Abnormality diagnosing method for sound or vibration and abnormality diagnosing apparatus for sound or vibration

• US 7,933,742 B2
• Filed: 01/31/2007
• Issued: 04/26/2011
• Est. Priority Date: 02/01/2006
• Status: Active Grant

• Current Assignee: Jtekt Corporation (Toyota Motor Corporation), Zhong Zhang
• Original Assignee: Jtekt Corporation (Toyota Motor Corporation), Zhong Zhang
• Inventors: Zhang, Zhong, Uemura, Hiroshi, Ishii, Hideaki
• Primary Examiner(s): Charioui; Mohamed
• Assistant Examiner(s): Ngon; Ricky
• Application Number:
  US11/700,188
• Publication Number:
  US 20070176759A1
• Time in Patent Office: 1,546 Days
• Field of Search: 702/183, 702/33, 702/56, 702/66, 702/182, 702/185, 702189-195, 382/278, 735/93, 736/02, 736/59, 736/60, 381/71.1, 381/71.8
• US Class Current: 702/183
• CPC Class Codes: G01H 3/125 for representing acoustic field distribution using optical means G01H9/002; sonar systems for imaging G01S7/56, G01S15/89; acoustic holography G03H3/00

Abstract

A noise detection signal detected at an evaluation point is filtered by a low pass filter, so that an evaluation waveform signal is extracted. A real signal mother wavelet of complex type is derived from the extracted evaluation waveform signal. Using this mother wavelet, an abnormal waveform signal stored in advance is processed by wavelet transformation. Then, a correlation value is calculated between the abnormal waveform signal and the mother wavelet. The correlation value is compared with a predetermined criterion, so that the presence or absence of abnormality in a sound source or vibration source corresponding to each abnormal waveform signal is determined.

Claims

1. An abnormality diagnosing method for sound or vibration for analyzing sound or vibration that appears at an evaluation point set up in a target space containing a plurality of sound sources or vibration sources, and thereby diagnosing the presence or absence of abnormality occurring in the sound source or vibration source, comprising:

a preliminary step of extracting abnormal waveform signals individually from a detection result of sound or vibration acquired in advance at the evaluation point at the time of occurrence of abnormality in each of the plurality of sound sources or vibration sources;

a first step of extracting an evaluation waveform signal serving as an analysis target from a detection result of sound or vibration at the evaluation point;

a second step of performing Hanning window processing, Fourier transformation, normalization processing, and Hilbert transformation on the evaluation waveform signal extracted at said first step, and thereby deriving a real signal mother wavelet of complex type;

a third step of performing wavelet transformation on each of the abnormal waveform signals extracted at said preliminary step by using the real signal mother wavelet derived at said second step, and thereby calculating a correlation value of each abnormal waveform signal with the real signal mother wavelet;

a fourth step of comparing the correlation value calculated at said third step with a predetermined criterion, and thereby determining the presence or absence of abnormality in each of the plurality of sound sources or vibration sources; and

a fifth step of displaying the correlation values calculated by the third step on a same time axis on a display section,

wherein the real signal mother wavelet derived at the second step has a start point and an end point at zeros, and an average over the domain of the real signal mother wavelet is zero while the function is bounded, and

wherein the third step of performing the wavelet transformation includes obtaining an inner product between a transform function to which each of the abnormal waveform signals is applied and the real signal mother wavelet.

2. An abnormality diagnosing apparatus for sound or vibration for analyzing sound or vibration that appears at an evaluation point set up in a target space containing a plurality of sound sources or vibration sources, and thereby diagnosing the presence or absence of abnormality occurring in the sound source or vibration source, comprising:

a detecting section for sound or vibration arranged at the evaluation point;

a storage section for storing abnormal waveform signals individually extracted from a detection result of said detecting section at the time of occurrence of abnormality in each of the plurality of sound sources or vibration sources;

an extracting section for extracting an evaluation waveform signal serving as an analysis target from the detection result of said detecting section;

a deriving section for performing Hanning window processing, Fourier transformation, normalization processing, and Hilbert transformation on the evaluation waveform signal extracted by said extracting section, and thereby deriving a real signal mother wavelet of complex type;

a correlation calculating section for performing wavelet transformation on each of the abnormal waveform signals stored in said storage section by using the real signal mother wavelet derived by said deriving section, and thereby calculating a correlation value of each abnormal waveform signal with the real signal mother wavelet;

a determining section for comparing the calculation result of said correlation calculating section with a predetermined criterion, and thereby determining the presence or absence of abnormality in each of the plurality of sound sources or vibration sources; and

a display section for displaying one or both of the calculation result obtained by said correlation calculating section and the determination result obtained by said determining section,

wherein said display section displays on a same time axis the correlation values calculated by said correlation calculating section,

wherein the real signal mother wavelet derived by the deriving section has a start point and an end point at zeros, and an average over the domain of the real signal mother wavelet is zero while the function is bounded, and

wherein the correlation calculating section performs the wavelet transformation by obtaining an inner product between a transform function to which each of the abnormal waveform signals is applied and the real signal mother wavelet.

3. An abnormality diagnosing apparatus for sound or vibration for analyzing sound or vibration that appears at an evaluation point set up in a target space containing a plurality of sound sources or vibration sources, and thereby diagnosing the presence or absence of abnormality occurring in the sound source or vibration source, comprising:

a first storage section for storing abnormal waveform signals extracted individually from a detection result of a detecting section for sound or vibration arranged at the evaluation point at the time of occurrence of abnormality in each of the plurality of sound sources or vibration sources;

a second storage section for storing a real signal mother wavelet of complex type derived by Hanning window processing, Fourier transformation, normalization processing, and Hilbert transformation performed on the evaluation waveform signal extracted from the detection result of said detecting section;

a correlation calculating section for performing wavelet transformation on each of the abnormal waveform signals stored in said first storage section by using the real signal mother wavelet stored in said second storage section, and thereby calculating a correlation value of each abnormal waveform signal with the real signal mother wavelet;

a determining section for comparing the calculation result of said correlation calculating section with a predetermined criterion, and thereby determining the presence or absence of abnormality in each of the plurality of sound sources or vibration sources; and

a display section for displaying one or both of the calculation result obtained by said correlation calculating section and the determination result obtained by said determining section,

wherein said display section displays on a same time axis the correlation values calculated by said correlation calculating section

wherein the real signal mother wavelet has a start point and an end point at zeros, and an average over the domain of the real signal mother wavelet is zero while the function is bounded, and

wherein the correlation calculating section performs the wavelet transformation obtaining an inner product between a transform function to which each of the abnormal waveform signals is applied and the real signal mother wavelet.