EP1920226A2 - Procede et dispositif pour reconnaitre des grincements - Google Patents

Procede et dispositif pour reconnaitre des grincements

Info

Publication number
EP1920226A2
EP1920226A2 EP06777653A EP06777653A EP1920226A2 EP 1920226 A2 EP1920226 A2 EP 1920226A2 EP 06777653 A EP06777653 A EP 06777653A EP 06777653 A EP06777653 A EP 06777653A EP 1920226 A2 EP1920226 A2 EP 1920226A2
Authority
EP
European Patent Office
Prior art keywords
signal
frequency
detected
squeaking
value
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
EP06777653A
Other languages
German (de)
English (en)
Inventor
Fawzi Attia
Michael Fischer
Winfried Keiper
Thomas Zoeller
Uwe Bretz
Mats Karlsson
Uwe Martin
Michael Raquet
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Robert Bosch GmbH
Original Assignee
Robert Bosch GmbH
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Robert Bosch GmbH filed Critical Robert Bosch GmbH
Publication of EP1920226A2 publication Critical patent/EP1920226A2/fr
Withdrawn legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
    • G01H3/00—Measuring characteristics of vibrations by using a detector in a fluid
    • G01H3/04—Frequency
    • G01H3/08—Analysing frequencies present in complex vibrations, e.g. comparing harmonics present
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M17/00—Testing of vehicles

Definitions

  • the invention relates to a method for detecting squeaking noise within a sound signal generated during operation of a motor vehicle or during operation of one of its components, in which the sound signal is detected at least two times the amplitude distribution in one
  • Frequency domain representing amplitude spectrum of the sound signal is generated and by evaluating the at least two amplitude spectra, the presence of a
  • An advantageous embodiment of the invention is characterized in that the squeaking noises are brake squeaking noises. Brake squeaks are particularly disturbing for the driver of a motor vehicle and the environment.
  • An advantageous embodiment of the invention is characterized in that each of the two amplitude spectra is determined by a frequency analysis of a temporal signal waveform, wherein the temporal signal waveform extends over a respective other temporal signal portion.
  • time-shifted signal analyzes are available and it is also possible to determine the length of time a squeaking noise.
  • An advantageous embodiment of the invention is characterized in that - the temporal signal waveform is sampled, and the temporal signal portion each extends over a predetermined number of sampling points.
  • An advantageous embodiment of the invention is characterized in that - the temporal signal waveform is sampled at a signal sampling frequency of 48 kHz and each signal portion comprises 4096 sampling points.
  • An advantageous embodiment of the invention is characterized in that the signal sections overlap, in particular the overlap is 50% of the temporal
  • An advantageous embodiment of the invention is characterized in that a smoothed amplitude spectrum is formed from each of the at least two amplitude spectra and the presence of a squeaking noise is detected by evaluating the at least two amplitude spectra and the at least two smoothed amplitude spectra.
  • An advantageous embodiment of the invention is characterized in that the smoothing by arithmetic averaging of the sound pressure level determined from the amplitude spectrum over predetermined frequency sections, wherein the frequency sections in particular have a length of 1 third.
  • An advantageous embodiment of the invention is characterized in that the presence of a squeaking noise is detected by evaluation of signal peaks in the amplitude spectra.
  • the property is exploited that expressing squeaks in signal peaks.
  • An advantageous embodiment of the invention is characterized in that for detecting the presence of a squeaking noise for a given minimum number of evaluations of temporally successive amplitude spectra signal peaks are detected within the same frequency interval. This rule out that randomly occurring individual signal peaks are erroneously detected as squeaking noise.
  • An advantageous embodiment of the invention is characterized in that a squeaking noise is detected as no longer present if, for a further predetermined minimum number of evaluations temporally immediately successive amplitude spectra in the same frequency interval no signal peak is detected.
  • An advantageous embodiment of the invention is characterized in that the entire frequency range to be examined is divided into overlapping frequency intervals.
  • the determined value of the signal peak by at least a predetermined value is greater than the value of the corresponding signal at the edges of the frequency interval.
  • Amplitude spectrum is formed and that a signal peak in a frequency interval is then detected as present, if in addition the determined value of the signal peak in this frequency interval is greater than the determined from the smoothed amplitude spectrum value of the signal at the same
  • Frequency of the frequency interval is and the determined value of the signal peak by at least a predetermined value is greater than the determined from the smoothed amplitude spectrum value of the signal at the edges of the frequency interval.
  • the average deviation of the frequency of the detected signal peaks from the center frequency of the
  • Presence of a squeaking noise is closed when the deviation quantity exceeds a predefinable threshold value in such a direction that increases the deviation associated with the deviation amount.
  • spurious signals have a lower standard deviation around their center frequency than brake generated squeaking events.
  • An advantageous embodiment of the invention is characterized in that the presence of a squeaking noise is concluded when the deviation quantity exceeds a predefinable threshold value in such a direction that the deviation associated with the deviation quantity is reduced.
  • the main squeal frequency is determined.
  • the invention comprises a device containing means for carrying out the method according to the invention.
  • FIG. 1 shows an example spectrum Xi and an associated smoothed spectrum Si.
  • FIG. 2 shows the principle of the search for squeaking events in a frequency-time domain. Signal peaks are marked as black squares.
  • Fig. 4 shows the basic sequence of the method according to the invention.
  • the spectrogram is a representation of the frequency content of the signal x (n) as a function of time.
  • N 4096 sampling points
  • the signal sections each overlap by 50%, i. by 2048 sample points, -
  • Each signal segment is multiplied by a Hanning window of length N.
  • a smoothed spectrogram S (f, n) is further calculated according to the above method.
  • spectra Xi (f) are calculated individually from the signal sections xi and thereby smoothed, that the amplitude values within certain frequency bands are replaced by their mean value.
  • Si (f) smoothed amplitude spectra
  • Fig. 1 is an example spectrum Xi (f) (jagged line) and an associated smoothed
  • the signal is searched for signal peaks using the signal peak detection method described below before searching for squeaking noises.
  • Broadband sounds are also characterized in that signal peaks are detected in a frequency-time range by this signal peak detection method. The extent of this frequency-time range is set according to practical requirements. It is also defined how many signal peaks must be detected by this signal peak detection method per frequency-time range, so that a broadband interference signal and thus no squeaking is detected as present.
  • Frequency-time ranges not already excluded by the search for noises are examined for signal peaks by the signal peak detection method described later.
  • Deviations from the stated values are also possible in particular for C and w.
  • C the value range from 7 dB to 17 dB has proven to be suitable; for w, the value range from 0.025 to 0.175 is suitable.
  • the information about found signal peaks at each time point n is available for each frequency fc in the considered area.
  • coherent frequency-time ranges are searched, which represent squeaking noises. For this purpose, for each frequency fc for which a signal peak has been found, all times n are considered and determined, whether to successive
  • Time points j and j + 1 a signal peak was detected by the signal peak detection method.
  • a minimum number k of immediately temporally successive signal peaks in a frequency range around fc must exist in order to be recognized as a squeaking noise. If there are at least k time-related signal peaks, during the further search for the frequency fc along the times n, temporal gaps of the maximum duration 1 between the signal peaks are also allowed and attributed to the duration of the squeaking sound found. This procedure is based on the fact that for the perception of a sound by humans this - must be present for a minimum period and on the other hand, temporal masking effects, which are shorter than a certain duration, are barely influenced in their perception by humans.
  • time gap between two signal peaks exceeds the duration 1, then this duration is no longer attributed to the total length of a squeaking noise. If the time gap between signal peaks or temporally related sequences of signal peaks exceeds a duration gmax, then the sequences of signal peaks found are treated no more than one but two squeak events. The start and end times of each detected squeak event q are stored as tqstart and tqend, respectively.
  • FIG. 2 shows the principle of searching for squeaking events in a frequency-time domain.
  • the signal peaks found by the signal peak detection method are marked as black squares.
  • the time t is plotted in the abscissa direction in FIG. 2 and the frequency f is plotted in the ordinate direction.
  • two squeaking events are identified with "squeaks 1" and
  • Each of the two squeaking events comprises 7 detected signal peaks, with "squeaking 1" there is a time gap between the third and fourth squeaking event, which, however, is not greater than the maximum permitted time duration 1. Between the two squealing events there is a time gap whose duration exceeds the value gmax. That's why there are two separate ones
  • This frequency is defined as the actual frequency fq of the squeaking event.
  • first all spectra Xi which lie within the times tqstart and tqend are combined to form a new averaged spectrum Xqavg, by replacing in each case all the amplitude values of a frequency f by their arithmetic mean.
  • the squeal frequency fq is the frequency within the window Fc, for which Xqavg has the maximum level. This level Lq is stored as a representative value for the squeak event q.
  • signal peaks are found by evaluating the spectra Xi and Si and a defined, frequency-dependent threshold Th (f).
  • Th the sound pressure level in dB can be used.
  • the search for peaks may be limited to frequencies between a lower limit frequency fGl and an upper limit frequency fGu.
  • Each spectrum Xi is divided into frequency windows Fc by center frequencies fc.
  • the value w 0.0276 has proven to be particularly suitable in experiments.
  • Successive windows Fc and Fc + 1 may overlap depending on the choice of center frequencies.
  • the maximum value of the spectrum Xi is now searched for all frequencies f with fcl ⁇ f ⁇ fcu within Fc. The detected maximum value is denoted Lmax.
  • Lmax is at least C higher than the value of the unsmoothed spectrum at boundaries fcl and fcu of the considered range (conditions 2) and 3)). This clearly shows that the maximum is sufficiently localized and pronounced have to be. This excludes the possibility that very loud noises, the spectrum of which has a very high value for all frequencies, are misleading
  • the windows are so narrow that only a squeak can ever fall into a window. For example, at a frequency of 800 Hz, the window size is approximately 90 Hz.
  • the first harmonic of a falling in the window is approximately 90 Hz.
  • Squeaking noise is at 1600 Hz and thus far out of the window.
  • Xi (f) that is the unsmoothed amplitude spectrum of the signal excerpt xi, - Si (f), that is the smoothed amplitude spectrum of the signal excerpt xi, and the threshold value Th assumed here as frequency-independent over the frequency f shown in the abscissa direction.
  • the edge points fcl and fcu of the frequency window are shown on the abscissa axis.
  • FIG. 3 under the abscissa axis, 4 further frequency windows labeled a, b, c and d are shown by way of example. This makes it clear that the frequency axis can be divided into several frequency windows, wherein the frequency windows overlap in particular.
  • squeaking frequencies In practice, it often comes to the simultaneous excitation of several squeaking frequencies during a braking operation. To select one of these frequencies as the main frequency, an approach can be used which is based on the sound perception of humans. For this purpose, the associated specific loudnesses or volume levels are calculated and compared for all squeaking frequencies. The frequency of the squeak event having the highest specific loudness is selected as the main frequency. The strong correlation between the loudness and the subjectively perceived annoyance of a sound proves in listening tests that the most annoying squeaking event for the listener is evaluated in one signal as representative of the overall signal.
  • the following procedure can be used.
  • the specific loudnesses of all squeaking events are summed up within a frequency window Fmain.
  • Fmain glides over all frequencies between FGl and FGu. This results in a representation of the summed specific loudnesses of all occurring squeaking events as a function of the frequency.
  • peak values are now sought in a sliding window Fmov, which has a larger bandwidth than Fmain. Starting with the highest peak values found, the corresponding frequencies result in individual main frequencies.
  • FIG. 4 The basic sequence of the method according to the invention is shown in FIG. 4.
  • the sound signal is detected in block 401 and determined in block 402 whose amplitude spectrum.
  • feedback is made to block 401, i. the sound signal is detected again at a later time.
  • the presence of a squeaking noise is detected by evaluating the at least two amplitude spectra.
  • the method ends in block 404.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)

Abstract

La présente invention concerne un procédé pour reconnaître des grincements dans un signal acoustique produit lors du fonctionnement d'un véhicule automobile ou de l'un de ses composants. Dans le procédé de l'invention, le signal acoustique est détecté, un spectre d'amplitude du signal acoustique, qui représente la répartition d'amplitude sur une plage de fréquences, est produit en au moins deux instants, et la présence d'un grincement est reconnue par évaluation des deux spectres d'amplitude ou plus.
EP06777653A 2005-08-25 2006-07-07 Procede et dispositif pour reconnaitre des grincements Withdrawn EP1920226A2 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
DE102005040192A DE102005040192A1 (de) 2005-08-25 2005-08-25 Verfahren und Vorrichtung zur Erkennung von Quietschgeräuschen
PCT/EP2006/064023 WO2007023021A2 (fr) 2005-08-25 2006-07-07 Procede et dispositif pour reconnaitre des grincements

Publications (1)

Publication Number Publication Date
EP1920226A2 true EP1920226A2 (fr) 2008-05-14

Family

ID=37650632

Family Applications (1)

Application Number Title Priority Date Filing Date
EP06777653A Withdrawn EP1920226A2 (fr) 2005-08-25 2006-07-07 Procede et dispositif pour reconnaitre des grincements

Country Status (5)

Country Link
US (1) US8028583B2 (fr)
EP (1) EP1920226A2 (fr)
KR (1) KR101113383B1 (fr)
DE (1) DE102005040192A1 (fr)
WO (1) WO2007023021A2 (fr)

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102005052630A1 (de) * 2005-08-25 2007-03-01 Robert Bosch Gmbh Verfahren und Vorrichtung zur Bewertung der Lästigkeit von Quietschgeräuschen
DE102005040192A1 (de) 2005-08-25 2007-03-01 Robert Bosch Gmbh Verfahren und Vorrichtung zur Erkennung von Quietschgeräuschen
DE102008052818B4 (de) * 2008-10-13 2012-07-19 Dr. Ing. H.C. F. Porsche Aktiengesellschaft Messvorrichtung
CN103780738B (zh) * 2012-10-17 2017-08-29 腾讯科技（深圳）有限公司 移动终端图像处理方法及移动终端
JP6089948B2 (ja) * 2013-05-14 2017-03-08 日産自動車株式会社 車両の異音判定装置および異音判定方法
DE102014013472A1 (de) * 2013-09-30 2015-04-02 Hella Kgaa Hueck & Co. Verfahren zur Erkennung und Klassifikation von Schadensereignissen an Kraftfahrzeugen und Vorrichtung hierfür
DE102014007797B4 (de) * 2014-05-24 2022-05-12 Audi Ag Verfahren zur Erfassung eines Betriebsgeräuschs, insbesondere eines Bremsgeräuschs, und Kraftfahrzeug
US10832673B2 (en) 2018-07-13 2020-11-10 International Business Machines Corporation Smart speaker device with cognitive sound analysis and response
US10832672B2 (en) 2018-07-13 2020-11-10 International Business Machines Corporation Smart speaker system with cognitive sound analysis and response
KR102071259B1 (ko) 2018-09-05 2020-01-30 엔브이에이치코리아(주) 시편 단위에서의 스퀵 소음 데이터 측정을 위한 장치
CN109297718B (zh) * 2018-09-29 2020-08-07 重庆长安汽车股份有限公司 一种阶次啸叫噪声的评价方法
CA3105646C (fr) 2020-01-31 2026-02-03 Lululemon Athletica Canada Inc. Attache ayant une entretoise de tissu integree
CN112348335B (zh) * 2020-10-26 2024-11-19 安徽江淮汽车集团股份有限公司 变速器声响测试评价方法、装置、设备及存储介质
EP4361582B1 (fr) * 2022-10-24 2025-02-26 Wilo Se Procédé d'inspection de l'état d'un groupe motopompe ainsi qu'application logicielle, support de stockage et appareil d'inspection pour la mise en oeuvre du procédé

Family Cites Families (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3948345A (en) * 1973-06-15 1976-04-06 Allan Rosencwaig Methods and means for analyzing substances
US3892133A (en) * 1973-09-13 1975-07-01 Us Transport Statistical sound level analyzer
JPS59178357A (ja) * 1983-03-29 1984-10-09 Mitsubishi Electric Corp 歯車の異常音検出装置
US5058434A (en) * 1990-02-27 1991-10-22 Carl Schenck Ag Process for early detection of damage to machine parts
US5586028A (en) * 1993-12-07 1996-12-17 Honda Giken Kogyo Kabushiki Kaisha Road surface condition-detecting system and anti-lock brake system employing same
JPH07243906A (ja) 1994-03-08 1995-09-19 Bridgestone Corp 音源・振動源の寄与診断方法およびその装置
JP3159596B2 (ja) * 1994-03-22 2001-04-23 本田技研工業株式会社 ハイドロプレーニング現象検出装置
KR100201492B1 (ko) * 1996-07-26 1999-06-15 류정열 디퍼렌셜 기어 조립체 소음 측정장치
US5852243A (en) * 1997-07-21 1998-12-22 J-Squared, Llc Method and apparatus for detecting a road pavement surface condition
DE19749372A1 (de) 1997-11-07 1999-05-12 Volkswagen Ag Elektronisches Erkennungssystem und -verfahren für akustische Signale
US6332362B1 (en) * 2000-04-18 2001-12-25 Lg Electronics Inc. Device and method for detecting anomaly of air conditioner by using acoustic emission method
US6439053B1 (en) * 2000-09-13 2002-08-27 Henry Alan Bobulski Acoustic spectrometer apparatus and method for cavity geometry verification
DE10133987C2 (de) 2001-07-17 2003-09-04 Volkswagen Ag Verfahren und Vorrichtung zur Erkennung von Quietschgeräuschen einer Fahrzeugbremse durch Erkennung von Signalspitzen in Messreihen im Zeit- oder Frequenzbereich
GB0126070D0 (en) 2001-10-31 2001-12-19 Ncode Internat Ltd Identification of noise sources
DE10260723B4 (de) 2002-08-22 2010-09-16 Robert Bosch Gmbh Verfahren und Vorrichtung zur Testansteuerung
DE10250739A1 (de) 2002-10-31 2004-05-13 Bayerische Motoren Werke Ag Verfahren zur Bewertung von Störgeräuschen
JP4046059B2 (ja) * 2002-11-08 2008-02-13 株式会社豊田中央研究所 路面状態推定装置
DE10259979A1 (de) 2002-12-19 2004-07-15 Daimlerchrysler Ag Verfahren zur Ermittlung eines Straßenzustands während des Fahrbetriebs eines Kraffahrzeugs
JP4200796B2 (ja) * 2003-03-24 2008-12-24 株式会社アドヴィックス ブレーキノイズ検出装置
US6822929B1 (en) * 2003-06-25 2004-11-23 Sandia Corporation Micro acoustic spectrum analyzer
US6925879B2 (en) * 2003-09-30 2005-08-09 Spx Corporation Vibration analyzer and method
ITFI20030254A1 (it) * 2003-10-08 2005-04-09 Actis Active Sensors S R L Metodo e dispositivo perfezionati per l'analisi spettrale
US7124637B2 (en) * 2004-03-22 2006-10-24 Johnson Controls Technology Company Determining amplitude limits for vibration spectra
WO2006043511A1 (fr) * 2004-10-18 2006-04-27 Nsk Ltd. Systeme de diagnostic d’anomalie pour une machinerie
DE102005040192A1 (de) 2005-08-25 2007-03-01 Robert Bosch Gmbh Verfahren und Vorrichtung zur Erkennung von Quietschgeräuschen

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See references of WO2007023021A3 *

Also Published As

Publication number Publication date
US20090019936A1 (en) 2009-01-22
WO2007023021A3 (fr) 2007-07-26
KR20080036113A (ko) 2008-04-24
US8028583B2 (en) 2011-10-04
DE102005040192A1 (de) 2007-03-01
WO2007023021A2 (fr) 2007-03-01
KR101113383B1 (ko) 2012-03-05

Similar Documents

Publication Publication Date Title
EP1920226A2 (fr) Procede et dispositif pour reconnaitre des grincements
DE602004001241T2 (de) Vorrichtung zur Unterdrückung von impulsartigen Windgeräuschen
EP1920230A1 (fr) Procede et dispositif pour evaluer le degre de nuisance de bruits de type grincement(s)
DE102014013472A1 (de) Verfahren zur Erkennung und Klassifikation von Schadensereignissen an Kraftfahrzeugen und Vorrichtung hierfür
DE102008044088A1 (de) Verfahren zur dynamischen Ermittlung des Rauschlevels
DE102011055523A1 (de) Verfahren und Vorrichtung zur Schwingungsanalyse von Anlagen oder Maschinenelementen
DE69803202T2 (de) Verfahren und vorrichtung zur sprachdetektion
EP1563266B1 (fr) Procede pour evaluer des bruits parasites
DE102020126043A1 (de) Verfahren zur Zustandserkennung einer Antriebskette oder eines Antriebsriemens an einer industriellen Förderanlage und Ketten- oder Riemenspannvorichtung dafür
EP1431728A1 (fr) Procédé pour déterminer l'état d'une route lors du passage d'un véhicule
DE10157535B4 (de) Verfahren und Vorrichtung zur Reduzierung zufälliger, kontinuierlicher, instationärer Störungen in Audiosignalen
Chandrika et al. Development of an algorithm for automatic detection and rating of squeak and rattle events
DE102010020023A1 (de) Verfahren und Vorrichtung zur Rauschoptimierung eines Echosignals zur Fahrzeugumfelderfassung sowie zugehörige Vorrichtung zur Fahrzeugumfelderfassung
DE69132148T2 (de) Vorrichtung zur Verarbeitung eines Signals
EP2031581A1 (fr) Procédé destiné à la reconnaissance d'un évènement acoustique dans un signal audio
EP3588498B1 (fr) Procédé de suppression d'une réverbération acoustique dans un signal audio
DE10133987C2 (de) Verfahren und Vorrichtung zur Erkennung von Quietschgeräuschen einer Fahrzeugbremse durch Erkennung von Signalspitzen in Messreihen im Zeit- oder Frequenzbereich
EP1462777A1 (fr) Procédure et dispositif pour l'analyse quantitative des bruits d'un moteur
DE102015120081B4 (de) Verbrennungslärm-Ausblendesteuervorrichtung und -Ausblendeverfahren
EP4202372B1 (fr) Procédé de filtrage d'un signal de capteur et dispositif de commande d'un actionneur par filtrage d'un signal de capteur
DE69802431T2 (de) Verfahren und vorrichtung zur verbesserung eines digitalen sprachsignals
DE19626329C2 (de) Verfahren zur Analyse von Schallereignissen
DE102005033865B4 (de) Verfahren und Vorrichtung zur Auswertung eines mit einem Rauschsignal behafteten Nutzsignals
DE3810068A1 (de) Verfahren zur erkennung von sprachsignalen
DE102008053226B4 (de) Verfahren zur Bewertung eines Körperschallsignals für ein Personenschutzsystem

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

17P Request for examination filed

Effective date: 20080325

AK Designated contracting states

Kind code of ref document: A2

Designated state(s): DE ES FR GB

DAX Request for extension of the european patent (deleted)
RBV Designated contracting states (corrected)

Designated state(s): DE ES FR GB

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE APPLICATION IS DEEMED TO BE WITHDRAWN

18D Application deemed to be withdrawn

Effective date: 20170201