current signature. Motor current noise is assumed to include the sum of all the mechanical load changes which refer back to the electric motor drive, and the changes are described as being separated on a frequency and amplitude basis such that the source of various changes in load, such as periodic gear mesh loading, friction events at frequencies corresponding to their origin, and other motor load varying characteristics of the device, can be identified. Motor current noise signatures are taken at different periods during the operating life of the device and compared to determine aging and wear or abnormal operating characteristics. The embodiment of Haynes et al. appears to assume a fixed frequency system with a signal-to-noise ratio that is high enough (that is, any interfering signals are low enough) for accurate signal detection. MOVs operate in highly-controlled environments with well-prescribed duty cycles and typically run steadily without experiencing operating conditions that generate confounding signals. More sophisticated processing techniques are desirable for general industrial environments and are particularly desirable for locomotive environments with varying speed and load conditions and non-uniform track-related signals.
BRIEF SUMMARY
Problems in the combination of the traction motor and wheel set of a locomotive which are related to gears (insufficient lubrication due to oil-loss, gear wear, or cracks, for example) and bearings (damaged bearings or loss of lubrication, for example) are reflected in the torque of the motor. Under substantially constant operating conditions (that is, the rotational direction does not change and neither the rotational speed nor the load deviates by more than about 5%), such problems lead to substantially periodic additive components of the torque which are not present for a healthy motor. The period of the additive components is determined by the rotational speed of the motor, the number of teeth on the gears, and/or characteristics of the bearings. According to one embodiment of the present invention, using spectral analysis, characteristic features in relevant frequency bands of machine rotational frequency, current, voltage, vibration or torque signals can be extracted and summarized in an indicator value for the level of wear.
REFERENCES:
patent: 4467895 (1984-08-01), Smith et al.
patent: 4965513 (1990-10-01), Haynes et al.
patent: 5510927 (1996-04-01), Noe
patent: 5566092 (1996-10-01), Wang et al.
patent: 5570256 (1996-10-01), Schoen et al.
patent: 5629870 (1997-05-01), Farag et al.
patent: 5644458 (1997-07-01), Schoen et al.
patent: 5680025 (1997-10-01), Bower, III et al.
patent: 5817936 (1998-10-01), Schricker
patent: 5943634 (1999-08-01), Piety et al.
patent: 6124692 (2000-09-01), Canada et al.
patent: 6199018 (2001-03-01), Quist et al.
L. Devroye, et al, “A Probabiliatic Theory of Pattern Recognition”, 1997, pp. 46-47.
B. Flury. et al, “Multiple Linear Regression”, Multivariate Statistics a Practical Approach, CH.5, pp. 55-75, 1988.
PJ Rousseeuw, “Robust Estimation and Identifying Outliers”, Handbook Of Statistical Methods for Engineers and Scientists, CH. 17, pp. 17.1-17.26.
R. Duda, et al, “Pattern Classification and Scene Analysis”, pp. 114-119, 1973.
US Patent Application Entitled “Gear Transmission Condition Monitoring Method and Apparatus”, Ser. No. 09/618,440, Filed Jul. 18, 2000, Attorney Docket RD-26,982 By Suresh Reddy, Et Al.
Chbat Nicolas Wadih
Claus Bernhard Erich Hermann
Kliman Gerald Burt
Agosti Ann M.
Breedlove Jill M.
General Electric Company
Hilten John S.
Sun Xiuquin
LandOfFree
Electric machine monitoring method and system does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Electric machine monitoring method and system, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Electric machine monitoring method and system will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2970781