Rotating machine abnormality diagnosis device, rotating machine abnormality diagnosis method, and rotating machine

Patent metadata

Field	Value
Publication number	US-2017205276-A1
Application number	US-201515327831-A
Country	US
Kind code	A1
Filing date	Aug 18, 2015
Priority date	Sep 12, 2014
Publication date	Jul 20, 2017
Grant date	—

Abstract

A rotating machine abnormality diagnosing apparatus for diagnosing an abnormality in a rotating machine includes: an analyzing part which performs a time-frequency analysis to a signal S representing intensity and time relationship of an elastic wave generating in the rotating machine; an extracting part which extracts second data representing intensity and time relationship of a predetermined frequency component from first data obtained by the time-frequency analysis; and a determining part which determines that a first rotor and a second rotor are in contact with each other when the second data contains intensity variation having a cycle represented by T 1× b=T 2× a, in which a rotational cycle of the first rotor is denoted by T 1 , a rotational cycle of the second rotor is denoted by T 2 , and a ratio of a number of protrusions to a number of recesses is denoted by a:b.

First claim

1 . A rotating machine abnormality diagnosing apparatus for diagnosing an abnormality inside a rotating machine which includes a first rotor having a plurality of protrusions, and a second rotor having a plurality of recesses whose number is different from that of the protrusions and arranged against the first rotor at a predetermined gap so that corresponding protrusion and recess among the plurality of protrusions and recesses meet one another sequentially in accordance with rotation of the first rotor and the second rotor, the rotating machine abnormality diagnosing apparatus, comprising:
a sensor which detects an elastic wave generating in the rotating machine, and outputs a signal representing intensity and time relationship of the elastic wave; an analyzing part which performs a time-frequency analysis to the signal output from the sensor; an extracting part which extracts second data representing intensity and time relationship of a predetermined frequency component from first data obtained by the time-frequency analysis of the analyzing part; and a determining part which determines that the first rotor and the second rotor are in contact with each other when the second data contains intensity variation having a cycle represented by T 1 ×b=T 2 ×a, in which a rotational cycle of the first rotor is denoted by T 1 , a rotational cycle of the second rotor is denoted by T 2 , and a ratio of the number of the protrusions and the number of the recesses is denoted by a:b.

2 . A rotating machine abnormality diagnosing apparatus according to claim 1 , wherein
the rotating machine further includes a casing for accommodating the first rotor and the second rotor, and the determining part determines that the first rotor and the casing are in contact with each other when the second data contains intensity variation having a cycle represented by T 1 , and determines that the second rotor and the casing are in contact with each other when the second data contains intensity variation having a cycle represented by T 2 .

3 . A rotating machine abnormality diagnosing apparatus according to claim 2 , wherein
the determining part determines whether or not the first rotor and the second rotor are in contact with each other, after determining whether or not the first rotor and the casing are in contact with each other, and whether or not the second rotor and the casing are in contact with each other.

4 . A rotating machine abnormality diagnosing method for diagnosing an abnormality inside a rotating machine which includes a first rotor having a plurality of protrusions, and a second rotor having a plurality of recesses whose number is different from that of the protrusions and arranged against the first rotor at a predetermined gap so that corresponding protrusion and recess among the plurality of protrusions and recesses meet one another sequentially in accordance with rotation of the first rotor and the second rotor, the rotating machine abnormality diagnosing method, comprising:
a detecting step of detecting an elastic wave generating in the rotating machine, and outputting a signal representing intensity and time relationship of the elastic wave; an analyzing step of performing a time-frequency analysis to the signal output in the detecting step; an extracting step of extracting second data representing intensity and time relationship of a predetermined frequency component from first data obtained by the time-frequency analysis in the analyzing step; and a determining step of determining that the first rotor and the second rotor are in contact with each other when the second data contains intensity variation having a cycle represented by T 1 ×b=T 2 ×a, in which a rotational cycle of the first rotor is denoted by T 1 , a rotational cycle of the second rotor is denoted by T 2 , and a ratio of the number of the protrusions and the recesses is denoted by a:b.

5 . A rotating machine including the rotating machine abnormality diagnosing apparatus according to claim 1 .

Assignees

• Kobe Steel Ltd

Inventors

• Araki Kaname
• Takahashi Eiji
• Ashida Tsuyoshi
• Takaki Shugo
• Matsumoto Yasuharu

Classifications

• G01H17/00 Primary

  Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the other groups of this subclass · CPC title

• G01M13/00

  Testing of machine parts · CPC title

• G01M13/028 Primary

  Acoustic or vibration analysis · CPC title

• G01M99/00

  Subject matter not provided for in other groups of this subclass · CPC title