WO2007090492A2 - Verfahren und einrichtung zur detektion einer impulsartigen mechanischen einwirkung auf ein anlagenteil - Google Patents
Verfahren und einrichtung zur detektion einer impulsartigen mechanischen einwirkung auf ein anlagenteil Download PDFInfo
- Publication number
- WO2007090492A2 WO2007090492A2 PCT/EP2007/000135 EP2007000135W WO2007090492A2 WO 2007090492 A2 WO2007090492 A2 WO 2007090492A2 EP 2007000135 W EP2007000135 W EP 2007000135W WO 2007090492 A2 WO2007090492 A2 WO 2007090492A2
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- time
- mean
- amounts
- determined
- deviation
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H1/00—Measuring characteristics of vibrations in solids by using direct conduction to the detector
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/02—Gearings; Transmission mechanisms
- G01M13/028—Acoustic or vibration analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
Definitions
- the invention relates to a method and to a device for detecting a pulse-like mechanical action on a plant part.
- DE 198 43 615 C2 it is proposed to carry out the diagnosis of the state of a combustion drive with the aid of an analysis of the frequency spectrum of the measurement signals, which are recorded by means of a sound pickup arranged in the air inlet or exhaust area.
- DE 197 27 114 C2 the structure-borne noise signals occurring to monitor a machine instead of the airborne sound are detected.
- the respectively determined frequency spectra of the measurement signals detected by the structure-borne sound absorber are analyzed.
- the frequency spectrum of the measuring signal detected by a monitoring sensor is also analyzed during operation of the machine and compared with a reference frequency spectrum.
- a sensor is arranged in the US 4,888,948 at the inlet of the turbine, with which an electrical charge induced by the foreign bodies is detected.
- measured values of a signal describing the operating state are continuously recorded and compared with adaptive threshold values.
- quantile values of the probability distribution of the measured values are recursively and adaptively recorded. From the number and height of the threshold value overruns, parameters for the state of the machine are determined.
- Such problems can occur, for example, in gas turbines, whose combustion chamber is lined with ceramic tiles to protect against overheating. These ceramic tiles are subjected to a high dynamic load due to pressure swing fluctuations occurring in the combustion chamber. It may happen that portions of the tiles break at the respective brackets, be entrained by the flow of the exhaust gas and strike the first row of guide blades of the gas turbine. This can lead to damage to the coating of the guide vanes and destruction of the blades arranged behind it.
- WO 03/071243 A1 discloses a method for detecting a pulse-like mechanical action on an installation part, in which the detected structure-borne sound signal is subjected to a windowed Fourier transformation. From a multiplicity of the Fourier spectra determined in this way, a valuation function K is derived with the algorithms explained in greater detail there, which indicates the occurrence of a pulse-like mechanical action on the part of the installation.
- the specified in this document algorithm for deriving the evaluation function K allows the precise detection of the noisy measurement signal superimposed, attributable to a pulse-like action signal component.
- An essential step in this proposed evaluation algorithm is that the deviation of the amount A of the Fourier transform from an absolute value A is determined for each time window and each of the predetermined frequencies.
- the formation of the mean value A is of crucial importance, since there may be random state changes in plant parts, in particular in a turbine, in which the plant part changes from one operating state to another and the operating or background noise very quickly to a much higher one Increase in level can.
- Such a state change in a turbine for example, the onset of a so-called humming. This is caused by the arrangement of the flames in an annulus, which may cause the entire combustion chamber to be excited to resonant vibrations, in particular, the vibration modes being preferred in the circumferential direction.
- the continuous background noise fluid flow, pump noise
- operational short-term operating noise is superimposed, as for example, by permissible changes in operating conditions and intentional interference with the operation (actuation of valves, driving control rods) are caused.
- the invention is based on the object of specifying a method for detecting a pulse-like mechanical action on a part of the plant, compared to that of WO 03/071243 known method is further improved.
- the invention has the object to provide a device for carrying out the method.
- the object is achieved according to the invention with a method having the features of claim 1.
- a) in time steps temporally successive time windows is with predetermined parameters of a mathematical transport the magnitude (a (fi, tj)) of the transform formation regulations of the measurement signal (M) determined, b) for each time step ( ⁇ t) and each of the predetermined parameters (fi), the deviation of the amount (A (fi, t j )) from an average value (A (fi, t)) is determined, c) from the for each deviations determined by the given parameters (fj .
- ⁇ (fi, t) (Q « (fi, t) + Q 1 - ⁇ (fi, t)) / 2
- the invention is based on the recognition that a simple numerical averaging of the amount A can lead to a rapid transition of the operating state into a state with a higher noise level as a burst signal, i. a signal due to a pulse-like action on the part of the plant is interpreted.
- a signal is shown in the diagram of FIG.
- an evaluation algorithm improved in this way is also unable to optimally adjust to rapid changes in the operating noise, ie the respectively valid mean value A (fi, t) and its standard deviation A (fi , t) of the operating noise with fast changes to determine exactly enough.
- the mean value A (fi, t) to adapt faster to the changed operating situation.
- the second object is achieved according to the invention with a device having the features of claim 6, the advantages of which correspond to the advantages specified in claim 1.
- FIG. 1 shows an embodiment of a device according to the invention in a schematic schematic diagram
- FIG. 2 shows a diagram in which the operating noise (measurement signal M) of a turbine detected by a measuring transducer is plotted against the time t,
- FIG. 3 shows a diagram in which the evaluation function K derived from the measurement signal M of FIG. 2 in accordance with an evaluation algorithm illustrated in WO 03/071243 is plotted against time, FIG.
- FIG. 5 shows a diagram in which the absolute value A determined from the measurement signal M by fast Fourier transformation Frequency spectrum is plotted against the time t at a given frequency fi,
- Fig. 6 is a graph in which the moving time average A of the amounts A of the spectrum at a given
- FIG. 7 shows a diagram in which the evaluation function K derived from the measurement signal of FIG. 2 with a method according to the invention is plotted against the time t, FIG.
- FIG. 8 shows a diagram in which a measurement signal M recorded on a turbine is plotted against the time t, over which a burst caused by an impact is superposed
- a plurality of transducers or sensors 4 are arranged on a plant part 2, for example a gas turbine, which continuously detect the operating noise present in the plant part 2 and propagating in the form of structure-borne noise in several measuring channels.
- the sensors 4 each convert the structure-borne sound signals into an electrical measurement signal, which is amplified in a preamplifier 6 and forwarded to an analog-to-digital converter 8, which is connected to a digital memory 10.
- the respective amplified measurement signal M is digitized, buffered and sent to a computer for further processing.
- tion 12 in which the evaluation algorithm according to the invention is implemented.
- the computing device 12 includes for each measurement channel a processor for a fast calculation of a transform of the data passed on from the analog / digital converter 8 and a ring memory for storing a number L of the transforms determined by the transformation.
- this transformation is a fast windowed Fourier transform FFT.
- FFT Fast windowed Fourier transform
- orthogonal basis functions may be formed by the exponential function e "1 COt , as is the case with the Fourier transform However , in principle, other periodic functions may also be used as the orthogonal basis system These transformations become discrete transforms with a predetermined set of discrete ones
- a weighting function K (t) dependent on the time t is determined for each measuring channel according to an algorithm implemented in the computing device 12 and explained in greater detail below, which is provided with a predetermined threshold value K in a comparison device 14 0 is compared.
- Exceeding the threshold K 0 is used as an indication of the presence of a caused by a transient mechanical action pulse-like signal component and generates a corresponding trigger signal S.
- the trigger signal S is fed to a transient recorder 16, in which the measurement data (measurement signals M 3 ) recorded for a time range of for example 10 s and to a Evaluation computer 18 are passed to be able to perform a subsequent analysis with this.
- FIG. 4 explains the first step of the procedure implemented in the arithmetic unit 12 (FIG. 1) on the basis of a schematic diagram.
- the measurement signal M digitized at a high clock rate (typically 80-100 kHz, associated time interval DT approximately 0.01-0.0125 ms, (enlarged in the figure)) (shown in the figure in an analogous manner in the figure for reasons of clarity) is used in each case for a time range
- the digitized measurement signal M is subjected to a fast discrete Fourier transformation in a time window ⁇ t within this time range TS.
- the time window ⁇ t is shifted by a time step ⁇ t and updated to ⁇
- time-dependent amounts A fi, t.
- Fig. 5 the time course of the amount A at a predetermined frequency fi is illustrated. It can be seen from this figure that the amount A varies with time at this predetermined frequency f i.
- the time interval entered in the FIGURE is the time step ⁇ t in which the fast Fourier transformation is updated.
- This mean value A (fi, t) is a time-sliding, which in time steps ⁇ t as a function of time t is derived from a data set A (fi, t) composed of M amounts A (fi, t m ) of a time segment T. with the help of the relationship
- Is ⁇ -1 by means of the OC and the (1- ⁇ ) ⁇ -Quantiis Q and Q are a mean deviation s (fj., T) of the magnitudes A (fi, t n) from the mean magnitude A (fi, t ) with the help of the relationship
- Q 1-01 is the (1- ⁇ ) quantile of the normalized Gaussian or normal distribution.
- the mean and deviation of a data set are thus calculated without taking into account the values which are outside the ranges defined by ⁇ and (1- ⁇ ). This means that significantly higher amounts, such as those that can be additionally exacerbated by a superimposed burst signal, are not taken into account and thus can not distort the result.
- the mean value A (fi, t) adapts to rapid changes in the operating environment, so that these do not adulterate the normalized ones Lead spectra or pretend false positives in the monitoring.
- the mean value A (fi, t) adapts to rapid changes in the operating environment, so that these do not adulterate the normalized ones Lead spectra or pretend false positives in the monitoring.
- an improved mean value A opt (fi / t) can now be determined in a further calculation step on the basis of the quantile method by the respective current data set A (fi, t) the amounts A (fi, t m) are eliminated, which is significantly greater than the above-calculated mean magnitude A (fi, t).
- M E is the number of amounts A (fi, t m ) greater than
- Mean value A is calculated according to the following equation.
- a (fi, t) or A opt (fi, t) as well as the mean deviation s (fi, t) or s opt (fi, t) are constantly updated with the help of M transformations. This update takes place in time steps ⁇ t. The calculation of the new mean value
- a (fi, t + ⁇ t) or A opt (fi / t + ⁇ t) and the new mean deviation s (fi, t + ⁇ t) or s opt (fi, t + ⁇ t) is the underlying data set A (t + ⁇ t) thereby formed by deleting the first (oldest) amount and adding the newest amount.
- the normalized deviation D (t, fi) is additionally averaged in a frequency range fi-L / fi-L + i, • • • fi + L surrounding the frequency fi, consisting of 2L + 1 frequencies and an average normalized frequency range Deviation D (t, fj . )
- a further improvement of the signal-to-background ratio is achieved if, in addition, a threshold D 0 is introduced and a normalized deviation D s (fi, t) with the aid of
- the weighting function may also be the difference between the root of the sum S (t) and a moving time average of that root
- K (t) or K (t) exceeds a threshold value K 0 (alarm threshold), which is between 1.5 and 2 for gas turbines, this is an indication of the impact of a loose part.
- Deviation D s obtained from the measurement signal M of Fig. 2 evaluation function K (t) is shown in Fig. 7. It is clear now to recognize that the buzzing of the turbine has no influence on the evaluation function K (t).
- FIG. 8 now shows a real measuring signal M (t) obtained at a turbine, to which a burst-caused burst is superimposed at a time t ⁇ 3.9 s, which can not be directly identified in the measuring signal M.
- FIG. 9 the evaluation function K derived from the measurement signal of FIG. 8 is shown in an enlarged time scale. It can be seen from the figure that the event taking place (the burst) is noticeable by a marked increase in the evaluation function K.
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Computational Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
- Investigating Strength Of Materials By Application Of Mechanical Stress (AREA)
- Indication And Recording Devices For Special Purposes And Tariff Metering Devices (AREA)
Abstract
Description
Claims
Priority Applications (6)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
BRPI0706269-9A BRPI0706269B1 (pt) | 2006-02-03 | 2007-01-10 | Método e dispositivo para detectar um efeito mecânico do tipo pulso sobre uma parte do sistema |
ES07702640T ES2383148T3 (es) | 2006-02-03 | 2007-01-10 | Procedimiento y dispositivo para la detección de un efecto mecánico a modo de impulso sobre una parte de una instalación |
EP07702640A EP1979724B1 (de) | 2006-02-03 | 2007-01-10 | Verfahren und einrichtung zur detektion einer impulsartigen mechanischen einwirkung auf ein anlagenteil |
AT07702640T ATE550635T1 (de) | 2006-02-03 | 2007-01-10 | Verfahren und einrichtung zur detektion einer impulsartigen mechanischen einwirkung auf ein anlagenteil |
US12/185,307 US7684951B2 (en) | 2006-02-03 | 2008-08-04 | Method and device for detecting a pulse-type mechanical effect on a system part |
NO20083633A NO339870B1 (no) | 2006-02-03 | 2008-08-22 | Fremgangsmåte og innretning for detektering av en impulslignende innvirkning på en anleggsdel |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
DE102006004947A DE102006004947B4 (de) | 2006-02-03 | 2006-02-03 | Verfahren und Einrichtung zur Detektion einer impulsartigen mechanischen Einwirkung auf ein Anlagenteil |
DE102006004947.0 | 2006-02-03 |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US12/185,307 Continuation US7684951B2 (en) | 2006-02-03 | 2008-08-04 | Method and device for detecting a pulse-type mechanical effect on a system part |
Publications (2)
Publication Number | Publication Date |
---|---|
WO2007090492A2 true WO2007090492A2 (de) | 2007-08-16 |
WO2007090492A3 WO2007090492A3 (de) | 2008-03-20 |
Family
ID=38231141
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/EP2007/000135 WO2007090492A2 (de) | 2006-02-03 | 2007-01-10 | Verfahren und einrichtung zur detektion einer impulsartigen mechanischen einwirkung auf ein anlagenteil |
Country Status (10)
Country | Link |
---|---|
US (1) | US7684951B2 (de) |
EP (1) | EP1979724B1 (de) |
KR (1) | KR101035469B1 (de) |
AT (1) | ATE550635T1 (de) |
BR (1) | BRPI0706269B1 (de) |
DE (1) | DE102006004947B4 (de) |
ES (1) | ES2383148T3 (de) |
NO (1) | NO339870B1 (de) |
TW (1) | TWI338126B (de) |
WO (1) | WO2007090492A2 (de) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6553970B2 (ja) * | 2015-07-24 | 2019-07-31 | Ntn株式会社 | 異常診断装置およびセンサ外れ検知方法 |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2003071243A1 (de) * | 2002-02-22 | 2003-08-28 | Framatome Anp Gmbh | Verfahren und einrichtung zur detektion einer impulsartigen mechanischen einwirkung auf ein anlagenteil |
Family Cites Families (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DD224934A1 (de) * | 1984-04-30 | 1985-07-17 | Maxhuette Unterwellenborn | Verfahren zum ermitteln charakteristischer maschinenzustandsaenderungen aus diagnosesignalen |
GB8707187D0 (en) | 1987-03-25 | 1987-04-29 | Hughes Ltd Stewart | Monitoring of foreign object in engines |
US5479826A (en) | 1994-06-17 | 1996-01-02 | Westinghouse Electric Corporation | Microwave system for monitoring turbine blade vibration |
DE19545008C5 (de) | 1995-12-02 | 2004-07-22 | Reilhofer Kg | Verfahren zur Überwachung von periodisch arbeitenden Maschinen |
DE19727114C2 (de) | 1997-06-26 | 2002-08-08 | Busch Dieter & Co Prueftech | Verfahren zur Ermittlung und Darstellung von Spektren für Schwingungssignale |
DE59903184D1 (de) | 1998-07-22 | 2002-11-28 | Friedmund Nagel | Vorrichtung und verfahren zur aktiven reduzierung der schallemission von strahltriebwerken und zu deren diagnose |
DE19843615C2 (de) | 1998-09-23 | 2001-02-01 | Friedmund Nagel | Vorrichtung und Verfahren zur Diagnose von Verbrennungsantrieben |
DE19857552A1 (de) | 1998-12-14 | 2000-06-15 | Rolls Royce Deutschland | Verfahren zum Erkennen eines Wellenbruches in einer Strömungskraftmaschine |
US6499350B1 (en) | 2000-04-04 | 2002-12-31 | Swantech, L.L.C. | Turbine engine foreign object damage detection system |
-
2006
- 2006-02-03 DE DE102006004947A patent/DE102006004947B4/de not_active Expired - Fee Related
-
2007
- 2007-01-10 TW TW096100911A patent/TWI338126B/zh active
- 2007-01-10 WO PCT/EP2007/000135 patent/WO2007090492A2/de active Application Filing
- 2007-01-10 KR KR1020087021599A patent/KR101035469B1/ko active IP Right Grant
- 2007-01-10 BR BRPI0706269-9A patent/BRPI0706269B1/pt active IP Right Grant
- 2007-01-10 ES ES07702640T patent/ES2383148T3/es active Active
- 2007-01-10 EP EP07702640A patent/EP1979724B1/de active Active
- 2007-01-10 AT AT07702640T patent/ATE550635T1/de active
-
2008
- 2008-08-04 US US12/185,307 patent/US7684951B2/en active Active
- 2008-08-22 NO NO20083633A patent/NO339870B1/no unknown
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2003071243A1 (de) * | 2002-02-22 | 2003-08-28 | Framatome Anp Gmbh | Verfahren und einrichtung zur detektion einer impulsartigen mechanischen einwirkung auf ein anlagenteil |
Also Published As
Publication number | Publication date |
---|---|
US7684951B2 (en) | 2010-03-23 |
DE102006004947A1 (de) | 2007-08-16 |
TW200734622A (en) | 2007-09-16 |
TWI338126B (en) | 2011-03-01 |
ATE550635T1 (de) | 2012-04-15 |
BRPI0706269B1 (pt) | 2018-03-27 |
WO2007090492A3 (de) | 2008-03-20 |
NO20083633L (no) | 2008-08-22 |
BRPI0706269A2 (pt) | 2011-03-22 |
NO339870B1 (no) | 2017-02-13 |
KR20080094092A (ko) | 2008-10-22 |
EP1979724A2 (de) | 2008-10-15 |
ES2383148T3 (es) | 2012-06-18 |
DE102006004947B4 (de) | 2007-12-27 |
KR101035469B1 (ko) | 2011-05-18 |
US20090048791A1 (en) | 2009-02-19 |
EP1979724B1 (de) | 2012-03-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP1922530B1 (de) | Verfahren und einrichtung zur detektion des ortes einer impulsartigen mechanischen einwirkung auf ein anlagenteil | |
EP1476733B1 (de) | Verfahren und einrichtung zur detektion einer impulsartigen mechanischen einwirkung auf ein anlagenteil | |
EP3234512B1 (de) | Durchflussmessanordnung nach dem differenzdruckmessprinzip | |
DE102012109393A1 (de) | Vorrichtung und Verfahren zur Bewertung von Schwingungen | |
DE102012220222A1 (de) | Vorrichtung und Verfahren zur Zustandsüberwachung eines Wälzlagers | |
EP2623949A1 (de) | Zustandsüberwachungsvorrichtung und Verfahren zur Zustandsüberwachung von rotierenden mechanischen Bauteilen | |
DE102011055523A1 (de) | Verfahren und Vorrichtung zur Schwingungsanalyse von Anlagen oder Maschinenelementen | |
EP3058430B1 (de) | Vorrichtung sowie verfahren zum erfassen des aktuellen schädigungszustandes einer maschine | |
EP3589843B1 (de) | Verfahren und vorrichtung zum bestimmen eines indikators für eine vorhersage einer instabilität in einem verdichter sowie verwendung | |
DE102010005525B4 (de) | Verfahren zur Zustandsüberwachung einer Maschine und Überwachungseinrichtung hierfür | |
EP0742372A1 (de) | Überwachungssystem zur Feststellung einer Kavitationsintensität | |
WO2011018289A1 (de) | Verfahren zur frühzeitigen schadenserkennung in einem kraftfahrzeuggetriebe | |
EP1979724B1 (de) | Verfahren und einrichtung zur detektion einer impulsartigen mechanischen einwirkung auf ein anlagenteil | |
DE102015201203A1 (de) | Verfahren zur Detektion von Kavitation | |
AT511773A1 (de) | Vorrichtung und verfahren zur detektion wenigstens eines periodisch auftretenden fehlers an einem gegenstand | |
CN115186707A (zh) | 一种简易齿轮箱故障检测方法 | |
EP2110649B1 (de) | Sensor für die Zustandsüberwachung von mechanischen Bauteilen | |
DE112017007461T5 (de) | Vibrationsdetektionsvorrichtung und Unregelmäßigkeitsbestimmungssystem | |
DE102016114846A1 (de) | Differenzdruckmessanordnung und Verfahren zum Erkennen von verstopften Wirkdruckleitungen | |
DE102014208681A1 (de) | Verfahren zur Überwachung eines Zustands eines in einem Fluid bewegten Rotorblatts |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2007702640 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1570/MUMNP/2008 Country of ref document: IN |
|
WWE | Wipo information: entry into national phase |
Ref document number: 12008501745 Country of ref document: PH |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1020087021599 Country of ref document: KR |
|
ENP | Entry into the national phase |
Ref document number: PI0706269 Country of ref document: BR Kind code of ref document: A2 Effective date: 20080627 |