LOOSE PART MONITORING METHOD AND SYSTEM

• US 20100321195A1
• Filed: 04/28/2010
• Published: 12/23/2010
• Est. Priority Date: 11/01/2007
• Status: Active Grant

• Current Assignee: Shikoku Research Institute, Inc. (Shikoku Electric Power Company Incorporated)
• Original Assignee: Shikoku Research Institute, Inc. (Shikoku Electric Power Company Incorporated)
• Inventors: MORISAWA, TOSHIZUMI
• Granted Patent: US 8,446,287 B2
• Time in Patent Office: Days
• US Class Current: 340/679
• CPC Class Codes: G01F 1/8468 vibrating measuring conduits
• CPC Class Codes: G01H 1/00 Measuring characteristics of vibrations in solids by using direct conduction to the detector G01H9/00, G01H11/00 take precedence

Abstract

A loose part monitoring method and system for preventing the generation of a false alarm as much as possible, including analyzing a group of detection signals that have been output from a plurality of sensors placed on a wall defining a fluid flow path, wherein the group of detection signals are analyzed for a rising gradient that is related to change in intensity of impulsive sound, a damping time that is related to change of intensity of the impulsive sound, and a frequency spectrum that is related to pitch of the impulsive sound in order to determine whether the group of detection signals are false or true.

Claims

1. A method for monitoring a loose part in a fluid flow path using a plurality of sensors which are placed in predetermined positions on a wall defining the fluid flow path for detecting information on sound or vibration, said plurality of sensors being, configured to output a group of detection signals upon detection of the information, said method comprising:

a first false signal determination step comprising comparing a pattern of the order and difference in time of output of the group of detection signals from the plurality of sensors with reference patterns to determine whether or not the group of signals are false signals;

a second false signal determination step comprising comparing a rising gradient of one detection signal which is selected from the group of detection signals and which has been output early with that of another detection signal which is selected from the group of detection signals and which has been output late to determine whether or not the group of signals are false signals;

a third false signal determination step comprising obtaining a damping time of the waveform of one detection signal which is selected from the group of detection signals and determining whether or not the selected detection signal is derived from an impact and thereby determining whether or not the group of signals are false signals;

a fourth false signal determination step comprising comparing the frequency spectra of the group of detection signals to determine whether or not the group of signals are false signals; and

a fifth false signal determination step comprising comparing the rising gradients, damping times, and frequency spectra of the group of detection signals obtained in the second to fourth false signal determination steps with those of known detection signals determined as false signals in the past, and determining whether or not there are any detection signals in the group of detection signals that have the same properties as the known detection signals, thereby to determine whether or not the group of signals are false signals.

2. The method according to claim 1, further comprising triggering an alarm when it is determined that the group of detection signals are not false signals in any of the first to fifth false signal determination steps.

3. The method according to claim 1, wherein the group of detection signals are filtered to remove excessive noise prior to the first to fifth false signal determination steps.

4. The method according to claim 1, wherein the first to fourth false signal determination steps are carried out in any desired sequence.

5. The method according to claim 4, wherein, when it is determined that the group of detection signals are false signals in any of the first to fourth false signal determination steps, the succeeding step or steps are excluded.

6. A method for monitoring a loose part in a fluid flow path defined by a wall member using a plurality of sensors which are placed in predetermined positions on the wall member and which are configured to output detection signals upon detecting information on sound or vibration, comprising:

a first false signal determination step comprising receiving the detection signals from the plurality of sensors, acquiring a pattern of the order in which selected two of the detection signals have been output and the difference in time of output between the selected two detection signals, and comparing the acquired pattern with reference pattern records to determine whether or not the detection signals are false signals;

a second false signal determination step comprising receiving the detection signals from the plurality of sensors, acquiring a first rising gradient of first one of the detection signals, acquiring a second rising gradient of second one of the detection signals that arrives later than the first one, and comparing the first and second rising gradients to determine whether or not the detection signals are false signals;

a third false signal determination step comprising receiving the detection signals from the plurality of sensors, obtaining a damping time of selected one of the detection signals, and determining whether or not the obtained damping time falls within a predetermined range and thereby to determine whether or not the detection signals are false signals; and

a fourth false signal determination step comprising receiving the detection signals from the plurality of sensors, obtaining frequency spectra of at least two of the detection signals, and determining whether or not the obtained frequency spectra of the at least two detection signals match with each other and thereby to determine whether or not the detection signals are false signals.

7. The method according to claim 6, further comprising a fifth false signal determination step comprising comparing the rising gradients, the damping times, the frequency spectra and optionally amplitudes of the detection signals obtained in the second to fourth false signal determination steps with those of detection signals determined as false signals in the past to determine whether or not the detection signals are false signals.

8. The method according to claim 6, further comprising triggering an alarm when it is determined that the detection signals are not false signals in any of the first to fourth false signal determination steps.

9. The method according to claim 7, further comprising triggering an alarm when it is determined that the detection signals are not false signals in any of the first to fifth false signal determination steps.

10. The method according to claim 6, wherein the detection signals are filtered to remove excessive noise prior to the first to fifth false signal determination steps.

11. The method according to claim 6, wherein the first to fourth false signal determination steps are carried out in any desired sequence.

12. The method according to claim 11, wherein, when it is determined that the signals are false signals in any of the first to fourth false signal determination steps, the succeeding step or steps are excluded.

13. A loose part monitoring system for monitoring a loose part in a fluid flow path defined by a wall member, comprising:

a plurality of sensors placed in predetermined positions on the wall member and configured to output a group of detection signals upon detecting information on sound or vibration;

a pattern analyzing section configured to receive the group of detection signals from the plurality of sensors, to obtain a pattern of the order and difference in time of output of the group of detection signals from the plurality of sensors with reference patterns, to compare the acquired pattern with reference pattern data, and to determine whether or not the group of detection signals are false signals;

a waveform rising gradient analyzing section configured to receive the group of detection signals from the plurality of sensors and to exclude the group of detection signals when an impact waveform from a first sensor that has detected a signal first among the plurality of sensors has a smaller rising gradient than that of an impact waveform from a second sensor that detected a signal second among the plurality of sensors;

a damped vibration analyzing section configured to receive the group of detection signals from the plurality of sensors, to obtain a damping time of the waveform of one detection signal selected from the group of detection signals and to determine whether or not the selected detection signal is derived from an impact and thereby to determine whether or not the group of detection signals are false signals; and

a frequency spectrum comparing section configured to receive the group of detection signals from the plurality of sensors and to determine whether or not the group of detection signals have generally the same frequency spectrum and thereby to determine whether or not the group of detection signals are false signals.

14. The loose part monitoring system according to claim 13, further comprising a specific abnormal sound excluding section configured to receive rising gradients, damping times, and frequency spectra of the detection signals obtained in said waveform rising gradient analyzing section, damped vibration analyzing section and frequency spectrum comparing section, to compare the rising gradients, damping times, and frequency spectra of the group of detection signals obtained in said waveform rising gradient analyzing section, damped vibration analyzing section and frequency spectrum comparing section with those of known detection signals determined as false signals in the past and to determine whether or not there are any detection signals in the group of detection signals that have the same properties as the known detection signals and thereby to determine whether or not the detection signals are false signals.

15. The loose part monitoring system according to claim 13, further comprising loose part detectors coupled to the plurality of sensors, respectively, for removing excessive noise from the corresponding detection signals so that the detection signals with relatively less noise are fed to said waveform rising gradient analyzing section, damped vibration analyzing section and frequency spectrum comparing section.

16. The loose part monitoring system according to claim 15, further comprising a detection signal monitoring section coupled to the plurality of sensors for sensing noise signals from the sensors and optimally adjusting the setting conditions of the loose part detectors.

17. The loose part monitoring system according to claim 13, further comprising an alarm part coupled to said pattern analyzing section, waveform rising gradient analyzing section, damped vibration analyzing section and frequency spectrum comparing section for triggering an alarm when it is determined that the detection signals are not false signals in any of said pattern analyzing section, waveform rising gradient analyzing section, damped vibration analyzing section and frequency spectrum comparing section.