CN105741850B - Utilize the method and device of audio signal diagnostic machinery equipment - Google Patents
Utilize the method and device of audio signal diagnostic machinery equipment Download PDFInfo
- Publication number
- CN105741850B CN105741850B CN201510977714.2A CN201510977714A CN105741850B CN 105741850 B CN105741850 B CN 105741850B CN 201510977714 A CN201510977714 A CN 201510977714A CN 105741850 B CN105741850 B CN 105741850B
- Authority
- CN
- China
- Prior art keywords
- score
- signal
- frequency
- audio signal
- energy 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
- G10L21/0264—Noise filtering characterised by the type of parameter measurement, e.g. correlation techniques, zero crossing techniques or predictive techniques
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
- G01N29/04—Analysing solids
- G01N29/09—Analysing solids by measuring mechanical or acoustic impedance
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H15/00—Measuring mechanical or acoustic impedance
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0224—Process history based detection method, e.g. whereby history implies the availability of large amounts of data
- G05B23/0227—Qualitative history assessment, whereby the type of data acted upon, e.g. waveforms, images or patterns, is not relevant, e.g. rule based assessment; if-then decisions
- G05B23/0232—Qualitative history assessment, whereby the type of data acted upon, e.g. waveforms, images or patterns, is not relevant, e.g. rule based assessment; if-then decisions based on qualitative trend analysis, e.g. system evolution
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/03—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters
- G10L25/18—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters the extracted parameters being spectral information of each sub-band
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/03—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters
- G10L25/21—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters the extracted parameters being power information
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2291/00—Indexing codes associated with group G01N29/00
- G01N2291/01—Indexing codes associated with the measuring variable
- G01N2291/018—Impedance
Abstract
A kind of diagnostic device for diagnostic machinery equipment state, include: judging unit, the judging unit accumulates audio signal received in multiple unit detection times using as judge data, and finds out by comparing the audio signal of accumulation and each frequency energy value of reference signal the first score, the by comparing total energy value of the audio signal of the accumulation and the total energy value of the reference signal that calculate come at least one of the second score calculated and the third score calculated based on the peak information of the audio signal of the accumulation;And state output unit, the state output unit export the state of the mechanical equipment based at least one of first score, second score calculated by the judging unit and the third score.
Description
Technical field
The present invention relates to a kind of method and devices using audio signal diagnostic machinery equipment.
Background technique
Diagnostic system is used for the state of diagnostic machinery equipment.Most of diagnostic system is by generating in checkout and diagnosis position
The state that the trembling that frequency band is 10~1khz is vibrated to diagnose tool equipment.However, existing when actually mechanical equipment drives
Under Run-Time environment, all components of mechanical equipment are in linking state on the whole, thus according to carry object, operating process,
Or joint and linearly operating, the vibration data for diagnosing position will receive influence.Thus, in the vibration data detected when driving
Vibration data not only comprising diagnosis position also includes the vibration data at other positions, it may be said that the reliability of data is very low.
Therefore, after the diagnostic system based on vibration needs to stop mechanical equipment, only starting diagnosis position is to detect vibration.
As described above, based on the diagnostic system of vibration in order to reduce diagnosis position except movement or environmental factor shadow
It rings, it is necessary to stop mechanical equipment, so being difficult to accurately be diagnosed to be the practical problem occurred when driving.Therefore, it is necessary to a kind of nothings
It need to stop mechanical equipment, at any driving condition the accurately method of diagnostic machinery equipment.
Summary of the invention
Technical problem
The technical problem to be solved in the present invention is to provide a kind of method and devices using audio signal diagnostic machinery equipment.
Technical solution
The diagnostic device for diagnostic machinery equipment state of an exemplary implementation scheme according to the present invention includes:
Judging unit, the judging unit accumulate audio signal received in multiple unit detection times using as judging data, and
The first score calculated is found out by comparing each frequency energy value of the audio signal and reference signal accumulated, by comparing institute
The total energy value of the audio signal of accumulation and the total energy value of the reference signal are stated the second score for calculating and based on described
The peak information of the audio signal of accumulation is come at least one of the third score that calculates;And state output unit, the shape
State output unit is based in first score, second score calculated by the judging unit and the third score
At least one export the state of the mechanical equipment.
The diagnostic device also may include: audio signal extraction unit, the audio signal extraction unit per unit detection
Time receives audio signal detected by least one microphone mounted in the mechanical equipment, and from received audio signal
The judging unit will be sent after sound attenuation to.
The judging unit leads in the case where being entered any audio signal detected in arbitrary unit detection time
Any audio signal and the reference signal are crossed to judge whether the mechanical equipment is driving, if the machinery
Equipment is driving, so that it may which any audio signal is determined as the judgement data.
The judging unit calculates the difference of each frequency energy of any audio signal and the reference signal, if with
The reference signal frequency few compared to energy value is reference value hereinafter, then may determine that as in the arbitrary unit detection time
The mechanical equipment is driving.
The judging unit when mechanical equipment stopping, can abandoning described appoint in the arbitrary unit detection time
Meaning audio signal.
The judging unit extracts the time-based energy value Trendline of each frequency based on the audio signal of the accumulation,
And the representative energy value of corresponding frequencies is extracted in each frequency trend line, it is extracted using the representative energy value of each frequency aobvious
Show the first curve graph of the energy value based on frequency of the audio signal of the accumulation, based on first curve graph and display institute
State the energy value based on frequency of reference signal the second curve graph can find out first score, second score and
At least one of described third score.
The judging unit extracts the time-based energy Distribution value of each frequency based on the audio signal of the accumulation, and
Time shaft is divided into certain section, energy value Trendline can be calculated by section.
The judging unit can calculate the energy value trend by the mode of density data in the energy Distribution value
Line.
The judging unit receives the first audio signal in the first unit detection time, and finds out by comparing described the
Each frequency energy value of one audio signal and the reference signal come calculate short-term-the first score, by comparing described first
The total energy value of the total energy value of audio signal and the reference signal is come short-term-the second score that calculates and based on described the
The peak information of one audio signal come at least one of short-term-third score for calculating,
The state output unit is based on short-term-the first score, described short-term-the calculated by the judging unit
At least one of two scores and the short-term-third score can export the transition condition of the mechanical equipment.
The state output unit can to first score, second score calculated by the judging unit and
The third score is applicable in weighted value respectively, to calculate the state score of the mechanical equipment.
The state output unit may determine that the state of the mechanical equipment based on the state score to alert, infusing
Any one of meaning, normal.
The method of the diagnostic device diagnostic machinery equipment state of another exemplary embodiment includes following according to the present invention
Step: the first unit detection time is inscribed to be received in the first audio signal that the mechanical equipment detects;By first audio
Signal and reference signal are compared, to judge whether the mechanical equipment is driving;If the mechanical equipment is driving,
Just judge that data are stored using first audio signal as first;The stored quantity for judging data of judgement whether with
Reference value is identical;And when the stored quantity for judging data is reference value, held based on stored multiple judgement data
The long-term diagnostic process of row.
Judge in the step of whether mechanical equipment is driving described, calculates first audio signal and the reference
The difference of each frequency energy of signal, if the few frequency of the energy value compared with the reference signal be reference value hereinafter, if can sentence
Break in the first unit detection time the mechanical equipment driving.
The step of execution long-term diagnostic process can include the following steps: to the multiple judgement data integrated with
Extract accumulation diagnosis signal;It finds out by comparing each frequency energy value of the accumulation diagnosis signal and reference signal and counts
First score of calculation is counted by comparing the total energy value of the total energy value for accumulating diagnosis signal and the reference signal
The second score calculated and at least one of the third score calculated based on the peak information of the accumulation diagnosis signal;
And it is set based at least one of first score, second score and described third score to diagnose the machinery
Standby state.
Diagnosis signals step is accumulated in the extraction can include the following steps: is extracted based on the multiple judgement data each
The time-based energy value Trendline of frequency;The representative energy value of corresponding frequencies is extracted in each frequency trend line;And make
The representativeness energy value corresponds to each frequency, to extract the accumulation diagnosis signal of energy value of the display based on frequency.
The diagnostic method also includes execution the step of executing short-term diagnostic process based on first audio signal
The step of short-term diagnostic process can include the following steps: find out by comparing first audio signal and the reference signal
Each frequency energy value come calculate short-term-the first score, by comparing first audio signal total energy value and the ginseng
It short-term-the second score examining the total energy value of signal to calculate and is calculated based on the peak information of first audio signal
At least one of short-term-third score;And based on short-term-the first score, short-term-the second score and institute
State the transition condition of at least one of short-term-third score output mechanical equipment.
The method of the diagnostic device diagnostic machinery equipment state of the another exemplary implementation scheme of the present invention comprises the steps of:
The multiple audio signals detected in multiple unit detection times are stored;The multiple audio signal is integrated,
Diagnosis signal is accumulated to extract;Each frequency energy value of the accumulation diagnosis signal and reference signal is counted by comparing
Calculate the first score;It is calculated by comparing the total energy value of the total energy value for accumulating diagnosis signal and the reference signal
Second score;Third score is calculated based on the peak information of the accumulation diagnosis signal;And to first score, institute
It states the second score and the third score is applicable in weighted value, to diagnose the shape of the mechanical equipment based on score calculated
State.
In described the step of being stored multiple audio signals, the audio that the detects when mechanical equipment is driven
Signal is stored, and the audio signal detected when the mechanical equipment stops being abandoned.
The step of extraction accumulation diagnosis signal can include the following steps: be based on each frequency of the multiple audio extraction
Time-based energy value Trendline;The representative energy value of corresponding frequencies is extracted in each frequency trend line;And make the generation
Table performance magnitude corresponds to each frequency, to extract the accumulation diagnosis signal of energy value of the display based on frequency.
The step of the first score of the calculating, which comprises the steps of:, calculates the accumulation diagnosis signal and the reference letter
Number each frequency energy difference;Output shows the difference curve figure of the difference of each frequency energy;And it is bent by comparing the difference
Line chart and reference value calculate first score, the output difference curve graph the step of in, can be that face by the difference of energy
The difference of dividing value frequency below is corrected to 0 and generates the difference curve figure.
Beneficial effect
In accordance with an exemplary embodiment of the invention, without stopping mechanical equipment, based in the mechanical equipment driven
The audio signal of generation can accurately be diagnosed to be the state of mechanical equipment.
Detailed description of the invention
Fig. 1 is structural schematic diagram, shows the diagnostic device of an exemplary implementation scheme according to the present invention.
Fig. 2 a is example view, shows an exemplary implementation scheme according to the present invention and has filtered out from diagnostic signal
The signal of signal specific.
Fig. 2 b is example view, in the time-frequency domain for showing an exemplary implementation scheme according to the present invention
Signal is used in diagnosis.
Fig. 3 is example view, shows the frequency-in the detection time of an exemplary implementation scheme according to the present invention
Diagnosis signal in energy domain.
Fig. 4 is flow chart, and show an exemplary implementation scheme according to the present invention judges follow-up based on audio signal
The method of disconnected equipment state.
Fig. 5 is method schematic diagram, for illustrate an exemplary implementation scheme according to the present invention based on audio signal
The method for judging whether to drive to diagnostic device.
Fig. 6 shows the respective judgement number of multiple unit detection times of an exemplary implementation scheme according to the present invention
According to.
Fig. 7 is flow chart, shows the short-term diagnostic process of an exemplary implementation scheme according to the present invention.
Fig. 8 is flow chart, shows the long-term diagnostic process of an exemplary implementation scheme according to the present invention.
Fig. 9 is used to illustrate the time-based energy value of a certain frequency of an exemplary implementation scheme according to the present invention
Trendline.
Figure 10 is flow chart, and the calculating long-term-the first for showing an exemplary implementation scheme according to the present invention judges item
The method of mesh score.
The output of the state output unit of an exemplary implementation scheme according to the present invention is shown respectively in Figure 11 and Figure 12
As a result.
Specific embodiment
In the following, the exemplary implementation scheme that present invention will be described in detail with reference to the accompanying, so that those skilled in the art can
To be easy to implement the present invention.The present invention being capable of deformation implementation in a variety of different ways, it is not limited to as described herein exemplary
Embodiment.The part unrelated with explanation is omitted in attached drawing, to clearly demonstrate the present invention, uses phase in specification in the whole text
With appended drawing reference indicate the same or similar part.
In specification in the whole text, when a certain component of certain a part of "comprising", unless there are especially opposite record, otherwise indicate
Can also comprising other components other non-excluded components.
Fig. 1 is structural schematic diagram, shows the diagnostic device of an exemplary implementation scheme according to the present invention;Fig. 2 a is to show
Example property view, shows the signal that an exemplary implementation scheme according to the present invention has filtered out signal specific from diagnostic signal;
Fig. 2 b is example view, shows the diagnosis letter in the time-frequency domain of an exemplary implementation scheme according to the present invention
Number;Fig. 3 is example view, shows frequency-energy in the detection time of an exemplary implementation scheme according to the present invention
Diagnosis signal in domain.
Referring to Fig.1, diagnostic device 10 in the various mechanical equipments such as industrial automation machinery, automobile, aircraft, ship by producing
Raw sound judges the state of mechanical equipment (to diagnostic device).It is set for this purpose, diagnostic device 10 collects machinery in certain time
The audio signal of standby middle generation, and according to the feature of established standards analysis audio signal, the state of mechanical equipment is judged with this.Example
Such as, diagnostic device 10 may determine that the state of joint, engine, motor, engine, bearing etc. based on audio signal.
Diagnostic device 10 includes audio signal extraction unit 100, judging unit 200 and state output unit 300.
Audio signal extraction unit 100 is from input from the audio letter mounted in diagnostic points or the microphone around diagnostic points
Diagnosis signal is extracted in number.At this point, since audio signal has noise signal (noi se), audio signal extraction unit 100
It can filter out or decay from audio input signal by noise signal.
Audio signal extraction unit 100 can carry out initial noise according to the characteristic or environment of audio input signal and filter out.
At this point, audio signal extraction unit 100 can using the physics phase difference for the audio signal for being input at least two microphones come into
Row initial noise filters out.
If thoroughly filtering out noise signal, signal meeting serious distortion, it is possible to extract with can not be successfully from diagnostic signal section
The signal for the sign that shows abnormality.Therefore, the setting of audio signal extraction unit 100 is for determining the noise signal which kind of being decayed to
The parameter of degree.Noise-filtering parameter may include environmental correclation parameter and relevant parameter.Environmental correclation parameter is according to detection ring
Border adjusts the parameter of noise signal attenuation degree, can be set based on signal-to-noise ratio (Sound to Noise Ratio, SNR).
Relevant parameter is comprising signal segment (diagnostic signal section) adjustment with the related audio signal of the condition diagnosing of mechanical equipment
The parameter of noise signal attenuation rate can be set based on the incidence coefficient for being overlapped how many noise signal in display diagnostic signal section
It is fixed.
It energy variation mode of the audio signal extraction unit 100 based on diagnostic signal section and adjacent with diagnostic signal section makes an uproar
The energy variation mode computation incidence relation of message number section.It include noise signal in diagnostic signal section, if thorough from diagnostic signal section
Bottom filters out noise signal, then the signal extracted can serious distortion.Therefore, it is true to be based on relevant parameter for audio signal extraction unit 100
It is scheduled on the attenuation degree of diagnostic signal section noise signal.
Audio signal extraction unit 100 is based on noise-filtering parameter and noise signal is decayed from diagnostic signal section, to extract
Signal is used in diagnosis.Audio signal extraction unit 100 can be decayed noise signal from noise signal section based on environmental correclation parameter, and
And noise signal can be decayed from diagnostic signal section based on environmental correclation parameter and relevant parameter.That is, audio signal extraction unit
100 extract the less diagnosis signal of distortion by suitably filtering out noise signal from diagnostic signal section.
Audio signal extraction unit 100 only can extract or filter out particular diagnosis from the diagnostic signal for filter out noise signal
Signal.
Audio signal extraction unit 100 is including the specific of parameter using the frequency content of particular diagnosis signal as parameter
In the case where the amplitude preservation certain time of diagnostic signal, this is considered as special circumstances, it can be only by the target signal filter or extraction.
Audio signal extraction unit 100, can be from filtering out using the feature for the automated machine for continuing to repeat required movement
Signal specific is only extracted or filtered out in the diagnostic signal extracted after noise.This is to be considered using the user of product of the present invention
, broadly the diagnosis of the associated components of diagnostics division position can be made more accurate in the diagnosis for the region of interest to be diagnosed, and
And energy value is far longer than under the noise (that is, the inverted situation of SNR) of diagnostic signal can also accurately extract diagnostic signal.
Referring to Fig. 2 a, audio signal extraction unit 100 can from filter out filtered out in the diagnostic signal extracted after noise it is specific
Signal.
Referring to Fig. 2 b, audio signal extraction unit 100 is from the audio signal generated in diagnostic device in detection time
Middle extraction diagnosis signal.The diagnosis signal being attenuated by various parameters noise is exported in time-frequency domain, so that it may
It is shown as shown in Fig. 2 b.
Judging unit 200 judges the state to diagnostic device based on the audio signal of input.It is described herein, in order to accurate
Diagnosis, judging unit 200 judged based on the audio signal that the noise of audio signal extraction unit 100 is attenuated to diagnostic device
State, but judging unit 200 can use and come input from the result that the audio signal of microphone is handled in various ways
Judge the state to diagnostic device.
Judging unit 200 will be given in certain time (detection time) with the frequency transformed data of about 0.03 second interval input
It is tired to deposit, and the frequency response data of the diagnosis signal stored is retrieved, and then recalculate the energy information of each frequency, so that
The characteristic of each frequency shows best.The even diagnostic data that is filtered out of noise, since the characteristic of audio may also generate respectively
Kind phenomenon, such as each data will not consistently show the energy value of special frequency band, and tire out and deposit in the time and will not always exist, and
It is that can disappear after being temporarily present.Therefore, judging unit 200 can will be directed to the dispersion of the accumulated value of each frequency energy value, mark
Quasi- deviation, change rate, distributive law etc. statistics calculation integrated, thus will test time memory all frequency contents
It restores (calculating again), and arranges as the final result value for the acoustic characteristic inputted in detection time.
For arranging each frequency energy information for final result value, frequency-energy domain can be displayed in as shown in Figure 3.This
When, judging unit 200 scans the audio signal stored in certain time when necessary may be used with fully reflecting energy value by frequency
It is corrected.
Judging unit 200 calculates the score of judgement project based on each frequency energy information of audio signal.Judgement project
Reach the frequency information of peak value, the energy value of each frequency, total energy value etc. comprising energy.
At this point, judging unit 200 can be calculated based on each frequency energy information in unit detection time (one-time detection)
The score of the judgement project of the unit detection time.Mechanical equipment state is judged based on this one-time detection result, can be described as
Short-term diagnosis.When diagnostic result is without diagnosing also obvious for a long time, short-term diagnosis can be executed.In addition, judging unit 200 can
Judgement project is calculated based on each frequency energy information obtained from the testing result accumulation more than certain number (such as 30 times)
Score, this can be described as diagnosing for a long time.It is unobvious or to need that long-term diagnosis can be described as a kind of diagnostic result diagnosed in short term
The diagnosis wanting the equipment of continuous observation and carrying out.
State output unit 300 weighted value is applicable in the score of the judgement project calculated of judging unit 200 with calculate to
The state score of diagnostic device, and the state to diagnostic device is exported according to established standards.State to diagnostic device be according to
What established standards exported in various ways, such as exported in a manner of any one of warning, attention, normal, and can with examine
Disconnected score exports together.
In the following, judging that the method for equipment state to be diagnosed is described in detail based on audio signal to judging unit.
Fig. 4 is flow chart, and show an exemplary implementation scheme according to the present invention judges follow-up based on audio signal
The method of disconnected equipment state;Fig. 5 is method schematic diagram, for illustrate an exemplary implementation scheme according to the present invention based on
Audio signal judges the method whether driven to diagnostic device;Fig. 6 shows an exemplary implementation scheme according to the present invention
Multiple respective judgement data of unit detection time.
Referring to Fig. 4, the audio signal (diagnosis signal) that is detected in 200 recruiting unit's detection time of judging unit
(S110).Unit detection time can be one detection period, and set the specified time.Audio signal can be follow-up
The signal generated in disconnected equipment or the signal corrected to improve diagnosis accuracy.For example, judging unit 200 can be believed from audio
Number extraction unit 100 receives the signal that specific audio as shown in Figure 2 a is filtered out or receives what noise as shown in Figure 2 b was attenuated
Audio signal.
Judging unit 200 is based on diagnosis and executes short-term diagnostic process (S120) with signal.That is, it is judged that unit 200 is based on
The a certain received audio signal of unit detection time calculates the score of judgement project, to judge in the temporary of the unit detection time
When state.
Judging unit 200 is based on diagnosis and judges whether driving (S130) to diagnostic device with signal.Referring to Fig. 5, sentence
Disconnected unit 200 is compared the energy of diagnosis signal and the energy of reference signal by frequency.Moreover, the energy value compared with reference signal
When few frequency is certain amount/certain proportion (such as 70%) or more, judging unit 200 may determine that not having to diagnostic device
There is starting, i.e., stops to diagnostic device.
Judging unit 200 will abandon received diagnosis signal if it is determined that do not start to diagnostic device
(S140).That is, judging unit 200 would not be used as diagnostic data if not the audio signal detected at any driving condition
To use.The reasons why discarding is because can be comprising the data of non-driven state, as long-term diagnosis for long-term diagnosis
Data using when the dispersion degree of each frequency can be produced a very large impact.For example, in the long-term diagnosis that detection number is 30 times,
If comprising 20 non-driven state data, length that the most point of distributive law for extracting frequency energy value calculates
For phase diagnosis, the distributive law of non-driven state data is certain to the distributive law higher than driving condition.It is thus also possible to cause
To non-driven state rather than the result that is diagnosed of driving condition.To sum up, obtaining noise in long-term diagnosis and being filtered out
Driving condition corresponding site signal and this is carried out diagnosis be considerable.
Received diagnosis will be determined as if it is determined that driving to diagnostic device by judging unit 200 with signal
Judgement data corresponding to unit detection time are simultaneously stored into repository (S150).It is interior during the test to accumulate respectively referring to Fig. 6
Unit detection time judges data.
Judging unit 200 judges whether the quantity for judging data stored is reference value N (such as N=30) (S160).
If the quantity of the judgement data stored is also not up to reference value, judging unit 200, which returns to, receives audio signal
Step (S110) receives the audio signal generated in mechanical equipment in next unit detection time.
If the quantity of the judgement data stored reaches reference value, judging unit 200 will be based on the judgement number of storage
According to the long-term diagnostic process (S170) of execution.That is, it is judged that unit 200 can carry out comprehensive audio letter based on the result that will test n times
Number each frequency energy value calculate the score of judgement project, to judge the state of mechanical equipment.
Short-term diagnostic process is highly effective to the diagnosis of following scenario described: checking transition condition or can obviously examine in a short time
It is disconnected clear, for example, cannot absolutely generate the apparent state of abnormal sound in finished product by boundary after finished product assembling in production technology
It is fixed, or the critical value that is defined suddenly well beyond production equipment.However, other than these apparent situations,
In most cases, state deteriorating or the probability to improve are extremely low suddenly for production equipment.Therefore, this portion is improved in order to supplement
Point, it preferably carries out simultaneously in short term and long-term diagnostic process, this can greatly improve the reliability and diagnosis of diagnostic result
Accuracy.Therefore, while diagnostic device 10 is diagnosed in short term, based on appropriate detection number (such as minimum 20 times or more)
It accumulates data and carries out long-term diagnostic process, to judge the state of mechanical equipment.
In the following, short-term diagnostic process and long-term diagnostic process are described in detail.
Fig. 7 is flow chart, shows the short-term diagnostic process of an exemplary implementation scheme according to the present invention.
Referring to Fig. 7, judging unit 200 receives diagnosis with signal 20 (S210).Judging unit 200 is used as diagnosis signal 20
It can use following input data, the input data is that input signal was converted to frequency data by FFT every 0.03 second
And it is tired deposit after, ingredient by tiring out the frequency data deposited in unit detection time analyzes the feature can most reflect each frequency again
The corrected data of mode.
Judging unit 200 is from diagnosis with extracting each frequency energy value in signal 20 and compare diagnosis signal 20 by each frequency
With the energy (S220) of reference signal 30.Reference signal 30 can be audio signal when diagnostic device is normal condition, or
Person can be from the data accumulated for a long time under normal operating condition with the model of the most energy value of each frequency energy distributive law
Corrected audio signal centered on enclosing.
Judging unit 200 is short to calculate based on each frequency energy comparison result of diagnosis signal 20 and reference signal 30
Phase-the first judges the score (S222) of project.Each frequency energy of judging unit 200 relatively can be each frequency energy difference or
Person's similarity.Judging unit 200 compare each frequency energy of diagnosis signal 20 and reference signal 30 as a result, based on energy value
Difference and each frequency similarity (change rate) compared result by each stage be applicable in critical value section value.It is according to comparison result
Below " normal " critical value, the situation of " attention " critical value range, " warning " critical value range, respectively by frequency assign critical value
Each basic unit score, to calculate score.But for diagnosing in short term, feelings that critical value preferably greater than diagnoses for a long time
Shape, wherein critical value can be set on the basis of according to technological know-how (knowhow) of acoustic characteristic.For example, to ginseng
Examine the energy value of signal difference and its deviation range lower than 7dB (normal critical value) or 7dB or more and 13dB or less (attention is faced
Dividing value), the situation of 13dB or more (warning critical value) assign score appropriate.Critical value and score are adjustable.
The total energy value (S230) of the calculating diagnosis signal 20 of judging unit 200.
Judging unit 200 calculates short-term-the second with the total energy value of signal 20 and reference energy value by comparing diagnosis
The score (S232) of judgement project.Reference energy value can be the total energy value of reference signal 30.- the second judges project in short term
Score judges that the score calculation of project is identical with short-term-the first, to the difference of the total energy value of diagnostic signal and reference signal
Setting is normal, pays attention to, the critical value of warning, and calculates the score relative to field corresponding to the difference with reference signal.Equally,
Preferably there is the critical value for being higher than and diagnosing for a long time in short-term diagnosis, wherein critical value is according to the technological know-how of acoustic characteristic
On the basis of set, assign score appropriate according to each critical value, and can judge that item score is worthwhile with short-term-the first.
Judging unit 200 extracts peak value (peak) (S240) in signal 20 from diagnosis.Signified peak value in diagnostic signal
The not instead of value that there were significant differences compared with reference signal in frequency domain constitutes the bandwidth of the frequency of peak value and the height of peak value
Condition and parameter defined it is consistent when, be defined as peak value.The width and height of peak value are independently carried out with reference signal
It calculates, refers to that almost without peak value, the energy value of all frequency bands, which is uniformly distributed, not to be protruded without abnormal good sound
's.Actually refer to the ingredient not highlighted.
After judging unit 200 assigns score to the peak information of the quantity comprising peak value, the height of peak value etc., by comparing
Reference signal and peak information included in together calculate the score (S242) that short-term-third judges project.Reference information can
To be the peak information (quantity, height etc.) of normal condition.Peak value is unrelated with reference signal, only merely to analyze at present
The audio status of the signal of input has the advantages that analysis can be compared with reference signal.For example, judging unit 200
It is found that number of peaks assigns score (1 point × quantity) by the frequency that is found, by with the peak information of reference signal ratio
Score can be added and subtracted compared with analysis.Assuming that judging list when peak information is consistent with the peak information of reference signal under a frequency
Member 200 can subtract at 1 point, keep when inconsistent, and it is consistent when can keep 30% according to the height change of peak value or add
Score etc..
Judging unit 200 to three of the above situation assign the score based on condition, and by the summation to these scores come
Terminate short-term diagnosis.Score summation based on condition can also include institute to three kinds of projects other than the situation of illustration
There is each frequency calculation condition.But when only carrying out unconditional worthwhile to three kinds of projects, due to worthwhile score mistake beyond expectation
Greatly, the critical value relative to worthwhile entire result may will become discommodity.Therefore, judging unit 200 can be according to each
Stage worthwhile result is worthwhile to carry out by adjusting range, so that worthwhile result summation is no more than particular fraction (such as 20).Example
Such as, if exporting in the first stage at 10 points, second stage may obtain 10 points or more as a result, still in this case can surpass
Cross at 20 points, therefore after using 10 as 100% adjust automatically ratios, it can be worthwhile on 10 points of first stage.If the first,
The result of two-stage is 15 points, then to obtain the result within 5 points in the phase III.In fact, 15 points are that its state is at least
A possibility that " attention " is more than the stage big points, thus finally final stage can export proceeded to relative to " attention " it is assorted
The score of the specific progress of degree.
Only, for short-term diagnosis, the case where deteriorating suddenly due to equipment, is rare, is different from long-term diagnosis,
It is applicable in high critical value in short-term diagnosis, though in order to avoid the low equipment for being likely to cause major accident of the frequency of occurrences is disliked suddenly
The case where change, simultaneously diagnoses this, and the situation preferably with generation abnormal sound is (for example, the scale sound of musical instrument, piano changes
Situation) equally handle.
As described above, on the basis of the score of each judgement project calculated of judging unit 200, state output unit
300 calculate the state score to diagnostic device, and export the short term state to diagnostic device according to established standards.
Fig. 8 is flow chart, shows the long-term diagnostic process of an exemplary implementation scheme according to the present invention;Fig. 9 is used for
Illustrate the time-based energy value Trendline of a certain frequency of an exemplary implementation scheme according to the present invention.
Referring to Fig. 8, judging unit 200 from the multiple diagnosis accumulated in certain time with extracting each frequency in signal and being based on when
Between energy value Trendline (S310).Referring to Fig. 9, multiple diagnosis are successively detected in chronological order with signal.Therefore, judge
Unit 200 finds out the energy value of specific frequency (for example, n-th frequency) from multiple diagnosis respectively with signal, finds out as shown in Figure 9
The distribution of energy value (longitudinal axis) based on time (horizontal axis).That is, Fig. 9 shows the time-based energy value point of specific frequency
Cloth.Judging unit 200 is divided into certain section in time-based energy Distribution value, by time shaft, and calculates energy by section
It is worth Trendline.For example, can be every 10 days calculating energy value Trendline.At this point, if based on all of time section are included in
Data find out Trendline, then noise may have an impact Trendline.Therefore, in order to reduce the influence of noise, judge list
Member 200 can find out Trendline using the central value of density data (the similar data of energy value).Audio signal due to environment because
Element or acoustic characteristic may temporarily include noise.Therefore, if making Trendline by density data, so that it may ignore noise, it can
Relevant data are judged to tool equipment state to reflect from probability.
Judging unit 200 generates accumulation diagnosis signal from the representative energy value of each each frequency of frequency trend line drawing
40(S320).Accumulation diagnosis is indicated with signal 40 with each frequency energy value curve graph (graph).That is, long-term diagnostic process benefit
The state of mechanical equipment is judged with frequency energy value 40 each derived from the testing result of accumulation, and short-term diagnostic process utilizes
Each frequency energy value 20 judges the transition condition of mechanical equipment derived from one-time detection result.
Judging unit 200 with extracting each frequency energy value in signal 40, and compares accumulation diagnosis by frequency from accumulation diagnosis
With the energy (S330) of signal 40 and reference signal 30.Wherein, the reference that reference signal 30 and short-term diagnostic process are utilized is believed
Number may it is identical may also be different.Each frequency energy of the judging unit 200 based on accumulation diagnosis signal 40 and reference signal 30
Comparison result and reference value calculate the long-term-the first score (S332) for judging project.Each frequency energy relatively can be each frequency
The difference of rate energy.
Judging unit 200 calculates the total energy value (S340) of accumulation diagnosis signal 40.Judging unit 200 is by comparing tired
The product diagnosis total energy value of signal 40 calculates the long-term-the second score (S342) for judging project with reference energy value.With reference to
Energy value can be the total energy value of reference signal 30.In mechanical equipment, lubricates shortage of oil or nut or bolt is slightly loose
When opening, whole audio power be will increase.Therefore, judge that the score of project can predict mechanical equipment from second to a certain extent
It is middle to lead to the problem of and state.
Judging unit 200 extracts peak value (peak) (S350) in signal 40 from accumulation diagnosis.Judging unit 200 passes through ratio
Peak information and reference information compared with the quantity comprising peak value, the height of peak value etc. judge dividing for project to calculate long-term-third
Number (S352).Reference information can be the peak information of normal condition.When there is mostly dynamic abnormal position in mechanical equipment, peak value ratio
Normal condition occurs more.Therefore, judge that the score of project can be predicted to produce in mechanical equipment from third to a certain extent
The reason of raw problem and state.
As described above, on the basis of the score of each judgement project calculated of judging unit 200, state output unit
300 calculate the state score to diagnostic device, and export the long term state to diagnostic device according to established standards.
Figure 10 is flow chart, and the calculating long-term-the first for showing an exemplary implementation scheme according to the present invention judges item
The method of purpose score.
Referring to Fig.1 0, the difference of each frequency energy of the judging unit 200 based on accumulation diagnosis signal 40 and reference signal 30,
The long-term-the first score for judging project can be calculated.
Judging unit 200 is calculated the difference (difference) of accumulation diagnosis signal 40 and the energy of reference signal 30 by frequency
(S410)。
When the difference of the energy of a certain frequency is critical value or less, the difference of the frequency is corrected to 0 simultaneously by judging unit 200
It exports difference curve graph (S420).When the difference of the energy of a certain frequency is more than critical value, the difference of the frequency is directly exported
Onto difference curve figure.
Judging unit 200 calculates the long-term-the first score for judging project by comparing difference curve figure and reference value
(S430)。
As described above, judging unit 200 the energy of a certain frequency difference be critical value below when, by the difference of the frequency
It is corrected to 0, so as to improve the efficiency of calculation.
The output of the state output unit of an exemplary implementation scheme according to the present invention is shown respectively in Figure 11 and Figure 12
As a result.
Referring to Fig.1 1, state output unit 300 is applicable in weighted value to the score of the judgement project calculated of judging unit 200
To calculate the state score to diagnostic device, and the state to diagnostic device is exported according to established standards.Only pass through a judgement
Project is difficult to accurately judge the state of mechanical equipment.Therefore, state output unit 300 is applicable in weighted value to each judgement project,
Accurately to judge the state of mechanical equipment.When actually occurring abnormal sound in the characteristic of audio, to comprising each frequency energy value it
Two kinds of projects in difference, three kinds of judgement projects of integral energy size, number of peaks can have an impact.For example, integral energy is big
Although the energy value for being less than reference signal but crest frequency is greater than reference signal, indicate that the size of peak value even larger than refers to
Signal.Alternatively, the frequency factors for being also possible to rise as integral energy value that the difference of each frequency energy is big more.With these
Part is foundation, more than specific quantity (such as three) is met in many material elements calculated using three kinds of judgement projects
When condition, weighted value can be assigned to the end value of calculation according to the degree of material elements.State to diagnostic device is according to setting
Calibration standard exports in various ways.For example, state output unit 300 can on picture visual display warning, pay attention to, be normal
It is any.State output unit 300 can show warning, attention, normal on picture in the form of key, press when user touches
When key, the particular state information of the mechanical equipment can be shown.
State output unit 300 not only can also show state, and can be by machine with visual display by alarming
The information such as the status information of tool equipment and position send wireless terminal or remote server to.User can save as a result,
The hard work that entire plant area is confirmed is obtained, only checks associated machines if necessary.Referring to Fig.1 2, state output unit
300 can set danger level and warning level based on energy value.Moreover, state output unit 300 from accumulation diagnosis signal
40 can be by danger level and warning level with visual display.
In Figure 12, each frequency energy value has exceeded danger level, therefore administrative staff can know easily wait diagnose
The state of equipment has reached danger level.
Diagnostic device 10 is connected to the multiple mechanical equipments driven, collects the audio signal of each mechanical equipment, is based on
Audio signal monitors the state of each mechanical equipment in real time, can identify the mechanical equipment to go wrong.
As described above, in accordance with an exemplary embodiment of the invention, based on the sound generated in the mechanical equipment driven
Frequency signal can accurately be diagnosed to be the state of mechanical equipment.
Exemplary implementation scheme of the invention is described in detail above, but scope of the presently claimed invention is simultaneously
It is not limited to this, what those skilled in the art was carried out using basic conception of the invention defined in claims
Various forms of changes and improvement, also belong to scope of the presently claimed invention.
Claims (13)
1. a kind of diagnostic device for diagnostic machinery equipment state, includes:
Judging unit, the judging unit accumulate audio signal received in multiple unit detection times using as judging number
According to, and find out by comparing each frequency energy value of the audio signal and reference signal accumulated the first score calculated, pass through
Compare the total energy value of the audio signal of the accumulation and the total energy value of the reference signal the second score and base that calculate
In the peak information of the audio signal of the accumulation come at least one of the third score that calculates, wherein the judging unit
The time-based energy value Trendline of each frequency is extracted based on the audio signal of the accumulation, and is mentioned in each frequency trend line
The representative energy value for taking corresponding frequencies extracts the audio signal for showing the accumulation using the representative energy value of each frequency
The energy value based on frequency the first curve graph, based on first curve graph and show the reference signal based on frequency
The second curve graph of energy value find out at least one in first score, second score and the third score
It is a;And
State output unit, the state output unit is based on first score calculated by the judging unit, described
At least one of two scores and the third score export the state of the mechanical equipment.
2. diagnostic device according to claim 1, also includes:
Audio signal extraction unit, the audio signal extraction unit per unit detection time are received mounted in the mechanical equipment
Audio signal detected by least one microphone, and the judgement will be sent after sound attenuation to from received audio signal
Unit.
3. diagnostic device according to claim 1, wherein
The judging unit passes through ratio in the case where being entered any audio signal detected in arbitrary unit detection time
Any audio signal and the reference signal judge whether the mechanical equipment is driving, if the mechanical equipment
It is driving, any audio signal is just determined as the judgement data.
4. diagnostic device according to claim 3, wherein
The judging unit calculates the difference of each frequency energy of any audio signal and the reference signal, if with described
The reference signal frequency few compared to energy value is reference value hereinafter, being then judged as the machinery described in the arbitrary unit detection time
Equipment is driving.
5. diagnostic device according to claim 3, wherein
The judging unit when mechanical equipment stopping, abandoning any audio letter in the arbitrary unit detection time
Number.
6. diagnostic device according to claim 1, wherein
The judging unit extracts the time-based energy Distribution value of each frequency based on the audio signal of the accumulation, and by when
Between axis be divided into certain section, by section calculate energy value Trendline.
7. diagnostic device according to claim 6, wherein
The judging unit calculates the energy value Trendline in a manner of by density data in the energy Distribution value.
8. diagnostic device according to claim 1, wherein
The judging unit receives the first audio signal in the first unit detection time, and finds out by comparing first sound
Each frequency energy value of frequency signal and the reference signal come calculate short-term-the first score, by comparing first audio
The total energy value of the total energy value of signal and the reference signal is come short-term-the second score that calculates and based on first sound
The peak information of frequency signal come at least one of short-term-third score for calculating,
The state output unit is based on short-term-the first score calculated by the judging unit, described-the second point short-term
The transition condition of at least one of several and described short-term-third score output mechanical equipment.
9. diagnostic device according to claim 1, wherein
The state output unit is to first score, second score calculated by the judging unit and described
Three scores are applicable in weighted value respectively, to calculate the state score of the mechanical equipment.
10. diagnostic device according to claim 9, wherein
The state of the mechanical equipment is judged as warning, attention, normal based on the state score by the state output unit
Any one of.
11. a kind of method of diagnostic device diagnostic machinery equipment state comprising the steps of:
The multiple audio signals detected in multiple unit detection times are stored;
The multiple audio signal is integrated, accumulates diagnosis signal to extract;
Each frequency energy value of the accumulation diagnosis signal and reference signal is calculated into the first score by comparing;
Second point is calculated by comparing the total energy value of the total energy value for accumulating diagnosis signal and the reference signal
Number;
Third score is calculated based on the peak information of the accumulation diagnosis signal;And
To first score, second score and the third score be applicable in weighted value, with based on score calculated come
The state of the mechanical equipment is diagnosed,
Wherein, the step of extraction accumulation diagnosis signal comprising the steps of:
Based on the time-based energy value Trendline of each frequency of the multiple audio extraction;
The representative energy value of corresponding frequencies is extracted in each frequency trend line;And
The representativeness energy value is set to correspond to each frequency, to extract the accumulation diagnosis letter of energy value of the display based on frequency
Number.
12. according to the method for claim 11, wherein
In described the step of being stored multiple audio signals,
The audio signal detected when by mechanical equipment driving is stored, and abandons detection when the mechanical equipment stops
The audio signal arrived.
13. according to the method for claim 11, wherein
The step of the first score of the calculating comprising the steps of:
Calculate the difference of each frequency energy of the accumulation diagnosis signal and the reference signal;
Output shows the difference curve figure of the difference of each frequency energy;And
First score is calculated by comparing the difference curve figure and reference value,
In the output difference curve graph the step of,
The difference that the difference of energy is critical value frequency below is corrected to 0 and generates the difference curve figure.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020140190823A KR101663820B1 (en) | 2014-12-26 | 2014-12-26 | Method and apparatus of machine diagnosis using sound signal |
KR10-2014-0190823 | 2014-12-26 |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105741850A CN105741850A (en) | 2016-07-06 |
CN105741850B true CN105741850B (en) | 2019-06-28 |
Family
ID=56296276
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510977714.2A Active CN105741850B (en) | 2014-12-26 | 2015-12-23 | Utilize the method and device of audio signal diagnostic machinery equipment |
Country Status (3)
Country | Link |
---|---|
JP (1) | JP6643079B2 (en) |
KR (1) | KR101663820B1 (en) |
CN (1) | CN105741850B (en) |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6447583B2 (en) | 2016-06-24 | 2019-01-09 | トヨタ自動車株式会社 | Recombinant yeast and method for producing ethanol using the same |
CN107346665A (en) * | 2017-06-29 | 2017-11-14 | 广州视源电子科技股份有限公司 | Method, apparatus, equipment and the storage medium of audio detection |
KR102039743B1 (en) * | 2017-08-11 | 2019-11-01 | (주)아이티공간 | Predictive maintenance method of driving part |
WO2019189950A1 (en) * | 2018-03-27 | 2019-10-03 | 테크와이드 주식회사 | Method, system, and program for remote diagnosis of facility |
CN110319918A (en) * | 2018-03-28 | 2019-10-11 | 深圳金智凌轩视讯技术有限公司 | Pass through the method and device of sound detection equipment state |
KR20210017577A (en) * | 2019-08-09 | 2021-02-17 | 주식회사 엘지화학 | Quantitatively diagnosis method for manufacturing facility |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1111778A (en) * | 1994-04-07 | 1995-11-15 | 西屋电气公司 | Diagnostic system responsive to learned audio signatures |
WO2000004359A1 (en) * | 1998-07-15 | 2000-01-27 | Horton, Inc. | Abnormalities detection with acoustic monitor |
JP2001087863A (en) * | 1999-09-20 | 2001-04-03 | Amada Co Ltd | Plate working method and device |
CN1926413A (en) * | 2004-10-18 | 2007-03-07 | 日本精工株式会社 | Abnormality diagnosis system for machinery |
CN101135614A (en) * | 2006-08-29 | 2008-03-05 | 松下电工株式会社 | Anomaly monitoring device and method |
CN202853912U (en) * | 2012-10-30 | 2013-04-03 | 上海电机学院 | Train fault detecting device |
CN103325387A (en) * | 2012-03-23 | 2013-09-25 | 三菱电机株式会社 | Abnormal sound diagnostic device |
CN103645060A (en) * | 2013-11-08 | 2014-03-19 | 北京能高自动化技术股份有限公司 | Method for recognizing and diagnosing sudden failures of wind generating set based on state monitoring system |
Family Cites Families (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US684312A (en) * | 1900-11-01 | 1901-10-08 | Jacob D Rush | Garment-holder. |
JPS5850435A (en) * | 1981-09-22 | 1983-03-24 | Kansai Electric Power Co Inc:The | Vibration monitor |
JPH0663916B2 (en) * | 1989-03-14 | 1994-08-22 | 富士電機株式会社 | Gear abnormality diagnosis device |
JP3251799B2 (en) * | 1995-02-13 | 2002-01-28 | 三菱電機株式会社 | Equipment diagnostic equipment |
JPH09229762A (en) * | 1996-02-23 | 1997-09-05 | Hitachi Ltd | Method and apparatus for monitoring of abnormality of instrument |
JP2001324381A (en) * | 2000-05-17 | 2001-11-22 | Nkk Corp | Abnormality diagnostic method of plunger pump |
JP4029405B2 (en) * | 2003-08-28 | 2008-01-09 | オムロン株式会社 | Abnormality determination method and apparatus |
JP2007248184A (en) * | 2006-03-15 | 2007-09-27 | Omron Corp | Impact sound detection method and detector |
KR20090010430A (en) * | 2007-07-23 | 2009-01-30 | 주식회사 싸이언 | Apparatus for detecting mechanical trouble |
JP4608564B2 (en) * | 2008-03-28 | 2011-01-12 | 株式会社高田工業所 | Abnormality diagnosis method for low-speed rotating machinery |
CN105549016B (en) * | 2011-09-26 | 2018-11-02 | 安大略发电有限公司 | Ultrasonic matrix check |
JP2013140135A (en) * | 2011-12-09 | 2013-07-18 | Tokyo Electron Ltd | Abnormality detection apparatus for periodic driving system, processing apparatus including periodic driving system, abnormality detection method for periodic driving system, and computer program |
-
2014
- 2014-12-26 KR KR1020140190823A patent/KR101663820B1/en active IP Right Grant
-
2015
- 2015-12-23 CN CN201510977714.2A patent/CN105741850B/en active Active
- 2015-12-25 JP JP2015254031A patent/JP6643079B2/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1111778A (en) * | 1994-04-07 | 1995-11-15 | 西屋电气公司 | Diagnostic system responsive to learned audio signatures |
WO2000004359A1 (en) * | 1998-07-15 | 2000-01-27 | Horton, Inc. | Abnormalities detection with acoustic monitor |
JP2001087863A (en) * | 1999-09-20 | 2001-04-03 | Amada Co Ltd | Plate working method and device |
CN1926413A (en) * | 2004-10-18 | 2007-03-07 | 日本精工株式会社 | Abnormality diagnosis system for machinery |
CN101135614A (en) * | 2006-08-29 | 2008-03-05 | 松下电工株式会社 | Anomaly monitoring device and method |
CN103325387A (en) * | 2012-03-23 | 2013-09-25 | 三菱电机株式会社 | Abnormal sound diagnostic device |
CN202853912U (en) * | 2012-10-30 | 2013-04-03 | 上海电机学院 | Train fault detecting device |
CN103645060A (en) * | 2013-11-08 | 2014-03-19 | 北京能高自动化技术股份有限公司 | Method for recognizing and diagnosing sudden failures of wind generating set based on state monitoring system |
Also Published As
Publication number | Publication date |
---|---|
KR101663820B1 (en) | 2016-10-10 |
JP2016126016A (en) | 2016-07-11 |
KR20160082716A (en) | 2016-07-11 |
CN105741850A (en) | 2016-07-06 |
JP6643079B2 (en) | 2020-02-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105741850B (en) | Utilize the method and device of audio signal diagnostic machinery equipment | |
DE69732569T2 (en) | SYSTEM FOR DETECTING FAULTS OF ELECTRIC MOTORS | |
US11467562B2 (en) | Online monitoring device and system for a 3D printing device | |
Wang et al. | Theoretical and experimental investigations on spectral Lp/Lq norm ratio and spectral Gini index for rotating machine health monitoring | |
CN102870057B (en) | Plant diagnosis device, diagnosis method, and diagnosis program | |
CN105041631B (en) | The detection method and system of a kind of drive shaft vibration signal of gas compressor | |
US20060210141A1 (en) | Inspection method and inspection apparatus | |
US10539602B2 (en) | Active shaft grounding system with diagnostic waveform analysis | |
US10311703B1 (en) | Detection of spikes and faults in vibration trend data | |
CN109416299A (en) | The method for identifying the failure in the gear train in wind turbine | |
CN104949840A (en) | Diesel engine fault diagnosis method based on vibration analysis | |
Romero et al. | Vestas V90-3MW wind turbine gearbox health assessment using a vibration-based condition monitoring system | |
CN114813124B (en) | Bearing fault monitoring method and device | |
CN103821749A (en) | On-line diagnosis method of stall and surge of axial fan | |
Cao et al. | Remaining useful life prediction of wind turbine generator bearing based on EMD with an indicator | |
JP6685124B2 (en) | Diagnostic device and diagnostic method | |
WO2017176401A1 (en) | Shaft ground monitoring with diagnostic waveform analysis | |
DE102017206040A1 (en) | System and method for condition monitoring and / or fault diagnosis | |
CN115270896A (en) | Intelligent diagnosis method for identifying loosening fault of main bearing of aircraft engine | |
Fumagalli et al. | Agile diagnostic tool based on electrical signature analysis | |
Gelman et al. | Diagnostics of local tooth damage in gears by the wavelet technology | |
CN205449521U (en) | Range hood failure diagnosis's system based on EMD falls and makes an uproar | |
CN113168739A (en) | Method for checking at least one vehicle and electronic computing device | |
DE60217419T2 (en) | MONITORING AND DIAGNOSIS OF A TECHNICAL INSTALLATION USING A MECHANICALLY ACTIVATED DRAWING AGENT | |
JP4209793B2 (en) | Abnormality diagnosis method based on acoustic signal and program used for executing the method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |