System, Method, and Apparatus for Using the Sound Signature of a Device to Determine its Operability

• US 20090091441A1
• Filed: 10/08/2008
• Published: 04/09/2009
• Est. Priority Date: 10/09/2007
• Status: Active Grant

• Current Assignee: Schweitzer Engineering Laboratories, Incorporated
• Original Assignee: Schweitzer Engineering Laboratories, Incorporated
• Inventors: Schweitzer, Edmund O. III, Minteer, Timothy M., Whitehead, David E.
• Granted Patent: US 8,155,326 B2
• Time in Patent Office: Days
• US Class Current: 340/531
• CPC Class Codes: G01M 13/028 Acoustic or vibration analysis
• CPC Class Codes: G01N 29/14 using acoustic emission techniques echo of particles G01N29/046; measuring mechanical vibrations or acoustic waves in solids in general G01H1/00

Abstract

A system for monitoring the operation of one or more devices is provided. The system includes a microphone acoustically coupled to one the monitored device. An analog-to-digital-converter samples the microphone and a processor examines the resultant digital signal for the occurrence of an abnormal event.

Claims

1. A system for monitoring the operation a device for the occurrence of an abnormal event comprising:

i) a microphone acoustically coupled to the device, the microphone producing an analog signal corresponding to sound waves observed by the microphone;

ii) an analog-to-digital converter for receiving the analog signal and producing a digital signal corresponding to the analog signal;

iii) a processor for receiving the digital signal and, for determining whether the abnormal event has occurred.

2. The system of claim 1 further comprising an alarm contact coupled to the processor, the alarm contact for operation when the abnormal event is detected.

3. The system of claim 1 further comprising a network interface coupled to the processor, the network interface for generating a network message when the abnormal event is detected.

4. The system of claim 1 further comprising a first network interface coupled to the analog-to-digital converter and a network, and a second network interface coupled to the processor and the network, and wherein the processor receives the digital signal via the network.

5. The system of claim 1 further comprising a memory and wherein the processor

(1) records in the memory an ambient sound signature and

(2) records in the memory a normal event sound signature corresponding to an operation of the device.

6. The system of claim 5 further comprising a Human Machine Interface coupled to the processor, the Human Machine Interface providing instructions to a user enabling the recordation of the ambient sound signature and the normal event sound signature.

7. The system of claim 1 further comprising a network interface coupled to the processor, wherein the processor

(1) accesses a database describing another device installed proximate to the system for monitoring, and

(2) operates the network interface to generate a sequence of network messages for operating the another device.

8. The system of claim 5 wherein the processor compares the digital signal to the ambient sound signature using a pattern-matching algorithm to determine if the abnormal event has occurred.

9. The system of claim 8 wherein the processor further compares the digital signal to the normal event sound signature using a pattern matching algorithm to determine if the digital signal corresponds to the normal event sound signature.

10. The system of claim 9 wherein the processor logs a normal event in the memory if the digital signal corresponds to the normal event sound signature.

11. The system of claim 8 wherein the digital signal consists of a latest sample and a plurality of previously stored samples, all of the samples generated periodically by the analog-to-digital converter, and wherein the processor further captures an event window consisting of a plurality of samples if a condition is met.

12. The system of claim 11 wherein the processor compares the latest sample to a threshold, and the processor captures the event window if the latest sample exceeds the threshold.

13. The system of claim 11 wherein the processor compares the latest sample to a previously acquired sample, and the processor captures the event window if the latest sample exceeds the previously acquired sample by a threshold.

14. The system of claim 11 wherein the processor calculates a root-mean-square value based on the latest sample and the previously stored samples, and compares the root-mean-square value to a threshold, and the processor captures the event window if the root-mean-square value exceeds the threshold.

15. The system of claim 11 wherein the processor calculates a first frequency spectrum describing a plurality of frequency components present in the digital signal, and wherein the processor compares the plurality of frequency components of the first frequency spectrum to a plurality of frequency components of a second frequency spectrum, the second frequency spectrum calculated prior to the calculation of the first frequency spectrum, and wherein the processor captures the event window if a frequency component of the first frequency spectrum differs from a corresponding frequency component in the second frequency spectrum by a threshold.

16. The system of claim 11 wherein the processor calculates a first frequency spectrum describing a plurality of frequency components present in the digital signal, and wherein the processor compares the plurality of frequency components of the first frequency spectrum to a plurality of frequency components of a second frequency spectrum, the second frequency spectrum calculated based on the recorded ambient sound signature, and wherein the processor captures the event window if a frequency component of the first frequency spectrum differs from a corresponding frequency component in the second frequency spectrum by a threshold.

17. The system of claim 11 wherein the processor determines a correlation of the digital signal to the ambient sound signature and wherein the processor captures an event window if the correlation is less than a threshold.

18. A method operating on a processor for monitoring the operation of a device for the occurrence of an abnormal event comprising the steps of:

i) receiving a digital signal indicative of a real-time sound produced by the monitored device; and

ii) determining whether an abnormal event has occurred based on said digital signal.

19. The method of claim 18, further comprising the step of operating an alarm contact on detection of an abnormal event.

20. The method of claim 18, further comprising the step of generating a network message on detection of an abnormal event.

21. The method of claim 18, wherein said step of receiving includes receiving the digital time-domain signal via a network.

22. The method of claim 18 further comprising the step of recording an ambient sound signature in a memory coupled to said processor, said ambient sound signature indicative of a normal sound produced by the monitored device.

23. The method of claim 22 further comprising the steps of:

i) accessing a database describing other devices installed proximate to the monitored device;

ii) issuing network messages to said other devices, the network messages adapted to cause one or more operations of said other devices; and

iii) recording the sound signatures of said operations.

24. The method of claim 22 wherein said step of determining comprises pattern-matching the digital signal to the ambient sound signature.

25. The method of claim 22 wherein digital signal comprises a latest sample and a plurality of previously stored samples, and wherein the step of determining includes identifying an event window comprising a plurality of samples if a condition is met.

26. The method of claim 25 wherein the step of determining comprises comparing the latest sample to a threshold and capturing an event window if the latest sample exceeds the threshold.

27. The method of claim 25 wherein the step of determining comprises comparing the latest sample to a previously acquired sample, and capturing an event window if the latest sample exceeds the previously acquired sample by a threshold.

28. The method of claim 25 wherein the step of determining comprises calculating a root-mean-square value based on the latest sample and the previously stored samples and comparing the root-mean-square value to a threshold, and capturing an event window if the root-mean-square value exceeds the threshold.

29. The method of claim 25 wherein the step of determining comprises:

i) calculating a first frequency spectrum describing a plurality of frequency components present in the digital signal;

ii) comparing the plurality of frequency components of the first frequency spectrum to a plurality of frequency components of a second frequency spectrum calculated prior to the calculation of the first frequency spectrum; and

iii) capturing an event window if a frequency component of the first frequency spectrum differs from a corresponding frequency component of the second frequency spectrum by a threshold.

30. The method of claim 25 wherein the step of determining comprises:

i) calculating a first frequency spectrum describing a plurality of frequency components present in the digital signal;

ii) comparing the plurality of frequency components of the first frequency spectrum to a plurality of frequency components of a second frequency spectrum describing said ambient sound signature; and

iii) capturing an event window if a frequency component of the first frequency spectrum differs from a corresponding frequency component of the second frequency spectrum by a threshold.

31. The method of claim 25 wherein the step of determining comprises correlating the digital signal to the ambient sound signature and capturing an event window if the correlation is less than a threshold.