Diagnostic system and method, especially for a valve

• US 20030019297A1
• Filed: 03/29/2002
• Published: 01/30/2003
• Est. Priority Date: 09/30/1999
• Status: Active Grant

• Current Assignee: Siemens AG
• Original Assignee: Siemens AG
• Inventors: Puettmer, Alf, Fiebelkorn, Klaus-Dieter, Klebert, Gerhard
• Granted Patent: US 6,637,267 B2
• Time in Patent Office: Days
• US Class Current: 73/587
• CPC Class Codes: F16K 37/0083 by measuring valve parameters

Abstract

The invention relates to a diagnostic system and method, especially for a valve (2) that can be actuated by a positioner (9) via a drive (6). The system utilizes a sensor (12) for recording structure-related sound and has a device (14) for evaluating a recorded measurement signal (13). A spectral region greater than 50 kHz is evaluated for conducting fault recognition. A fault indication signal is generated if the intensity of the measurement signal in this spectral region exceeds a defined threshold value. The sound caused by cavitation is primarily evaluated for conducting fault recognition.

Claims

1. A diagnostic system comprising:

a sensor operable to sense structure-borne noise; and

an evaluation unit operable to evaluate a recorded measurement signal,

wherein said evaluation unit is configured such that a spectral region of the measurement signal above a first limit frequency, where the first limit frequency is greater than 50 kHz, is evaluated for fault detection, and a fault indication signal is generated if an intensity of the measurement signal in the spectral region exceeds a threshold value.

2. A diagnostic system as claimed in claim 1, further comprising:

a positioner operable to actuate a valve; and

a drive unit operable to drive said positioner,

wherein the structure-borne noise sensed by said sensor is located at the valve.

3. A diagnostic system as claimed in claim 1, wherein said sensor comprises:

an active high-pass filter operable to reduce a component in the measurement signal, wherein the component is caused by low-frequency operating noise.

4. A diagnostic system as claimed in claim 1, wherein the sensor comprises a transducer element that is relatively insensitive to lower-frequency operating noise and relatively sensitive to higher-frequency noise.

5. A diagnostic system as claimed in claim 4, wherein said transducer element comprises a piezo-electric material.

6. A diagnostic system as claimed in claim 1, wherein said sensor comprises a housing operable to shield components of said sensor from electromagnetic and electrostatic fields.

7. A diagnostic system as claimed in claim 1, wherein said sensor comprises:

an active high-pass filter;

a mounting location; and

means for thermally decoupling said active high-pass filter from said mounting location.

8. A diagnostic system as claimed in claim 1, wherein said sensor comprises a threaded mounting stem.

9. A diagnostic system as claimed in claim 1, wherein said evaluation unit comprises a high-pass filter with an adjustable limit frequency.

10. A diagnostic system as claimed in claim 1, wherein said evaluation unit comprises a bandpass filter with adjustable upper and lower limit frequencies.

11. A diagnostic method comprising:

sensing structure-borne noise;

evaluating a recorded measurement signal for fault detection, wherein a spectral region of the measurement signal above a first limit frequency greater than 50 kHz is evaluated; and

generating a fault indication signal if the intensity of the measurement signal in the spectral region exceeds a threshold value.

12. A diagnostic method as claimed in claim 11, further comprising:

actuating a valve using a positioner via a drive, and wherein the sensed structure-borne noise is located at the valve.

13. A diagnostic method as claimed in claim 11, further comprising:

determining a strength of background noise in the measurement signal by measuring a signal strength in a spectral region above a third limit frequency, the third limit frequency being greater than a second limit frequency and selected such that influence on the measured value by cavitation noise is minimized;

determining a strength of the cavitation noise in the measurement signal by measuring the signal strength in a spectral region between the second limit frequency and the third limit frequency;

determining the threshold value based on the strength of the background noise;

and

generating the fault indication signal if the strength of the cavitation noise exceeds the threshold value.

14. A diagnostic method as claimed in claim 13, further comprising:

determining a ratio relative to the strength of the cavitation noise and the strength of the background noise, wherein the strength of the cavitation noise and the strength of the background noise correspond to a fault-free state of a valve; and

modifying the threshold value to be higher than the strength of the background noise multiplied by the determined ratio.

15. A diagnostic method as claimed in claim 14, wherein the valve in said determining step is a new valve.

16. A diagnostic method as claimed in claim 14, further comprising:

determining a trend indicative of when the threshold is likely to be exceeded, wherein the trend is based on successive diagnostic measurements.

17. A method for determining a fault in a valve through which a gaseous or a liquid material flows, the method comprising:

measuring an acoustic signal generated in a vicinity of the valve as the gaseous or liquid material flows through the valve;

separating the acoustic signal into a low frequency portion and a high frequency portion; and

determining a likelihood value representing a likelihood that the valve has a fault, wherein the likelihood value is based on the high frequency portion of the acoustic signal.

18. A method as claimed in claim 17, wherein the high frequency portion consists essentially of signals having a frequency above approximately 50 kHz.

19. A method as claimed in claim 18, wherein the high frequency portion consists essentially of signals having a frequency above approximately 200 kHz.

20. A method as claimed in claim 18, wherein the high frequency portion consists essentially of signals having a frequency above approximately 500 kHz.

21. A method as claimed in claim 17, wherein said determining comprises using fuzzy logic or a neural network.

22. A method as claimed in claim 17, further comprising:

determining an estimated time after which the valve is more likely than not to exhibit a fault, wherein said determining of the likelihood value is based on a frequency spectrum of the acoustic signal.

23. A method as claimed in claim 17, further comprising:

determining a type of fault, wherein said type of fault is based on a frequency spectrum of the acoustic signal.